From 84c9abc3e9762c4486ddc5ca0352a5d697a51987 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Wed, 25 Feb 2015 23:09:26 +0100 Subject: start of branch, commit log will be rewritten --- test/core/parser/stack/StackTest.cpp | 15 +- test/core/parser/utils/TokenizedDataTest.cpp | 602 +++++++++++---------------- test/core/parser/utils/TokenizerTest.cpp | 248 +++++------ test/formats/osml/OsmlStreamParserTest.cpp | 79 ++-- test/formats/osxml/OsxmlEventParserTest.cpp | 47 +-- 5 files changed, 413 insertions(+), 578 deletions(-) (limited to 'test') diff --git a/test/core/parser/stack/StackTest.cpp b/test/core/parser/stack/StackTest.cpp index a93f14a..83966d5 100644 --- a/test/core/parser/stack/StackTest.cpp +++ b/test/core/parser/stack/StackTest.cpp @@ -24,6 +24,7 @@ #include #include #include +#include #include @@ -53,7 +54,7 @@ struct Tracker { Variant::mapType annotationStartArgs; Variant annotationEndClassName; Variant annotationEndElementName; - Variant dataData; + TokenizedData dataData; bool startResult; bool fieldStartSetIsDefault; @@ -81,7 +82,7 @@ struct Tracker { annotationStartArgs = Variant::mapType{}; annotationEndClassName = Variant::fromString(std::string{}); annotationEndElementName = Variant::fromString(std::string{}); - dataData = Variant::fromString(std::string{}); + dataData = TokenizedData(); startResult = true; fieldStartSetIsDefault = false; @@ -157,7 +158,7 @@ public: return tracker.annotationEndResult; } - bool data(Variant &data) override + bool data(TokenizedData &data) override { tracker.dataCount++; tracker.dataData = data; @@ -363,7 +364,7 @@ TEST(Stack, multipleFields) s.data("test"); tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test", tracker.dataData); + EXPECT_EQ("test", tracker.dataData.text().asString()); s.fieldEnd(); tracker.expect(1, 0, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc @@ -375,7 +376,7 @@ TEST(Stack, multipleFields) s.data("test2"); tracker.expect(1, 0, 2, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test2", tracker.dataData); + EXPECT_EQ("test2", tracker.dataData.text().asString()); s.fieldEnd(); tracker.expect(1, 0, 2, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc @@ -387,7 +388,7 @@ TEST(Stack, multipleFields) s.data("test3"); tracker.expect(1, 0, 3, 2, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test3", tracker.dataData); + EXPECT_EQ("test3", tracker.dataData.text().asString()); s.fieldEnd(); tracker.expect(1, 0, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc @@ -744,4 +745,4 @@ TEST(Stack, fieldAfterDefaultField) ASSERT_FALSE(logger.hasError()); } } -} \ No newline at end of file +} diff --git a/test/core/parser/utils/TokenizedDataTest.cpp b/test/core/parser/utils/TokenizedDataTest.cpp index 231bad9..dfe2526 100644 --- a/test/core/parser/utils/TokenizedDataTest.cpp +++ b/test/core/parser/utils/TokenizedDataTest.cpp @@ -22,6 +22,43 @@ namespace ousia { +void assertToken(TokenizedDataReader &reader, TokenId id, + const std::string &text, const TokenSet &tokens = TokenSet{}, + WhitespaceMode mode = WhitespaceMode::TRIM, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset, + SourceId sourceId = InvalidSourceId) +{ + Token token; + ASSERT_TRUE(reader.read(token, tokens, mode)); + EXPECT_EQ(id, token.id); + EXPECT_EQ(text, token.content); + if (start != InvalidSourceOffset) { + EXPECT_EQ(start, token.getLocation().getStart()); + } + if (end != InvalidSourceOffset) { + EXPECT_EQ(end, token.getLocation().getEnd()); + } + EXPECT_EQ(sourceId, token.getLocation().getSourceId()); +} + +void assertText(TokenizedDataReader &reader, const std::string &text, + const TokenSet &tokens = TokenSet{}, + WhitespaceMode mode = WhitespaceMode::TRIM, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset, + SourceId id = InvalidSourceId) +{ + assertToken(reader, Tokens::Data, text, tokens, mode, start, end, id); +} + +void assertEnd(TokenizedDataReader &reader) +{ + Token token; + ASSERT_TRUE(reader.atEnd()); + ASSERT_FALSE(reader.read(token)); +} + TEST(TokenizedData, dataWhitespacePreserve) { TokenizedData data; @@ -29,15 +66,10 @@ TEST(TokenizedData, dataWhitespacePreserve) // 0123456789012345 // 0 1 - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" test1 test2 ", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(16U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, " test1 test2 ", TokenSet{}, WhitespaceMode::PRESERVE, + 0, 16); + assertEnd(reader); } TEST(TokenizedData, dataWhitespaceTrim) @@ -47,15 +79,10 @@ TEST(TokenizedData, dataWhitespaceTrim) // 0123456789012345 // 0 1 - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("test1 test2", token.content); - EXPECT_EQ(1U, token.getLocation().getStart()); - EXPECT_EQ(14U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::TRIM)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "test1 test2", TokenSet{}, WhitespaceMode::TRIM, 1, + 14); + assertEnd(reader); } TEST(TokenizedData, dataWhitespaceCollapse) @@ -65,15 +92,10 @@ TEST(TokenizedData, dataWhitespaceCollapse) // 0123456789012345 // 0 1 - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("test1 test2", token.content); - EXPECT_EQ(1U, token.getLocation().getStart()); - EXPECT_EQ(14U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::COLLAPSE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "test1 test2", TokenSet{}, WhitespaceMode::COLLAPSE, 1, + 14); + assertEnd(reader); } TEST(TokenizedData, singleToken) @@ -82,17 +104,9 @@ TEST(TokenizedData, singleToken) ASSERT_EQ(2U, data.append("$$")); data.mark(5, 0, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 0, 2); + assertEnd(reader); } TEST(TokenizedData, singleDisabledToken) @@ -101,15 +115,9 @@ TEST(TokenizedData, singleDisabledToken) ASSERT_EQ(2U, data.append("$$")); data.mark(5, 0, 2); - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "$$", TokenSet{}, WhitespaceMode::COLLAPSE, 0, 2); + assertEnd(reader); } TEST(TokenizedData, dualToken) @@ -120,18 +128,10 @@ TEST(TokenizedData, dualToken) data.mark(5, 0, 2); data.mark(6, 1, 1); - data.enableToken(5); - data.enableToken(6); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5, 6}, WhitespaceMode::COLLAPSE, 0, + 2); + assertEnd(reader); } TEST(TokenizedData, dualTokenShorterEnabled) @@ -142,385 +142,281 @@ TEST(TokenizedData, dualTokenShorterEnabled) data.mark(5, 0, 2); data.mark(6, 1, 1); - data.enableToken(6); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(6U, token.id); - EXPECT_EQ("$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(1U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(6U, token.id); - EXPECT_EQ("$", token.content); - EXPECT_EQ(1U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 6, "$", TokenSet{6}, WhitespaceMode::COLLAPSE, 0, 1); + assertToken(reader, 6, "$", TokenSet{6}, WhitespaceMode::COLLAPSE, 1, 2); + assertEnd(reader); } TEST(TokenizedData, dualTokenLongerEnabled) { TokenizedData data; ASSERT_EQ(2U, data.append("$$")); + data.mark(6, 0, 1); data.mark(5, 0, 2); + data.mark(6, 1, 1); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 0, 2); + assertEnd(reader); } TEST(TokenizedData, tokensAndDataPreserveWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ test $$")); - // 0123456789 + ASSERT_EQ(18U, data.append("$$ test text $$")); + // 012345678901234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" test ", token.content); - EXPECT_EQ(2U, token.getLocation().getStart()); - EXPECT_EQ(8U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::PRESERVE, 0, 2); + assertText(reader, " test text ", TokenSet{5}, WhitespaceMode::PRESERVE, + 2, 16); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::PRESERVE, 16, 18); + assertEnd(reader); } TEST(TokenizedData, tokensAndDataTrimWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ test $$")); - // 0123456789 + ASSERT_EQ(18U, data.append("$$ test text $$")); + // 012345678901234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("test", token.content); - EXPECT_EQ(3U, token.getLocation().getStart()); - EXPECT_EQ(7U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::TRIM)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::TRIM, 0, 2); + assertText(reader, "test text", TokenSet{5}, WhitespaceMode::TRIM, 3, + 15); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::TRIM, 16, 18); + assertEnd(reader); } TEST(TokenizedData, tokensAndDataCollapseWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ test $$")); - // 0123456789 + ASSERT_EQ(18U, data.append("$$ test text $$")); + // 012345678901234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("test", token.content); - EXPECT_EQ(3U, token.getLocation().getStart()); - EXPECT_EQ(7U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::COLLAPSE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 0, 2); + assertText(reader, "test text", TokenSet{5}, WhitespaceMode::COLLAPSE, 3, + 15); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 16, 18); + assertEnd(reader); } TEST(TokenizedData, tokensAndWhitespacePreserveWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ $$")); - // 0123456789 + ASSERT_EQ(8U, data.append("$$ $$")); + // 01234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" ", token.content); - EXPECT_EQ(2U, token.getLocation().getStart()); - EXPECT_EQ(8U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::PRESERVE, 0, 2); + assertText(reader, " ", TokenSet{5}, WhitespaceMode::PRESERVE, 2, 6); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::PRESERVE, 6, 8); + assertEnd(reader); } TEST(TokenizedData, tokensAndWhitespaceTrimWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ $$")); - // 0123456789 + ASSERT_EQ(8U, data.append("$$ $$")); + // 01234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::TRIM)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::TRIM, 0, 2); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::TRIM, 6, 8); + assertEnd(reader); } TEST(TokenizedData, tokensAndWhitespaceCollapseWhitespace) { TokenizedData data; - ASSERT_EQ(10U, data.append("$$ $$")); - // 0123456789 + ASSERT_EQ(8U, data.append("$$ $$")); + // 01234567 data.mark(5, 0, 2); data.mark(5, 2); - data.enableToken(5); - - Token token; - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(8U, token.getLocation().getStart()); - EXPECT_EQ(10U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.next(token, WhitespaceMode::COLLAPSE)); + TokenizedDataReader reader = data.reader(); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 0, 2); + assertToken(reader, 5, "$$", TokenSet{5}, WhitespaceMode::COLLAPSE, 6, 8); + assertEnd(reader); } -TEST(TokenizedData, textPreserveWhitespace) +TEST(TokenizedData, appendChars) { TokenizedData data; - ASSERT_EQ(6U, data.append(" $$ ")); - // 012345 - data.mark(5, 2, 2); - - data.enableToken(5); + ASSERT_EQ(1U, data.append('t', 5, 7)); + ASSERT_EQ(2U, data.append('e', 7, 8)); + ASSERT_EQ(3U, data.append('s', 8, 10)); + ASSERT_EQ(4U, data.append('t', 10, 12)); - Token token; - ASSERT_TRUE(data.text(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" ", token.content); - EXPECT_EQ(0U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.next(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(2U, token.getLocation().getStart()); - EXPECT_EQ(4U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.text(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" ", token.content); - EXPECT_EQ(4U, token.getLocation().getStart()); - EXPECT_EQ(6U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.text(token, WhitespaceMode::PRESERVE)); - ASSERT_FALSE(data.next(token, WhitespaceMode::PRESERVE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "test", TokenSet{5}, WhitespaceMode::COLLAPSE, 5, 12); + assertEnd(reader); } -TEST(TokenizedData, textTrimWhitespace) +TEST(TokenizedData, protectedWhitespace) { TokenizedData data; - ASSERT_EQ(6U, data.append(" $$ ")); - // 012345 - data.mark(5, 2, 2); + ASSERT_EQ(4U, data.append("test", 10)); + ASSERT_EQ(11U, data.append(" test", 14, true)); - data.enableToken(5); - - Token token; - ASSERT_FALSE(data.text(token, WhitespaceMode::TRIM)); - - ASSERT_TRUE(data.next(token, WhitespaceMode::TRIM)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(2U, token.getLocation().getStart()); - EXPECT_EQ(4U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); + TokenizedDataReader reader = data.reader(); + assertText(reader, "test test", TokenSet{5}, WhitespaceMode::COLLAPSE, 10, + 21); + assertEnd(reader); +} - ASSERT_FALSE(data.text(token, WhitespaceMode::TRIM)); - ASSERT_FALSE(data.next(token, WhitespaceMode::TRIM)); +TEST(TokenizedData, specialNewlineToken) +{ + TokenizedData data; + data.append("a\nb\n \nc\n"); + // 0 12 3456 78 9 + + const TokenSet tokens{Tokens::Newline}; + + TokenizedDataReader reader = data.reader(); + assertText(reader, "a", tokens, WhitespaceMode::COLLAPSE, 0, 1); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 1, 2); + assertText(reader, "b", tokens, WhitespaceMode::COLLAPSE, 2, 3); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 3, 4); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 7, 8); + assertText(reader, "c", tokens, WhitespaceMode::COLLAPSE, 8, 9); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 9, 10); + assertEnd(reader); } -TEST(TokenizedData, textCollapseWhitespace) +TEST(TokenizedData, specialParagraphToken) { TokenizedData data; - ASSERT_EQ(6U, data.append(" $$ ")); - // 012345 - data.mark(5, 2, 2); + data.append("a\nb\n \nc\n"); + // 0 12 3456 78 9 - data.enableToken(5); + const TokenSet tokens{Tokens::Paragraph}; - Token token; - ASSERT_FALSE(data.text(token, WhitespaceMode::COLLAPSE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "a b", tokens, WhitespaceMode::COLLAPSE, 0, 3); + assertToken(reader, Tokens::Paragraph, "\n \n", tokens, + WhitespaceMode::COLLAPSE, 3, 8); + assertText(reader, "c", tokens, WhitespaceMode::COLLAPSE, 8, 9); + assertEnd(reader); +} - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(5U, token.id); - EXPECT_EQ("$$", token.content); - EXPECT_EQ(2U, token.getLocation().getStart()); - EXPECT_EQ(4U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); +TEST(TokenizedData, specialSectionToken) +{ + TokenizedData data; + data.append("a\nb\n \n \t \n"); + // 0 12 3456 789 01 2 + // 0 1 + + const TokenSet tokens{Tokens::Section}; - ASSERT_FALSE(data.text(token, WhitespaceMode::COLLAPSE)); - ASSERT_FALSE(data.next(token, WhitespaceMode::COLLAPSE)); + TokenizedDataReader reader = data.reader(); + assertText(reader, "a b", tokens, WhitespaceMode::COLLAPSE, 0, 3); + assertToken(reader, Tokens::Section, "\n \n \t \n", tokens, + WhitespaceMode::COLLAPSE, 3, 13); + assertEnd(reader); } -TEST(TokenizedData, appendChars) +TEST(TokenizedData, specialTokenPrecedence) { TokenizedData data; - ASSERT_EQ(1U, data.append('t', 5, 7)); - ASSERT_EQ(2U, data.append('e', 7, 8)); - ASSERT_EQ(3U, data.append('s', 8, 10)); - ASSERT_EQ(4U, data.append('t', 10, 12)); + data.append("a\nb\n\nc\n\n\nd"); + // 0 12 3 45 6 7 89 + + const TokenSet tokens{Tokens::Newline, Tokens::Paragraph, Tokens::Section}; + + TokenizedDataReader reader = data.reader(); + assertText(reader, "a", tokens, WhitespaceMode::COLLAPSE, 0, 1); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 1, 2); + assertText(reader, "b", tokens, WhitespaceMode::COLLAPSE, 2, 3); + assertToken(reader, Tokens::Paragraph, "\n\n", tokens, + WhitespaceMode::COLLAPSE, 3, 5); + assertText(reader, "c", tokens, WhitespaceMode::COLLAPSE, 5, 6); + assertToken(reader, Tokens::Section, "\n\n\n", tokens, + WhitespaceMode::COLLAPSE, 6, 9); + assertText(reader, "d", tokens, WhitespaceMode::COLLAPSE, 9, 10); + assertEnd(reader); +} - Token token; - ASSERT_TRUE(data.text(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("test", token.content); - EXPECT_EQ(5U, token.getLocation().getStart()); - EXPECT_EQ(12U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.text(token, WhitespaceMode::COLLAPSE)); - ASSERT_FALSE(data.next(token, WhitespaceMode::COLLAPSE)); +TEST(TokenizedData, specialTokenPrecedence2) +{ + TokenizedData data; + data.append("\nb\n\nc\n\n\n"); + // 0 12 3 45 6 7 + + const TokenSet tokens{Tokens::Newline, Tokens::Paragraph, Tokens::Section}; + + TokenizedDataReader reader = data.reader(); + assertToken(reader, Tokens::Newline, "\n", tokens, WhitespaceMode::COLLAPSE, + 0, 1); + assertText(reader, "b", tokens, WhitespaceMode::COLLAPSE, 1, 2); + assertToken(reader, Tokens::Paragraph, "\n\n", tokens, + WhitespaceMode::COLLAPSE, 2, 4); + assertText(reader, "c", tokens, WhitespaceMode::COLLAPSE, 4, 5); + assertToken(reader, Tokens::Section, "\n\n\n", tokens, + WhitespaceMode::COLLAPSE, 5, 8); + assertEnd(reader); } -TEST(TokenizedData, copy) +TEST(TokenizedData, specialTokenIndent) { TokenizedData data; - ASSERT_EQ(7U, data.append(" a $ b ")); - // 0123456 - data.mark(6, 3, 1); - data.enableToken(6); + data.append(" test\n\ttest2\n test3 \ttest4\ntest5"); + // 01234567 8 901234 5678901234567890 123456 789012 + // 0 1 2 3 4 + const TokenSet tokens{Tokens::Indent, Tokens::Dedent}; + + TokenizedDataReader reader = data.reader(); + assertToken(reader, Tokens::Indent, "", tokens, WhitespaceMode::COLLAPSE, + 4, 4); + assertText(reader, "test", tokens, WhitespaceMode::COLLAPSE, 4, 8); + assertToken(reader, Tokens::Indent, "", tokens, WhitespaceMode::COLLAPSE, + 10, 10); + assertText(reader, "test2 test3 test4", tokens, WhitespaceMode::COLLAPSE, 10, 37); + assertToken(reader, Tokens::Dedent, "", tokens, WhitespaceMode::COLLAPSE, + 38, 38); + assertText(reader, "test5", tokens, WhitespaceMode::COLLAPSE, 38, 43); + assertEnd(reader); +} - Token token; - ASSERT_TRUE(data.text(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("a", token.content); - EXPECT_EQ(1U, token.getLocation().getStart()); - EXPECT_EQ(2U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_FALSE(data.text(token, WhitespaceMode::COLLAPSE)); - - TokenizedData dataCopy = data; - - ASSERT_TRUE(data.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(6U, token.id); - EXPECT_EQ("$", token.content); - EXPECT_EQ(3U, token.getLocation().getStart()); - EXPECT_EQ(4U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(dataCopy.next(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(6U, token.id); - EXPECT_EQ("$", token.content); - EXPECT_EQ(3U, token.getLocation().getStart()); - EXPECT_EQ(4U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - - ASSERT_TRUE(data.text(token, WhitespaceMode::PRESERVE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ(" b ", token.content); - EXPECT_EQ(4U, token.getLocation().getStart()); - EXPECT_EQ(7U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - ASSERT_FALSE(data.next(token)); - - ASSERT_TRUE(dataCopy.text(token, WhitespaceMode::COLLAPSE)); - EXPECT_EQ(Tokens::Data, token.id); - EXPECT_EQ("b", token.content); - EXPECT_EQ(5U, token.getLocation().getStart()); - EXPECT_EQ(6U, token.getLocation().getEnd()); - EXPECT_EQ(InvalidSourceId, token.getLocation().getSourceId()); - ASSERT_FALSE(dataCopy.next(token)); +TEST(TokenizedData, specialTokenIndentOverlap) +{ + TokenizedData data; + data.append(" test\n\ttest2\n test3 \ttest4\ntest5"); + // 01234567 8 901234 5678901234567890 123456 789012 + // 0 1 2 3 4 + const TokenSet tokens{Tokens::Indent, Tokens::Dedent, 5}; + + data.mark(5, 4, 4); + + TokenizedDataReader reader = data.reader(); + assertToken(reader, Tokens::Indent, "", tokens, WhitespaceMode::COLLAPSE, + 4, 4); + assertToken(reader, 5, "test", tokens, WhitespaceMode::COLLAPSE, 4, 8); + assertToken(reader, Tokens::Indent, "", tokens, WhitespaceMode::COLLAPSE, + 10, 10); + assertText(reader, "test2 test3 test4", tokens, WhitespaceMode::COLLAPSE, 10, 37); + assertToken(reader, Tokens::Dedent, "", tokens, WhitespaceMode::COLLAPSE, + 38, 38); + assertText(reader, "test5", tokens, WhitespaceMode::COLLAPSE, 38, 43); + assertEnd(reader); } + } diff --git a/test/core/parser/utils/TokenizerTest.cpp b/test/core/parser/utils/TokenizerTest.cpp index 3809a12..0f2bfb7 100644 --- a/test/core/parser/utils/TokenizerTest.cpp +++ b/test/core/parser/utils/TokenizerTest.cpp @@ -20,6 +20,7 @@ #include #include +#include namespace ousia { @@ -31,23 +32,40 @@ TEST(Tokenizer, tokenRegistration) ASSERT_EQ(0U, tokenizer.registerToken("a")); ASSERT_EQ(Tokens::Empty, tokenizer.registerToken("a")); - ASSERT_EQ("a", tokenizer.getTokenString(0U)); + ASSERT_EQ("a", tokenizer.lookupToken(0U).string); ASSERT_EQ(1U, tokenizer.registerToken("b")); ASSERT_EQ(Tokens::Empty, tokenizer.registerToken("b")); - ASSERT_EQ("b", tokenizer.getTokenString(1U)); + ASSERT_EQ("b", tokenizer.lookupToken(1U).string); ASSERT_EQ(2U, tokenizer.registerToken("c")); ASSERT_EQ(Tokens::Empty, tokenizer.registerToken("c")); - ASSERT_EQ("c", tokenizer.getTokenString(2U)); + ASSERT_EQ("c", tokenizer.lookupToken(2U).string); ASSERT_TRUE(tokenizer.unregisterToken(1U)); ASSERT_FALSE(tokenizer.unregisterToken(1U)); - ASSERT_EQ("", tokenizer.getTokenString(1U)); + ASSERT_EQ("", tokenizer.lookupToken(1U).string); ASSERT_EQ(1U, tokenizer.registerToken("d")); ASSERT_EQ(Tokens::Empty, tokenizer.registerToken("d")); - ASSERT_EQ("d", tokenizer.getTokenString(1U)); + ASSERT_EQ("d", tokenizer.lookupToken(1U).string); +} + +void expectData(const std::string &expected, SourceOffset tokenStart, + SourceOffset tokenEnd, SourceOffset textStart, + SourceOffset textEnd, const Token &token, TokenizedData &data, + WhitespaceMode mode = WhitespaceMode::PRESERVE) +{ + ASSERT_EQ(Tokens::Data, token.id); + + Variant text = data.text(mode); + ASSERT_TRUE(text.isString()); + + EXPECT_EQ(expected, text.asString()); + EXPECT_EQ(tokenStart, token.location.getStart()); + EXPECT_EQ(tokenEnd, token.location.getEnd()); + EXPECT_EQ(textStart, text.getLocation().getStart()); + EXPECT_EQ(textEnd, text.getLocation().getEnd()); } TEST(Tokenizer, textTokenPreserveWhitespace) @@ -56,36 +74,34 @@ TEST(Tokenizer, textTokenPreserveWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::PRESERVE}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ(" this \t is only a \n\n test text ", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(36U, loc.getEnd()); + expectData(" this \t is only a \n\n test text ", 0, 36, 0, 36, + token, data, WhitespaceMode::PRESERVE); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::PRESERVE}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("this \t is only a \n\n test text", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(32U, loc.getEnd()); + expectData("this \t is only a \n\n test text", 0, 32, 0, 32, + token, data, WhitespaceMode::PRESERVE); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } } @@ -95,36 +111,34 @@ TEST(Tokenizer, textTokenTrimWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::TRIM}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("this \t is only a \n\n test text", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(1U, loc.getStart()); - ASSERT_EQ(33U, loc.getEnd()); + expectData("this \t is only a \n\n test text", 0, 36, 1, 33, token, + data, WhitespaceMode::TRIM); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::TRIM}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("this \t is only a \n\n test text", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(32U, loc.getEnd()); + expectData("this \t is only a \n\n test text", 0, 32, 0, 32, + token, data, WhitespaceMode::TRIM); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } } @@ -134,36 +148,34 @@ TEST(Tokenizer, textTokenCollapseWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::COLLAPSE}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("this is only a test text", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(1U, loc.getStart()); - ASSERT_EQ(33U, loc.getEnd()); + expectData("this is only a test text", 0, 36, 1, 33, token, data, + WhitespaceMode::COLLAPSE); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - Tokenizer tokenizer{WhitespaceMode::COLLAPSE}; + Tokenizer tokenizer; Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("this is only a test text", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(32U, loc.getEnd()); + expectData("this is only a test text", 0, 32, 0, 32, token, data, + WhitespaceMode::COLLAPSE); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } } @@ -177,14 +189,12 @@ TEST(Tokenizer, simpleReadToken) { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test1", token.content); - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); + expectData("test1", 0, 5, 0, 5, token, data); char c; ASSERT_TRUE(reader.peek(c)); @@ -193,7 +203,8 @@ TEST(Tokenizer, simpleReadToken) { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); ASSERT_EQ(tid, token.id); ASSERT_EQ(":", token.content); @@ -209,14 +220,10 @@ TEST(Tokenizer, simpleReadToken) { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test2", token.content); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); - SourceLocation loc = token.location; - ASSERT_EQ(6U, loc.getStart()); - ASSERT_EQ(11U, loc.getEnd()); + expectData("test2", 6, 11, 6, 11, token, data); char c; ASSERT_FALSE(reader.peek(c)); @@ -233,21 +240,17 @@ TEST(Tokenizer, simplePeekToken) { Token token; - ASSERT_TRUE(tokenizer.peek(reader, token)); - - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test1", token.content); - - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); + TokenizedData data; + ASSERT_TRUE(tokenizer.peek(reader, token, data)); + expectData("test1", 0, 5, 0, 5, token, data); ASSERT_EQ(0U, reader.getOffset()); ASSERT_EQ(5U, reader.getPeekOffset()); } { Token token; - ASSERT_TRUE(tokenizer.peek(reader, token)); + TokenizedData data; + ASSERT_TRUE(tokenizer.peek(reader, token, data)); ASSERT_EQ(tid, token.id); ASSERT_EQ(":", token.content); @@ -261,35 +264,26 @@ TEST(Tokenizer, simplePeekToken) { Token token; - ASSERT_TRUE(tokenizer.peek(reader, token)); - - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test2", token.content); - - SourceLocation loc = token.location; - ASSERT_EQ(6U, loc.getStart()); - ASSERT_EQ(11U, loc.getEnd()); + TokenizedData data; + ASSERT_TRUE(tokenizer.peek(reader, token, data)); + expectData("test2", 6, 11, 6, 11, token, data); ASSERT_EQ(0U, reader.getOffset()); ASSERT_EQ(11U, reader.getPeekOffset()); } { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test1", token.content); - - SourceLocation loc = token.location; - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + expectData("test1", 0, 5, 0, 5, token, data); ASSERT_EQ(5U, reader.getOffset()); ASSERT_EQ(5U, reader.getPeekOffset()); } { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); ASSERT_EQ(tid, token.id); ASSERT_EQ(":", token.content); @@ -303,14 +297,9 @@ TEST(Tokenizer, simplePeekToken) { Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); - - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("test2", token.content); - - SourceLocation loc = token.location; - ASSERT_EQ(6U, loc.getStart()); - ASSERT_EQ(11U, loc.getEnd()); + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + expectData("test2", 6, 11, 6, 11, token, data); ASSERT_EQ(11U, reader.getOffset()); ASSERT_EQ(11U, reader.getPeekOffset()); } @@ -320,6 +309,7 @@ TEST(Tokenizer, ambiguousTokens) { CharReader reader{"abc"}; Tokenizer tokenizer; + TokenizedData data; TokenId t1 = tokenizer.registerToken("abd"); TokenId t2 = tokenizer.registerToken("bc"); @@ -328,16 +318,17 @@ TEST(Tokenizer, ambiguousTokens) ASSERT_EQ(1U, t2); Token token; - ASSERT_TRUE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_TRUE(tokenizer.read(reader, token, data)); - ASSERT_EQ(Tokens::Data, token.id); - ASSERT_EQ("a", token.content); + expectData("a", 0, 1, 0, 1, token, data); SourceLocation loc = token.location; ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(1U, loc.getEnd()); - ASSERT_TRUE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_TRUE(tokenizer.read(reader, token, data)); ASSERT_EQ(t2, token.id); ASSERT_EQ("bc", token.content); @@ -346,7 +337,8 @@ TEST(Tokenizer, ambiguousTokens) ASSERT_EQ(1U, loc.getStart()); ASSERT_EQ(3U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(reader, token)); + data.clear(); + ASSERT_FALSE(tokenizer.read(reader, token, data)); } TEST(Tokenizer, commentTestWhitespacePreserve) @@ -354,7 +346,7 @@ TEST(Tokenizer, commentTestWhitespacePreserve) CharReader reader{"Test/Test /* Block Comment */", 0}; // 012345678901234567890123456789 // 0 1 2 - Tokenizer tokenizer(WhitespaceMode::PRESERVE); + Tokenizer tokenizer; const TokenId t1 = tokenizer.registerToken("/"); const TokenId t2 = tokenizer.registerToken("/*"); @@ -370,45 +362,23 @@ TEST(Tokenizer, commentTestWhitespacePreserve) Token t; for (auto &te : expected) { - EXPECT_TRUE(tokenizer.read(reader, t)); + TokenizedData data(0); + EXPECT_TRUE(tokenizer.read(reader, t, data)); EXPECT_EQ(te.id, t.id); - EXPECT_EQ(te.content, t.content); + if (te.id != Tokens::Data) { + EXPECT_EQ(te.content, t.content); + } else { + Variant text = data.text(WhitespaceMode::PRESERVE); + ASSERT_TRUE(text.isString()); + EXPECT_EQ(te.content, text.asString()); + } EXPECT_EQ(te.location.getSourceId(), t.location.getSourceId()); EXPECT_EQ(te.location.getStart(), t.location.getStart()); EXPECT_EQ(te.location.getEnd(), t.location.getEnd()); } - ASSERT_FALSE(tokenizer.read(reader, t)); -} - -TEST(Tokenizer, commentTestWhitespaceCollapse) -{ - CharReader reader{"Test/Test /* Block Comment */", 0}; - // 012345678901234567890123456789 - // 0 1 2 - Tokenizer tokenizer(WhitespaceMode::COLLAPSE); - const TokenId t1 = tokenizer.registerToken("/"); - const TokenId t2 = tokenizer.registerToken("/*"); - const TokenId t3 = tokenizer.registerToken("*/"); - - std::vector expected = { - {Tokens::Data, "Test", SourceLocation{0, 0, 4}}, - {t1, "/", SourceLocation{0, 4, 5}}, - {Tokens::Data, "Test", SourceLocation{0, 5, 9}}, - {t2, "/*", SourceLocation{0, 10, 12}}, - {Tokens::Data, "Block Comment", SourceLocation{0, 13, 26}}, - {t3, "*/", SourceLocation{0, 27, 29}}}; - - Token t; - for (auto &te : expected) { - EXPECT_TRUE(tokenizer.read(reader, t)); - EXPECT_EQ(te.id, t.id); - EXPECT_EQ(te.content, t.content); - EXPECT_EQ(te.location.getSourceId(), t.location.getSourceId()); - EXPECT_EQ(te.location.getStart(), t.location.getStart()); - EXPECT_EQ(te.location.getEnd(), t.location.getEnd()); - } - ASSERT_FALSE(tokenizer.read(reader, t)); + TokenizedData data; + ASSERT_FALSE(tokenizer.read(reader, t, data)); } } diff --git a/test/formats/osml/OsmlStreamParserTest.cpp b/test/formats/osml/OsmlStreamParserTest.cpp index d52fa5b..3d01007 100644 --- a/test/formats/osml/OsmlStreamParserTest.cpp +++ b/test/formats/osml/OsmlStreamParserTest.cpp @@ -30,11 +30,21 @@ namespace ousia { static TerminalLogger logger(std::cerr, true); // static ConcreteLogger logger; +static OsmlStreamParser::State skipEmptyData(OsmlStreamParser &reader) +{ + OsmlStreamParser::State res = reader.parse(); + if (res == OsmlStreamParser::State::DATA) { + EXPECT_FALSE(reader.getData().hasNonWhitespaceText()); + res = reader.parse(); + } + return res; +} + static void assertCommand(OsmlStreamParser &reader, const std::string &name, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::COMMAND, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::COMMAND, skipEmptyData(reader)); EXPECT_EQ(name, reader.getCommandName().asString()); if (start != InvalidSourceOffset) { EXPECT_EQ(start, reader.getCommandName().getLocation().getStart()); @@ -57,16 +67,19 @@ static void assertCommand(OsmlStreamParser &reader, const std::string &name, static void assertData(OsmlStreamParser &reader, const std::string &data, SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset) + SourceOffset end = InvalidSourceOffset, + WhitespaceMode mode = WhitespaceMode::COLLAPSE) { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - EXPECT_EQ(data, reader.getData().asString()); + Variant text = reader.getText(mode); + ASSERT_TRUE(text.isString()); + EXPECT_EQ(data, text.asString()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getData().getLocation().getStart()); + EXPECT_EQ(start, text.getLocation().getStart()); EXPECT_EQ(start, reader.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getData().getLocation().getEnd()); + EXPECT_EQ(end, text.getLocation().getEnd()); EXPECT_EQ(end, reader.getLocation().getEnd()); } } @@ -75,7 +88,7 @@ static void assertFieldStart(OsmlStreamParser &reader, bool defaultField, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::FIELD_START, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::FIELD_START, skipEmptyData(reader)); EXPECT_EQ(defaultField, reader.inDefaultField()); if (start != InvalidSourceOffset) { EXPECT_EQ(start, reader.getLocation().getStart()); @@ -89,7 +102,7 @@ static void assertFieldEnd(OsmlStreamParser &reader, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::FIELD_END, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::FIELD_END, skipEmptyData(reader)); if (start != InvalidSourceOffset) { EXPECT_EQ(start, reader.getLocation().getStart()); } @@ -103,7 +116,7 @@ static void assertAnnotationStart(OsmlStreamParser &reader, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_START, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_START, skipEmptyData(reader)); EXPECT_EQ(name, reader.getCommandName().asString()); if (start != InvalidSourceOffset) { EXPECT_EQ(start, reader.getCommandName().getLocation().getStart()); @@ -131,7 +144,7 @@ static void assertAnnotationEnd(OsmlStreamParser &reader, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_END, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_END, skipEmptyData(reader)); ASSERT_EQ(name, reader.getCommandName().asString()); if (!elementName.empty()) { ASSERT_EQ(1U, reader.getCommandArguments().asMap().size()); @@ -152,7 +165,7 @@ static void assertEnd(OsmlStreamParser &reader, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + ASSERT_EQ(OsmlStreamParser::State::END, skipEmptyData(reader)); if (start != InvalidSourceOffset) { EXPECT_EQ(start, reader.getLocation().getStart()); } @@ -205,26 +218,14 @@ TEST(OsmlStreamParser, whitespaceEliminationWithLinebreak) assertData(reader, "hello world", 1, 14); } -TEST(OsmlStreamParser, escapeWhitespace) -{ - const char *testString = " hello\\ \\ world "; - // 012345 67 89012345 - // 0 1 - CharReader charReader(testString); - - OsmlStreamParser reader(charReader, logger); - - assertData(reader, "hello world", 1, 15); -} - static void testEscapeSpecialCharacter(const std::string &c) { CharReader charReader(std::string("\\") + c); OsmlStreamParser reader(charReader, logger); EXPECT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - EXPECT_EQ(c, reader.getData().asString()); + EXPECT_EQ(c, reader.getText().asString()); - SourceLocation loc = reader.getData().getLocation(); + SourceLocation loc = reader.getText().getLocation(); EXPECT_EQ(0U, loc.getStart()); EXPECT_EQ(1U + c.size(), loc.getEnd()); } @@ -253,16 +254,16 @@ TEST(OsmlStreamParser, singleLineComment) OsmlStreamParser reader(charReader, logger); { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("a", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(1U, loc.getEnd()); } { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("b", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(33U, loc.getStart()); ASSERT_EQ(34U, loc.getEnd()); } @@ -279,16 +280,16 @@ TEST(OsmlStreamParser, multilineComment) OsmlStreamParser reader(charReader, logger); { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("a", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(1U, loc.getEnd()); } { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("b", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(40U, loc.getStart()); ASSERT_EQ(41U, loc.getEnd()); } @@ -305,16 +306,16 @@ TEST(OsmlStreamParser, nestedMultilineComment) OsmlStreamParser reader(charReader, logger); { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("a", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(1U, loc.getEnd()); } { ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getData().asString()); - SourceLocation loc = reader.getData().getLocation(); + ASSERT_EQ("b", reader.getText().asString()); + SourceLocation loc = reader.getText().getLocation(); ASSERT_EQ(40U, loc.getStart()); ASSERT_EQ(41U, loc.getEnd()); } @@ -569,8 +570,11 @@ TEST(OsmlStreamParser, multipleCommands) OsmlStreamParser reader(charReader, logger); assertCommand(reader, "a", 0, 2); + assertData(reader, " ", 2, 3, WhitespaceMode::PRESERVE); assertCommand(reader, "b", 3, 5); + assertData(reader, " ", 5, 6, WhitespaceMode::PRESERVE); assertCommand(reader, "c", 6, 8); + assertData(reader, " ", 8, 9, WhitespaceMode::PRESERVE); assertCommand(reader, "d", 9, 11); assertEnd(reader, 11, 11); } @@ -584,10 +588,13 @@ TEST(OsmlStreamParser, fieldsWithSpaces) OsmlStreamParser reader(charReader, logger); assertCommand(reader, "a", 0, 2); + assertData(reader, " ", 2, 3, WhitespaceMode::PRESERVE); assertFieldStart(reader, false, 3, 4); assertCommand(reader, "b", 4, 6); + assertData(reader, " ", 6, 7, WhitespaceMode::PRESERVE); assertCommand(reader, "c", 7, 9); assertFieldEnd(reader, 9, 10); + assertData(reader, " \n\n {", 10, 12, WhitespaceMode::PRESERVE); assertFieldStart(reader, false, 16, 17); assertCommand(reader, "d", 17, 19); assertFieldEnd(reader, 19, 20); diff --git a/test/formats/osxml/OsxmlEventParserTest.cpp b/test/formats/osxml/OsxmlEventParserTest.cpp index 3293370..6942166 100644 --- a/test/formats/osxml/OsxmlEventParserTest.cpp +++ b/test/formats/osxml/OsxmlEventParserTest.cpp @@ -21,6 +21,7 @@ #include #include #include +#include #include @@ -74,13 +75,11 @@ public: }; static std::vector> parseXml( - const char *testString, - WhitespaceMode whitespaceMode = WhitespaceMode::TRIM) + const char *testString) { TestOsxmlEventListener listener; CharReader reader(testString); OsxmlEventParser parser(reader, listener, logger); - parser.setWhitespaceMode(whitespaceMode); parser.parse(); return listener.events; } @@ -157,7 +156,7 @@ TEST(OsxmlEventParser, magicTopLevelTagInside) ASSERT_EQ(expectedEvents, events); } -TEST(OsxmlEventParser, commandWithDataPreserveWhitespace) +TEST(OsxmlEventParser, commandWithData) { const char *testString = " hello \n world "; // 012345678901 234567890123 @@ -168,50 +167,12 @@ TEST(OsxmlEventParser, commandWithDataPreserveWhitespace) {OsxmlEvent::DATA, Variant::arrayType{" hello \n world "}}, {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; - auto events = parseXml(testString, WhitespaceMode::PRESERVE); + auto events = parseXml(testString); ASSERT_EQ(expectedEvents, events); // Check the location of the text ASSERT_EQ(3U, events[1].second.asArray()[0].getLocation().getStart()); ASSERT_EQ(20U, events[1].second.asArray()[0].getLocation().getEnd()); } - -TEST(OsxmlEventParser, commandWithDataTrimWhitespace) -{ - const char *testString = " hello \n world "; - // 012345678901 234567890123 - // 0 1 2 - - std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, Variant::arrayType{"a", Variant::mapType{}}}, - {OsxmlEvent::DATA, Variant::arrayType{"hello \n world"}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; - - auto events = parseXml(testString, WhitespaceMode::TRIM); - ASSERT_EQ(expectedEvents, events); - - // Check the location of the text - ASSERT_EQ(5U, events[1].second.asArray()[0].getLocation().getStart()); - ASSERT_EQ(19U, events[1].second.asArray()[0].getLocation().getEnd()); -} - -TEST(OsxmlEventParser, commandWithDataCollapseWhitespace) -{ - const char *testString = " hello \n world "; - // 012345678901 234567890123 - // 0 1 2 - - std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, Variant::arrayType{"a", Variant::mapType{}}}, - {OsxmlEvent::DATA, Variant::arrayType{"hello world"}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; - - auto events = parseXml(testString, WhitespaceMode::COLLAPSE); - ASSERT_EQ(expectedEvents, events); - - // Check the location of the text - ASSERT_EQ(5U, events[1].second.asArray()[0].getLocation().getStart()); - ASSERT_EQ(19U, events[1].second.asArray()[0].getLocation().getEnd()); -} } -- cgit v1.2.3 From 041a2dd18050e9e26ca1ee00851461dff1e1f90c Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Thu, 26 Feb 2015 00:22:12 +0100 Subject: Moved "assert" functions to own header --- test/core/parser/utils/TokenizedDataTest.cpp | 39 +------------- test/core/parser/utils/TokenizedDataTestUtils.hpp | 64 +++++++++++++++++++++++ 2 files changed, 66 insertions(+), 37 deletions(-) create mode 100644 test/core/parser/utils/TokenizedDataTestUtils.hpp (limited to 'test') diff --git a/test/core/parser/utils/TokenizedDataTest.cpp b/test/core/parser/utils/TokenizedDataTest.cpp index dfe2526..8488459 100644 --- a/test/core/parser/utils/TokenizedDataTest.cpp +++ b/test/core/parser/utils/TokenizedDataTest.cpp @@ -20,44 +20,9 @@ #include -namespace ousia { - -void assertToken(TokenizedDataReader &reader, TokenId id, - const std::string &text, const TokenSet &tokens = TokenSet{}, - WhitespaceMode mode = WhitespaceMode::TRIM, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset, - SourceId sourceId = InvalidSourceId) -{ - Token token; - ASSERT_TRUE(reader.read(token, tokens, mode)); - EXPECT_EQ(id, token.id); - EXPECT_EQ(text, token.content); - if (start != InvalidSourceOffset) { - EXPECT_EQ(start, token.getLocation().getStart()); - } - if (end != InvalidSourceOffset) { - EXPECT_EQ(end, token.getLocation().getEnd()); - } - EXPECT_EQ(sourceId, token.getLocation().getSourceId()); -} - -void assertText(TokenizedDataReader &reader, const std::string &text, - const TokenSet &tokens = TokenSet{}, - WhitespaceMode mode = WhitespaceMode::TRIM, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset, - SourceId id = InvalidSourceId) -{ - assertToken(reader, Tokens::Data, text, tokens, mode, start, end, id); -} +#include "TokenizedDataTestUtils.hpp" -void assertEnd(TokenizedDataReader &reader) -{ - Token token; - ASSERT_TRUE(reader.atEnd()); - ASSERT_FALSE(reader.read(token)); -} +namespace ousia { TEST(TokenizedData, dataWhitespacePreserve) { diff --git a/test/core/parser/utils/TokenizedDataTestUtils.hpp b/test/core/parser/utils/TokenizedDataTestUtils.hpp new file mode 100644 index 0000000..c384f9d --- /dev/null +++ b/test/core/parser/utils/TokenizedDataTestUtils.hpp @@ -0,0 +1,64 @@ +/* + Ousía + Copyright (C) 2014, 2015 Benjamin Paaßen, Andreas Stöckel + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ + +#ifndef _OUSIA_TOKENIZED_DATA_TEST_UTILS_HPP_ +#define _OUSIA_TOKENIZED_DATA_TEST_UTILS_HPP_ + +namespace ousia { + +static void assertToken(TokenizedDataReader &reader, TokenId id, + const std::string &text, const TokenSet &tokens = TokenSet{}, + WhitespaceMode mode = WhitespaceMode::TRIM, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset, + SourceId sourceId = InvalidSourceId) +{ + Token token; + ASSERT_TRUE(reader.read(token, tokens, mode)); + EXPECT_EQ(id, token.id); + EXPECT_EQ(text, token.content); + if (start != InvalidSourceOffset) { + EXPECT_EQ(start, token.getLocation().getStart()); + } + if (end != InvalidSourceOffset) { + EXPECT_EQ(end, token.getLocation().getEnd()); + } + EXPECT_EQ(sourceId, token.getLocation().getSourceId()); +} + +static void assertText(TokenizedDataReader &reader, const std::string &text, + const TokenSet &tokens = TokenSet{}, + WhitespaceMode mode = WhitespaceMode::TRIM, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset, + SourceId id = InvalidSourceId) +{ + assertToken(reader, Tokens::Data, text, tokens, mode, start, end, id); +} + +static void assertEnd(TokenizedDataReader &reader) +{ + Token token; + ASSERT_TRUE(reader.atEnd()); + ASSERT_FALSE(reader.read(token)); +} + +} + +#endif /* _OUSIA_TOKENIZED_DATA_TEST_UTILS_HPP_ */ + -- cgit v1.2.3 From 19dd5946125e90dcbd61966896c9f6cfc4451d80 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Thu, 26 Feb 2015 00:22:23 +0100 Subject: Reactivated TokenizerTest --- CMakeLists.txt | 2 +- test/core/parser/utils/TokenizerTest.cpp | 94 ++++++++++++++++++++++++++++---- 2 files changed, 83 insertions(+), 13 deletions(-) (limited to 'test') diff --git a/CMakeLists.txt b/CMakeLists.txt index 225e63d..75909e9 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -327,7 +327,7 @@ IF(TEST) test/core/parser/stack/StateTest test/core/parser/utils/SourceOffsetVectorTest test/core/parser/utils/TokenizedDataTest -# test/core/parser/utils/TokenizerTest + test/core/parser/utils/TokenizerTest test/core/parser/utils/TokenTrieTest test/core/resource/ResourceLocatorTest test/core/resource/ResourceRequestTest diff --git a/test/core/parser/utils/TokenizerTest.cpp b/test/core/parser/utils/TokenizerTest.cpp index 0f2bfb7..785bd81 100644 --- a/test/core/parser/utils/TokenizerTest.cpp +++ b/test/core/parser/utils/TokenizerTest.cpp @@ -22,6 +22,8 @@ #include #include +#include "TokenizedDataTestUtils.hpp" + namespace ousia { TEST(Tokenizer, tokenRegistration) @@ -58,14 +60,16 @@ void expectData(const std::string &expected, SourceOffset tokenStart, { ASSERT_EQ(Tokens::Data, token.id); - Variant text = data.text(mode); - ASSERT_TRUE(text.isString()); + Token textToken; + TokenizedDataReader reader = data.reader(); + ASSERT_TRUE(reader.read(textToken, TokenSet{}, mode)); - EXPECT_EQ(expected, text.asString()); + EXPECT_EQ(expected, textToken.content); EXPECT_EQ(tokenStart, token.location.getStart()); EXPECT_EQ(tokenEnd, token.location.getEnd()); - EXPECT_EQ(textStart, text.getLocation().getStart()); - EXPECT_EQ(textEnd, text.getLocation().getEnd()); + EXPECT_EQ(textStart, textToken.getLocation().getStart()); + EXPECT_EQ(textEnd, textToken.getLocation().getEnd()); + EXPECT_TRUE(reader.atEnd()); } TEST(Tokenizer, textTokenPreserveWhitespace) @@ -97,8 +101,8 @@ TEST(Tokenizer, textTokenPreserveWhitespace) TokenizedData data; ASSERT_TRUE(tokenizer.read(reader, token, data)); - expectData("this \t is only a \n\n test text", 0, 32, 0, 32, - token, data, WhitespaceMode::PRESERVE); + expectData("this \t is only a \n\n test text", 0, 32, 0, 32, token, + data, WhitespaceMode::PRESERVE); data.clear(); ASSERT_FALSE(tokenizer.read(reader, token, data)); @@ -134,8 +138,8 @@ TEST(Tokenizer, textTokenTrimWhitespace) TokenizedData data; ASSERT_TRUE(tokenizer.read(reader, token, data)); - expectData("this \t is only a \n\n test text", 0, 32, 0, 32, - token, data, WhitespaceMode::TRIM); + expectData("this \t is only a \n\n test text", 0, 32, 0, 32, token, + data, WhitespaceMode::TRIM); data.clear(); ASSERT_FALSE(tokenizer.read(reader, token, data)); @@ -368,9 +372,12 @@ TEST(Tokenizer, commentTestWhitespacePreserve) if (te.id != Tokens::Data) { EXPECT_EQ(te.content, t.content); } else { - Variant text = data.text(WhitespaceMode::PRESERVE); - ASSERT_TRUE(text.isString()); - EXPECT_EQ(te.content, text.asString()); + TokenizedDataReader dataReader = data.reader(); + Token textToken; + ASSERT_TRUE(dataReader.read(textToken, TokenSet{}, + WhitespaceMode::PRESERVE)); + EXPECT_TRUE(dataReader.atEnd()); + EXPECT_EQ(te.content, textToken.content); } EXPECT_EQ(te.location.getSourceId(), t.location.getSourceId()); EXPECT_EQ(te.location.getStart(), t.location.getStart()); @@ -380,5 +387,68 @@ TEST(Tokenizer, commentTestWhitespacePreserve) TokenizedData data; ASSERT_FALSE(tokenizer.read(reader, t, data)); } + +TEST(Tokenizer, nonPrimaryTokens) +{ + CharReader reader{ + "<>"}; + // 012345678901234567890 12345678901234567890123456789012345678901234567 + // 0 1 2 3 4 5 6 + + Tokenizer tokenizer; + + TokenId tBackslash = tokenizer.registerToken("\\"); + TokenId tDollar = tokenizer.registerToken("$", false); + TokenId tSpeechStart = tokenizer.registerToken("<<", false); + TokenId tSpeechEnd = tokenizer.registerToken(">>", false); + + TokenSet tokens = TokenSet{tDollar, tSpeechStart, tSpeechEnd}; + + Token token, textToken; + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertToken(dataReader, tSpeechStart, "<<", tokens, + WhitespaceMode::TRIM, 0, 2); + assertText(dataReader, "switch to", tokens, WhitespaceMode::TRIM, 2, + 11); + assertToken(dataReader, tDollar, "$", tokens, WhitespaceMode::TRIM, 12, + 13); + assertText(dataReader, "inline", tokens, WhitespaceMode::TRIM, 13, 19); + assertEnd(dataReader); + } + + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(tBackslash, token.id); + ASSERT_EQ(20U, token.location.getStart()); + ASSERT_EQ(21U, token.location.getEnd()); + } + + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertText(dataReader, "math mode", tokens, WhitespaceMode::TRIM, 21, + 30); + assertToken(dataReader, tDollar, "$", tokens, WhitespaceMode::TRIM, 30, + 31); + assertText(dataReader, "they said, see the world they said", tokens, + WhitespaceMode::TRIM, 32, 66); + assertToken(dataReader, tSpeechEnd, ">>", tokens, WhitespaceMode::TRIM, + 66, 68); + assertEnd(dataReader); + } + + TokenizedData data; + ASSERT_FALSE(tokenizer.read(reader, token, data)); +} } -- cgit v1.2.3 From 12e10d18810b7ea4ce142d76e846b4faf0c33488 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Fri, 27 Feb 2015 18:52:43 +0100 Subject: Made OsmlStreamParser ready for user defined tokens, started to adapt unit tests. --- CMakeLists.txt | 32 +- src/formats/osml/OsmlStreamParser.cpp | 701 +++++++++---- src/formats/osml/OsmlStreamParser.hpp | 298 ++---- test/formats/osml/OsmlStreamParserTest.cpp | 1542 ++++++++++++++-------------- 4 files changed, 1355 insertions(+), 1218 deletions(-) (limited to 'test') diff --git a/CMakeLists.txt b/CMakeLists.txt index 75909e9..4e2d7f7 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -212,14 +212,14 @@ ADD_LIBRARY(ousia_core # ousia_core #) -#ADD_LIBRARY(ousia_osml +ADD_LIBRARY(ousia_osml # src/formats/osml/OsmlParser -# src/formats/osml/OsmlStreamParser -#) + src/formats/osml/OsmlStreamParser +) -#TARGET_LINK_LIBRARIES(ousia_osml -# ousia_core -#) +TARGET_LINK_LIBRARIES(ousia_osml + ousia_core +) ADD_LIBRARY(ousia_osxml src/formats/osxml/OsxmlAttributeLocator @@ -383,17 +383,17 @@ IF(TEST) # ousia_mozjs # ) -# ADD_EXECUTABLE(ousia_test_osml + ADD_EXECUTABLE(ousia_test_osml # test/formats/osml/OsmlParserTest -# test/formats/osml/OsmlStreamParserTest -# ) + test/formats/osml/OsmlStreamParserTest + ) -# TARGET_LINK_LIBRARIES(ousia_test_osml -# ${GTEST_LIBRARIES} -# ousia_core -# ousia_osml -# ousia_filesystem -# ) + TARGET_LINK_LIBRARIES(ousia_test_osml + ${GTEST_LIBRARIES} + ousia_core + ousia_osml + ousia_filesystem + ) # ADD_EXECUTABLE(ousia_test_osxml # test/formats/osxml/OsxmlEventParserTest @@ -423,7 +423,7 @@ IF(TEST) ADD_TEST(ousia_test_filesystem ousia_test_filesystem) ADD_TEST(ousia_test_html ousia_test_html) # ADD_TEST(ousia_test_mozjs ousia_test_mozjs) -# ADD_TEST(ousia_test_osml ousia_test_osml) + ADD_TEST(ousia_test_osml ousia_test_osml) # ADD_TEST(ousia_test_osxml ousia_test_osxml) ADD_TEST(ousia_test_xml ousia_test_xml) ENDIF() diff --git a/src/formats/osml/OsmlStreamParser.cpp b/src/formats/osml/OsmlStreamParser.cpp index d4cdbf8..7e01a3c 100644 --- a/src/formats/osml/OsmlStreamParser.cpp +++ b/src/formats/osml/OsmlStreamParser.cpp @@ -19,92 +19,411 @@ #include #include #include +#include #include +#include +#include + #include "OsmlStreamParser.hpp" +#include +#include + namespace ousia { +namespace { /** - * Plain format default tokenizer. + * Osml format default tokenizer. Registers the primary tokens in its + * constructor. A single, static instance of this class is created as + * "OsmlTokens", which is copied to the Tokenizer instance of + * OsmlStreamParserImpl. */ -class PlainFormatTokens : public Tokenizer { +class OsmlFormatTokens : public Tokenizer { public: + TokenId Backslash; + TokenId LineComment; + TokenId BlockCommentStart; + TokenId BlockCommentEnd; + TokenId FieldStart; + TokenId FieldEnd; + TokenId DefaultFieldStart; + TokenId AnnotationStart; + TokenId AnnotationEnd; + /** - * Id of the backslash token. + * Registers the plain format tokens in the internal tokenizer. */ - TokenId Backslash; + OsmlFormatTokens() + { + Backslash = registerToken("\\"); + LineComment = registerToken("%"); + BlockCommentStart = registerToken("%{"); + BlockCommentEnd = registerToken("}%"); + FieldStart = registerToken("{"); + FieldEnd = registerToken("}"); + DefaultFieldStart = registerToken("{!"); + AnnotationStart = registerToken("<\\"); + AnnotationEnd = registerToken("\\>"); + } +}; + +/** + * Instance of OsmlFormatTokens used to initialize the internal tokenizer + * instance of OsmlStreamParserImpl. + */ +static const OsmlFormatTokens OsmlTokens; +/** + * Structure representing a field. + */ +struct Field { /** - * Id of the line comment token. + * Specifies whether this field was marked as default field. */ - TokenId LineComment; + bool defaultField; /** - * Id of the block comment start token. + * Location at which the field was started. */ - TokenId BlockCommentStart; + SourceLocation location; /** - * Id of the block comment end token. + * Constructor of the Field structure, initializes all member variables with + * the given values. + * + * @param defaultField is a flag specifying whether this field is a default + * field. + * @param location specifies the location at which the field was started. */ - TokenId BlockCommentEnd; + Field(bool defaultField = false, + const SourceLocation &location = SourceLocation{}) + : defaultField(defaultField), location(location) + { + } +}; +/** + * Entry used for the command stack. + */ +class Command { +private: /** - * Id of the field start token. + * Name and location of the current command. */ - TokenId FieldStart; + Variant name; /** - * Id of the field end token. + * Arguments that were passed to the command. */ - TokenId FieldEnd; + Variant arguments; /** - * Id of the default field start token. + * Vector used as stack for holding the number of opening/closing braces + * and the corresponding "isDefaultField" flag. */ - TokenId DefaultFieldStart; + std::vector fields; /** - * Id of the annotation start token. + * Set to true if this is a command with clear begin and end. */ - TokenId AnnotationStart; + bool hasRange; +public: /** - * Id of the annotation end token. + * Default constructor, marks this command as normal, non-range command. */ - TokenId AnnotationEnd; + Command() : hasRange(false) {} /** - * Registers the plain format tokens in the internal tokenizer. + * Constructor of the Command class. + * + * @param name is a string variant with name and location of the + * command. + * @param arguments is a map variant with the arguments given to the + * command. + * @param hasRange should be set to true if this is a command with + * explicit range. */ - PlainFormatTokens() + Command(Variant name, Variant arguments, bool hasRange) + : name(std::move(name)), + arguments(std::move(arguments)), + hasRange(hasRange) { - Backslash = registerToken("\\"); - LineComment = registerToken("%"); - BlockCommentStart = registerToken("%{"); - BlockCommentEnd = registerToken("}%"); - FieldStart = registerToken("{"); - FieldEnd = registerToken("}"); - DefaultFieldStart = registerToken("{!"); - AnnotationStart = registerToken("<\\"); - AnnotationEnd = registerToken("\\>"); + } + + /** + * Returns a reference at the variant representing name and location of the + * command. + * + * @return a variant containing name and location of the command. + */ + const Variant &getName() const { return name; } + + /** + * Returns a reference at the variant containing name, value and location of + * the arguments. + * + * @return the arguments stored for the command. + */ + const Variant &getArguments() const { return arguments; } + + /** + * Returns a reference at the internal field list. This list should be used + * for printing error messages when fields are still open although the outer + * range field closes. + * + * @return a const reference at the internal field vector. + */ + const std::vector &getFields() const { return fields; } + + /** + * Returns true if this command is currently in a default field. + * + * @return true if the current field on the field stack was explicitly + * marked as default field. If the field stack is empty, true is returned + * if this is a range command. + */ + bool inDefaultField() const + { + return (!fields.empty() && fields.back().defaultField) || + (fields.empty() && hasRange); + } + + /** + * Returns true if this command currently is in any field. + * + * @return true if a field is on the stack or this is a range commands. + * Range commands always are in a field. + */ + bool inField() const { return !fields.empty() || hasRange; } + + /** + * Returns true if this command currently is in a range field. + * + * @return true if the command has a range and no other ranges are on the + * stack. + */ + bool inRangeField() const { return fields.empty() && hasRange; } + + /** + * Returns true if this command currently is in a non-range field. + * + * @return true if the command is in a field, but the field is not the field + * constructed by the "range" + */ + bool inNonRangeField() const { return !fields.empty(); } + + /** + * Pushes another field onto the field stack of this command. + * + * @param defaultField if true, explicitly marks this field as default + * field. + * @param location is the source location at which the field was started. + * Used for error messages in which the user is notified about an error with + * too few closing fields. + */ + void pushField(bool defaultField = false, + const SourceLocation &location = SourceLocation{}) + { + fields.emplace_back(defaultField, location); + } + + /** + * Removes another field from the field stack of this command, returns true + * if the operation was successful. + * + * @return true if there was a field to pop on the stack, false otherwise. + */ + bool popField() + { + if (!fields.empty()) { + fields.pop_back(); + return true; + } + return false; } }; +} -static const PlainFormatTokens OsmlTokens; +/* Class OsmlStreamParserImpl */ -OsmlStreamParser::OsmlStreamParser(CharReader &reader, Logger &logger) - : reader(reader), - logger(logger), - tokenizer(OsmlTokens), - data(reader.getSourceId()) +/** + * Internal implementation of OsmlStreamParser. + */ +class OsmlStreamParserImpl { +public: + /** + * State enum compatible with OsmlStreamParserState but extended by two more + * entries (END and NONE). + */ + enum class State : uint8_t { + COMMAND_START = 0, + COMMAND_END = 1, + FIELD_START = 2, + FIELD_END = 3, + ANNOTATION_START = 4, + ANNOTATION_END = 5, + DATA = 6, + END = 7, + RECOVERABLE_ERROR = 8, + IRRECOVERABLE_ERROR = 9 + }; + +private: + /** + * Reference to the CharReader instance from which the incomming bytes are + * read. + */ + CharReader &reader; + + /** + * Reference at the logger instance to which all error messages are sent. + */ + Logger &logger; + + /** + * Tokenizer instance used to read individual tokens from the text. + */ + Tokenizer tokenizer; + + /** + * Stack containing the current commands. + */ + std::stack commands; + + /** + * Variant containing the tokenized data that was returned from the + * tokenizer as data. + */ + TokenizedData data; + + /** + * Variable containing the current location of the parser. + */ + SourceLocation location; + + /** + * Function used internally to parse an identifier. + * + * @param start is the start byte offset of the identifier (including the + * backslash). + * @param allowNSSep should be set to true if the namespace separator is + * allowed in the identifier name. Issues error if the namespace separator + * is placed incorrectly. + */ + Variant parseIdentifier(size_t start, bool allowNSSep = false); + + /** + * Function used internally to handle the special "\begin" command. + * + * @return an internal State specifying whether an error occured (return + * values State::REOVERABLE_ERROR or State::IRRECOVERABLE_ERROR) or a + * command was actually started (return value State::COMMAND_START). + */ + State parseBeginCommand(); + + /** + * Function used internally to handle the special "\end" command. + * + * @return an internal State specifying whether an error occured (return + * values State::REOVERABLE_ERROR or State::IRRECOVERABLE_ERROR) or a + * command was actually ended (return value State::COMMAND_END). + */ + State parseEndCommand(); + + /** + * Parses the command arguments. Handles errors if the name of the command + * was given using the hash notation and as a name field. + * + * @param commandArgName is the name argument that was given using the hash + * notation. + * @return a map variant containing the arguments. + */ + Variant parseCommandArguments(Variant commandArgName); + + /** + * Function used internally to parse a command. + * + * @param start is the start byte offset of the command (including the + * backslash) + * @param isAnnotation if true, the command is not returned as command, but + * as annotation start. + * @return true if a command was actuall parsed, false otherwise. + */ + State parseCommand(size_t start, bool isAnnotation); + + /** + * Function used internally to parse a block comment. + */ + void parseBlockComment(); + + /** + * Function used internally to parse a generic comment. + */ + void parseLineComment(); + + /** + * Pushes the parsed command onto the command stack. + */ + void pushCommand(Variant commandName, Variant commandArguments, + bool hasRange); + + /** + * Checks whether there is any data pending to be issued, if yes, resets the + * currently peeked characters and returns true. + * + * @return true if there was any data and DATA should be returned by the + * parse function, false otherwise. + */ + bool checkIssueData(); + + /** + * Returns a reference at the current command at the top of the command + * stack. + * + * @return a reference at the top command in the command stack. + */ + Command &cmd() { return commands.top(); } + + /** + * Returns a reference at the current command at the top of the command + * stack. + * + * @return a reference at the top command in the command stack. + */ + const Command &cmd() const { return commands.top(); } + +public: + /** + * Constructor of the OsmlStreamParserImpl class. Attaches the new + * OsmlStreamParserImpl to the given CharReader and Logger instances. + * + * @param reader is the reader instance from which incomming characters + * should be read. + * @param logger is the logger instance to which errors should be written. + */ + OsmlStreamParserImpl(CharReader &reader, Logger &logger); + + State parse(); + + const TokenizedData &getData() const { return data; } + const Variant &getCommandName() const { return cmd().getName(); } + const Variant &getCommandArguments() const { return cmd().getArguments(); } + const SourceLocation &getLocation() const { return location; } + bool inRangeCommand() const { return cmd().inRangeField(); }; + bool inDefaultField() const { return cmd().inDefaultField(); } +}; + +/* Class OsmlStreamParserImpl */ + +OsmlStreamParserImpl::OsmlStreamParserImpl(CharReader &reader, Logger &logger) + : reader(reader), logger(logger), tokenizer(OsmlTokens) { - // Place an intial command representing the complete file on the stack - commands.push(Command{"", Variant::mapType{}, true, true, true, false}); + commands.emplace("", Variant::mapType{}, true); } -Variant OsmlStreamParser::parseIdentifier(size_t start, bool allowNSSep) +Variant OsmlStreamParserImpl::parseIdentifier(size_t start, bool allowNSSep) { bool first = true; bool hasCharSinceNSSep = false; @@ -147,20 +466,20 @@ Variant OsmlStreamParser::parseIdentifier(size_t start, bool allowNSSep) return res; } -OsmlStreamParser::State OsmlStreamParser::parseBeginCommand() +OsmlStreamParserImpl::State OsmlStreamParserImpl::parseBeginCommand() { // Expect a '{' after the command reader.consumeWhitespace(); if (!reader.expect('{')) { logger.error("Expected \"{\" after \\begin", reader); - return State::NONE; + return State::RECOVERABLE_ERROR; } // Parse the name of the command that should be opened Variant commandName = parseIdentifier(reader.getOffset(), true); if (commandName.asString().empty()) { logger.error("Expected identifier", commandName); - return State::ERROR; + return State::IRRECOVERABLE_ERROR; } // Check whether the next character is a '#', indicating the start of the @@ -176,7 +495,7 @@ OsmlStreamParser::State OsmlStreamParser::parseBeginCommand() if (!reader.expect('}')) { logger.error("Expected \"}\"", reader); - return State::ERROR; + return State::IRRECOVERABLE_ERROR; } // Parse the arguments @@ -185,28 +504,15 @@ OsmlStreamParser::State OsmlStreamParser::parseBeginCommand() // Push the command onto the command stack pushCommand(std::move(commandName), std::move(commandArguments), true); - return State::COMMAND; -} - -static bool checkStillInField(const OsmlStreamParser::Command &cmd, - const Variant &endName, Logger &logger) -{ - if (cmd.inField && !cmd.inRangeField) { - logger.error(std::string("\\end in open field of command \"") + - cmd.name.asString() + std::string("\""), - endName); - logger.note(std::string("Open command started here:"), cmd.name); - return true; - } - return false; + return State::COMMAND_START; } -OsmlStreamParser::State OsmlStreamParser::parseEndCommand() +OsmlStreamParserImpl::State OsmlStreamParserImpl::parseEndCommand() { // Expect a '{' after the command if (!reader.expect('{')) { logger.error("Expected \"{\" after \\end", reader); - return State::NONE; + return State::RECOVERABLE_ERROR; } // Fetch the name of the command that should be ended here @@ -215,56 +521,58 @@ OsmlStreamParser::State OsmlStreamParser::parseEndCommand() // Make sure the given command name is not empty if (name.asString().empty()) { logger.error("Expected identifier", name); - return State::ERROR; + return State::IRRECOVERABLE_ERROR; } // Make sure the command name is terminated with a '}' if (!reader.expect('}')) { logger.error("Expected \"}\"", reader); - return State::ERROR; + return State::IRRECOVERABLE_ERROR; } - // Unroll the command stack up to the last range command - while (!commands.top().hasRange) { - if (checkStillInField(commands.top(), name, logger)) { - return State::ERROR; + // Unroll the command stack up to the last range command, make sure we do + // not intersect with any open field + while (!cmd().inRangeField()) { + if (cmd().inField()) { + logger.error(std::string("\\end in open field of command \"") + + cmd().getName().asString() + std::string("\""), + name); + const std::vector &fields = cmd().getFields(); + for (const Field &field : fields) { + logger.note(std::string("Still open field started here: "), + field.location); + } + return State::IRRECOVERABLE_ERROR; } commands.pop(); } - // Make sure we're not in an open field of this command - if (checkStillInField(commands.top(), name, logger)) { - return State::ERROR; - } - // Special error message if the top-level command is reached if (commands.size() == 1) { logger.error(std::string("Cannot end command \"") + name.asString() + std::string("\" here, no command open"), name); - return State::ERROR; + return State::IRRECOVERABLE_ERROR; } - // Inform the about command mismatches - const Command &cmd = commands.top(); - if (commands.top().name.asString() != name.asString()) { - logger.error(std::string("Trying to end command \"") + - cmd.name.asString() + + // Inform the user about command mismatches, copy the current command + // descriptor before popping it from the stack + if (getCommandName().asString() != name.asString()) { + logger.error(std::string("Trying to end command \"") + name.asString() + std::string("\", but open command is \"") + - name.asString() + std::string("\""), + getCommandName().asString() + std::string("\""), name); - logger.note("Last command was opened here:", cmd.name); - return State::ERROR; + logger.note("Open command started here:", getCommandName()); + return State::IRRECOVERABLE_ERROR; } - // Set the location to the location of the command that was ended, then end - // the current command + // End the current command location = name.getLocation(); commands.pop(); - return cmd.inRangeField ? State::FIELD_END : State::NONE; + return State::COMMAND_END; } -Variant OsmlStreamParser::parseCommandArguments(Variant commandArgName) +Variant OsmlStreamParserImpl::parseCommandArguments(Variant commandArgName) { // Parse the arguments using the universal VariantReader Variant commandArguments; @@ -290,29 +598,14 @@ Variant OsmlStreamParser::parseCommandArguments(Variant commandArgName) return commandArguments; } -void OsmlStreamParser::pushCommand(Variant commandName, - Variant commandArguments, bool hasRange) -{ - // Store the location on the stack - location = commandName.getLocation(); - - // Place the command on the command stack, remove the last commands if we're - // not currently inside a field of these commands - while (!commands.top().inField) { - commands.pop(); - } - commands.push(Command{std::move(commandName), std::move(commandArguments), - hasRange, false, false, false}); -} - -OsmlStreamParser::State OsmlStreamParser::parseCommand(size_t start, - bool isAnnotation) +OsmlStreamParserImpl::State OsmlStreamParserImpl::parseCommand( + size_t start, bool isAnnotation) { // Parse the commandName as a first identifier Variant commandName = parseIdentifier(start, true); if (commandName.asString().empty()) { logger.error("Empty command name", reader); - return State::NONE; + return State::RECOVERABLE_ERROR; } // Handle the special "begin" and "end" commands @@ -322,7 +615,7 @@ OsmlStreamParser::State OsmlStreamParser::parseCommand(size_t start, const bool isEnd = commandNameComponents[0] == "end"; // Parse the begin or end command - State res = State::COMMAND; + State res = State::COMMAND_START; if (isBegin || isEnd) { if (commandNameComponents.size() > 1) { logger.error( @@ -378,12 +671,13 @@ OsmlStreamParser::State OsmlStreamParser::parseCommand(size_t start, } else { // Make sure no arguments apart from the "name" argument are given // to an annotation end - Variant::mapType &map = commands.top().arguments.asMap(); + const Variant::mapType &map = getCommandArguments().asMap(); if (!map.empty()) { if (map.count("name") == 0 || map.size() > 1U) { logger.error( "An annotation end command may not have any arguments " - "other than \"name\""); + "other than \"name\"", + reader); return res; } } @@ -397,13 +691,13 @@ OsmlStreamParser::State OsmlStreamParser::parseCommand(size_t start, // If we're starting an annotation, return the command as annotation start // instead of command - if (isAnnotation && res == State::COMMAND) { + if (isAnnotation && res == State::COMMAND_START) { return State::ANNOTATION_START; } return res; } -void OsmlStreamParser::parseBlockComment() +void OsmlStreamParserImpl::parseBlockComment() { Token token; size_t depth = 1; @@ -426,7 +720,7 @@ void OsmlStreamParser::parseBlockComment() logger.error("File ended while being in a block comment", reader); } -void OsmlStreamParser::parseLineComment() +void OsmlStreamParserImpl::parseLineComment() { char c; while (reader.read(c)) { @@ -436,65 +730,34 @@ void OsmlStreamParser::parseLineComment() } } -bool OsmlStreamParser::checkIssueData() -{ - if (!data.empty()) { - location = data.getLocation(); - reader.resetPeek(); - return true; - } - return false; -} - -bool OsmlStreamParser::checkIssueFieldStart() +void OsmlStreamParserImpl::pushCommand(Variant commandName, + Variant commandArguments, bool hasRange) { - // Fetch the current command, and check whether we're currently inside a - // field of this command - Command &cmd = commands.top(); - if (!cmd.inField) { - // If this is a range command, we're now implicitly inside the field of - // this command -- we'll have to issue a field start command! - if (cmd.hasRange) { - cmd.inField = true; - cmd.inRangeField = true; - reader.resetPeek(); - return true; - } + // Store the location of the command + location = commandName.getLocation(); - // This was not a range command, so obviously we're now inside within - // a field of some command -- so unroll the commands stack until a - // command with open field is reached - while (!commands.top().inField) { - commands.pop(); - } + // Place the command on the command stack, remove the last commands if we're + // not currently inside a field of these commands + while (!cmd().inField()) { + commands.pop(); } - return false; + + // Push the new command onto the command stack + commands.emplace(std::move(commandName), std::move(commandArguments), + hasRange); } -bool OsmlStreamParser::closeField() +bool OsmlStreamParserImpl::checkIssueData() { - // Try to end an open field of the current command -- if the current command - // is not inside an open field, end this command and try to close the next - // one - for (int i = 0; i < 2 && commands.size() > 1; i++) { - Command &cmd = commands.top(); - if (!cmd.inRangeField) { - if (cmd.inField) { - cmd.inField = false; - if (cmd.inDefaultField) { - commands.pop(); - } - return true; - } - commands.pop(); - } else { - return false; - } + if (!data.empty()) { + location = data.getLocation(); + reader.resetPeek(); + return true; } return false; } -OsmlStreamParser::State OsmlStreamParser::parse() +OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() { // Reset the data handler data.clear(); @@ -507,14 +770,6 @@ OsmlStreamParser::State OsmlStreamParser::parse() // Special handling for Backslash and Text if (type == OsmlTokens.Backslash || type == OsmlTokens.AnnotationStart) { - // Before appending anything to the output data or starting a new - // command, check whether FIELD_START has to be issued, as the - // current command is a command with range - if (checkIssueFieldStart()) { - location = token.location; - return State::FIELD_START; - } - // Check whether a command starts now, without advancing the peek // cursor char c; @@ -535,11 +790,11 @@ OsmlStreamParser::State OsmlStreamParser::parse() State res = parseCommand(token.location.getStart(), type == OsmlTokens.AnnotationStart); switch (res) { - case State::ERROR: + case State::IRRECOVERABLE_ERROR: throw LoggableException( "Last error was irrecoverable, ending parsing " "process"); - case State::NONE: + case State::RECOVERABLE_ERROR: continue; default: return res; @@ -558,15 +813,12 @@ OsmlStreamParser::State OsmlStreamParser::parse() token.location.getStart() + 1); } - data.append(c, token.location.getStart(), reader.getPeekOffset()); + // Append the character to the output data, mark it as protected + data.append(c, token.location.getStart(), reader.getPeekOffset(), + true); reader.consumePeek(); continue; } else if (type == Tokens::Data) { - // Check whether FIELD_START has to be issued before appending text - if (checkIssueFieldStart()) { - location = token.location; - return State::FIELD_START; - } reader.consumePeek(); continue; } @@ -580,7 +832,7 @@ OsmlStreamParser::State OsmlStreamParser::parse() // We will handle the token now, consume the peeked characters reader.consumePeek(); - // Update the location to the current token location + // Synchronize the location with the current token location location = token.location; if (token.id == OsmlTokens.LineComment) { @@ -588,39 +840,27 @@ OsmlStreamParser::State OsmlStreamParser::parse() } else if (token.id == OsmlTokens.BlockCommentStart) { parseBlockComment(); } else if (token.id == OsmlTokens.FieldStart) { - Command &cmd = commands.top(); - if (!cmd.inField) { - cmd.inField = true; - } + cmd().pushField(false, token.location); return State::FIELD_START; -/* logger.error( - "Got field start token \"{\", but no command for which to " - "start the field. Write \"\\{\" to insert this sequence as " - "text.", - token);*/ } else if (token.id == OsmlTokens.FieldEnd) { - closeField(); - return State::FIELD_END; -/* if (closeField()) { + // Remove all commands from the list that currently are not in any + // field + while (!cmd().inField()) { + commands.pop(); + } + + // If the remaining command is not in a range field, remove this + // command + if (cmd().inNonRangeField()) { + cmd().popField(); return State::FIELD_END; } logger.error( - "Got field end token \"}\", but there is no field to end. " - "Write \"\\}\" to insert this sequence as text.", - token);*/ + "Got field end token \"}\", but there is no field to end.", + token); } else if (token.id == OsmlTokens.DefaultFieldStart) { - // Try to start a default field the first time the token is reached - Command &topCmd = commands.top(); - if (!topCmd.inField) { - topCmd.inField = true; - topCmd.inDefaultField = true; - } + cmd().pushField(true, token.location); return State::FIELD_START; -/* logger.error( - "Got default field start token \"{!\", but no command for " - "which to start the field. Write \"\\{!\" to insert this " - "sequence as text", - token);*/ } else if (token.id == OsmlTokens.AnnotationEnd) { // We got a single annotation end token "\>" -- simply issue the // ANNOTATION_END event @@ -641,11 +881,25 @@ OsmlStreamParser::State OsmlStreamParser::parse() // Make sure all open commands and fields have been ended at the end of the // stream while (commands.size() > 1) { - Command &cmd = commands.top(); - if (cmd.inField || cmd.hasRange) { - logger.error("Reached end of stream, but command \"" + - cmd.name.asString() + "\" has not been ended", - cmd.name); + if (cmd().inField()) { + // If the stream ended with an open range field, issue information + // about the range field + if (cmd().inRangeField()) { + // Inform about the still open command itself + logger.error("Reached end of stream, but command \"" + + getCommandName().asString() + + "\" has not been ended", + getCommandName()); + } else { + // Issue information about still open fields + const std::vector &fields = cmd().getFields(); + if (!fields.empty()) { + logger.error( + std::string( + "Reached end of stream, but field is still open."), + fields.back().location); + } + } } commands.pop(); } @@ -654,26 +908,45 @@ OsmlStreamParser::State OsmlStreamParser::parse() return State::END; } -Variant OsmlStreamParser::getText(WhitespaceMode mode) +/* Class OsmlStreamParser */ + +OsmlStreamParser::OsmlStreamParser(CharReader &reader, Logger &logger) + : impl(new OsmlStreamParserImpl(reader, logger)) +{ +} + +OsmlStreamParser::~OsmlStreamParser() +{ + // Stub needed because OsmlStreamParserImpl is incomplete in header +} + +OsmlStreamParser::State OsmlStreamParser::parse() +{ + return static_cast(impl->parse()); +} + +const TokenizedData &OsmlStreamParser::getData() const { - TokenizedData dataFork = data; - Variant text = dataFork.text(mode); - location = text.getLocation(); - return text; + return impl->getData(); } const Variant &OsmlStreamParser::getCommandName() const { - return commands.top().name; + return impl->getCommandName(); } const Variant &OsmlStreamParser::getCommandArguments() const { - return commands.top().arguments; + return impl->getCommandArguments(); } -bool OsmlStreamParser::inDefaultField() const +const SourceLocation &OsmlStreamParser::getLocation() const { - return commands.top().inRangeField || commands.top().inDefaultField; + return impl->getLocation(); } + +bool OsmlStreamParser::inDefaultField() const { return impl->inDefaultField(); } + +bool OsmlStreamParser::inRangeCommand() const { return impl->inRangeCommand(); } + } diff --git a/src/formats/osml/OsmlStreamParser.hpp b/src/formats/osml/OsmlStreamParser.hpp index 453a2bb..1fee90b 100644 --- a/src/formats/osml/OsmlStreamParser.hpp +++ b/src/formats/osml/OsmlStreamParser.hpp @@ -29,30 +29,29 @@ #ifndef _OUSIA_OSML_STREAM_PARSER_HPP_ #define _OUSIA_OSML_STREAM_PARSER_HPP_ +#include #include -#include -#include -#include -#include - namespace ousia { // Forward declarations class CharReader; class Logger; class OsmlStreamParserImpl; +class TokenizedData; +class Variant; /** * The OsmlStreamParser class provides a low-level reader for the TeX-esque osml * format. The parser is constructed around a "parse" function, which reads data * from the underlying CharReader until a new state is reached and indicates * this state in a return value. The calling code then has to pull corresponding - * data from the stream reader. The reader makes sure the incommind file is + * data from the stream reader. The reader makes sure the incomming stream is * syntactically valid and tries to recorver from most errors. If an error is * irrecoverable (this is the case for errors with wrong nesting of commands or * fields, as this would lead to too many consecutive errors) a - * LoggableException is thrown. + * LoggableException is thrown. The OsmlStreamParser can be compared to a SAX + * parser for XML. */ class OsmlStreamParser { public: @@ -60,39 +59,21 @@ public: * Enum used to indicate which state the OsmlStreamParser class is in * after calling the "parse" function. */ - enum class State { - /** - * State returned if a fully featured command has been read. A command - * consists of the command name and its arguments (which optionally - * includes the name). - */ - COMMAND, - - /** - * State returned if data is given. The reader must decide which field - * or command this should be routed to. Trailing or leading whitespace - * has been removed. Only called if the data is non-empty. - */ - DATA, - + enum class State : uint8_t { /** - * A user-defined entity has been found. The entity sequence is stored - * in the command name. + * State returned if the start of a command has been read. Use the + * getCommandName(), getCommandArguments() and inRangeCommand() + * functions the retrieve more information about the command that was + * just started. */ - ENTITY, + COMMAND_START = 0, /** - * State returned if an annotation was started. An annotation consists - * of the command name and its arguments (which optionally include the - * name). + * State returned if a range command has just ended. This state is not + * returned for non-range commands (as the actual end of a command is + * context dependant). */ - ANNOTATION_START, - - /** - * State returned if an annotation ends. The reader indicates which - * annotation ends. - */ - ANNOTATION_END, + COMMAND_END = 1, /** * State returned if a new field started. The reader assures that the @@ -100,200 +81,47 @@ public: * is not started if data has been given outside of a field. The * field number is set to the current field index. */ - FIELD_START, + FIELD_START = 2, /** * State returned if the current field ends. The reader assures that a * field was actually open. */ - FIELD_END, + FIELD_END = 3, /** - * The end of the stream has been reached. + * State returned if an annotation was started. An annotation consists + * of the command name and its arguments (which optionally include the + * name). */ - END, + ANNOTATION_START = 4, /** - * Returned from internal functions if nothing should be done. + * State returned if an annotation ends. The reader indicates which + * annotation ends. */ - NONE, + ANNOTATION_END = 5, /** - * Returned from internal function to indicate irrecoverable errors. + * State returned if data is given. The reader must decide which field + * or command this should be routed to. Trailing or leading whitespace + * has been removed. Only called if the data is non-empty. */ - ERROR - }; - - /** - * Entry used for the command stack. - */ - struct Command { - /** - * Name and location of the current command. - */ - Variant name; - - /** - * Arguments that were passed to the command. - */ - Variant arguments; - - /** - * Vector used as stack for holding the number of opening/closing braces - * and the corresponding "isDefaultField" flag. - */ - std::vector fields; - - /** - * Set to true if this is a command with clear begin and end. - */ - bool hasRange; - - /** - * Default constructor. - */ - Command() - : hasRange(false), - inField(false), - inDefaultField() - { - } + DATA = 6, /** - * Constructor of the Command class. - * - * @param name is a string variant with name and location of the - * command. - * @param arguments is a map variant with the arguments given to the - * command. - * @param hasRange should be set to true if this is a command with - * explicit range. - * @param inDefaultField is set to true if we currently are in a - * specially marked default field. - */ - Command(Variant name, Variant arguments, bool hasRange) - : name(std::move(name)), - arguments(std::move(arguments)), - hasRange(hasRange), - inField(inField), - inRangeField(inRangeField), - inDefaultField(inDefaultField) - { - } + * The end of the stream has been reached. + */ + END = 7 }; private: /** - * Reference to the CharReader instance from which the incomming bytes are - * read. - */ - CharReader &reader; - - /** - * Reference at the logger instance to which all error messages are sent. - */ - Logger &logger; - - /** - * Tokenizer instance used to read individual tokens from the text. - */ - Tokenizer tokenizer; - - /** - * Variant containing the tokenized data that was returned from the - * tokenizer as data. - */ - TokenizedData data; - - /** - * Stack containing the current commands. - */ - std::stack commands; - - /** - * Pointer at + * Pointer at the class containing the internal implementation (according + * to the PIMPL idiom). */ std::unique_ptr impl; - /** - * Function used internall to parse an identifier. - * - * @param start is the start byte offset of the identifier (including the - * backslash). - * @param allowNSSep should be set to true if the namespace separator is - * allowed in the identifier name. Issues error if the namespace separator - * is placed incorrectly. - */ - Variant parseIdentifier(size_t start, bool allowNSSep = false); - - /** - * Function used internally to handle the special "\begin" command. - */ - State parseBeginCommand(); - - /** - * Function used internally to handle the special "\end" command. - */ - State parseEndCommand(); - - /** - * Pushes the parsed command onto the command stack. - */ - void pushCommand(Variant commandName, Variant commandArguments, - bool hasRange); - - /** - * Parses the command arguments. - */ - Variant parseCommandArguments(Variant commandArgName); - - /** - * Function used internally to parse a command. - * - * @param start is the start byte offset of the command (including the - * backslash) - * @param isAnnotation if true, the command is not returned as command, but - * as annotation start. - * @return true if a command was actuall parsed, false otherwise. - */ - State parseCommand(size_t start, bool isAnnotation); - - /** - * Function used internally to parse a block comment. - */ - void parseBlockComment(); - - /** - * Function used internally to parse a generic comment. - */ - void parseLineComment(); - - /** - * Checks whether there is any data pending to be issued, if yes, issues it. - * - * @return true if there was any data and DATA should be returned by the - * parse function, false otherwise. - */ - bool checkIssueData(); - - /** - * Called before any data is appended to the internal data handler. Checks - * whether a new field should be started or implicitly ended. - * - * @return true if FIELD_START should be returned by the parse function. - */ - bool checkIssueFieldStart(); - - /** - * Closes a currently open field. Note that the command will be removed from - * the internal command stack if the field that is being closed is a - * field marked as default field. - * - * @return true if the field could be closed, false if there was no field - * to close. - */ - bool closeField(); - public: /** * Constructor of the OsmlStreamParser class. Attaches the new @@ -321,30 +149,10 @@ public: */ State parse(); - /** - * Returns a reference at the internally stored data. Only valid if - * State::DATA was returned by the "parse" function. - * - * @return a reference at a variant containing the data parsed by the - * "parse" function. - */ - const TokenizedData &getData() const { return data; } - - /** - * Returns the complete content of the internal TokenizedData instance as - * a single string Variant. This method is mainly used in the unit tests for - * this class, it simply calls the text() method of TokenizedData. - * - * @param mode is the WhitespaceMode that should be used for returning the - * text. - * @return a string variant containing the text content of the internal - * TokenizedData instance or a nullptr variant if there is no text. - */ - Variant getText(WhitespaceMode mode = WhitespaceMode::COLLAPSE); - /** * Returns a reference at the internally stored command name. Only valid if - * State::COMMAND was returned by the "parse" function. + * State::COMMAND_START, State::ANNOTATION_START or State::ANNOTATION_END + * was returned by the "parse" function. * * @return a reference at a variant containing name and location of the * parsed command. @@ -353,18 +161,46 @@ public: /** * Returns a reference at the internally stored command name. Only valid if - * State::COMMAND was returned by the "parse" function. + * State::COMMAND_START, State::ANNOTATION_START or State::ANNOTATION_END + * was returned by the "parse" function. * * @return a reference at a variant containing arguments given to the * command. */ const Variant &getCommandArguments() const; + /** + * Returns a reference at the internally stored data. Only valid if + * State::DATA was returned by the "parse" function. + * + * @return a reference at a variant containing the data parsed by the + * "parse" function. + */ + const TokenizedData &getData() const; + + /** + * Returns the location of the current token. + */ + const SourceLocation &getLocation() const; + + /** + * Returns true if the currently started command is a range command, only + * valid if State::COMMAND_START was returned by the "parse" function. + * + * @return true if the command is started is a range command, false + * otherwise. + */ + bool inRangeCommand() const; + /** * Returns true if the current field is the "default" field. This is true if * the parser either is in the outer range of a range command or inside a - * field that has been especially marked as "default" field (using the "|" - * syntax). + * field that has been especially marked as "default" field (using the "{!" + * syntax). Only valid if State::FIELD_START was returned by the "parse" + * function. + * + * @return true if the current field was marked as default field (using the + * "{!" syntax). */ bool inDefaultField() const; }; diff --git a/test/formats/osml/OsmlStreamParserTest.cpp b/test/formats/osml/OsmlStreamParserTest.cpp index 3d01007..8b64e51 100644 --- a/test/formats/osml/OsmlStreamParserTest.cpp +++ b/test/formats/osml/OsmlStreamParserTest.cpp @@ -21,7 +21,9 @@ #include #include +#include #include +#include #include @@ -30,147 +32,196 @@ namespace ousia { static TerminalLogger logger(std::cerr, true); // static ConcreteLogger logger; -static OsmlStreamParser::State skipEmptyData(OsmlStreamParser &reader) +static void assertCommandStart(OsmlStreamParser &parser, + const std::string &name, + bool rangeCommand, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) { - OsmlStreamParser::State res = reader.parse(); - if (res == OsmlStreamParser::State::DATA) { - EXPECT_FALSE(reader.getData().hasNonWhitespaceText()); - res = reader.parse(); - } - return res; -} - -static void assertCommand(OsmlStreamParser &reader, const std::string &name, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset) -{ - ASSERT_EQ(OsmlStreamParser::State::COMMAND, skipEmptyData(reader)); - EXPECT_EQ(name, reader.getCommandName().asString()); + ASSERT_EQ(OsmlStreamParser::State::COMMAND_START, parser.parse()); + EXPECT_EQ(name, parser.getCommandName().asString()); + EXPECT_EQ(rangeCommand, parser.inRangeCommand()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getCommandName().getLocation().getStart()); - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getCommandName().getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getCommandName().getLocation().getEnd()); - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getCommandName().getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } -static void assertCommand(OsmlStreamParser &reader, const std::string &name, - const Variant::mapType &args, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset) +static void assertCommandStart(OsmlStreamParser &parser, + const std::string &name, + bool rangeCommand, + const Variant::mapType &args, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) { - assertCommand(reader, name, start, end); - EXPECT_EQ(args, reader.getCommandArguments()); + assertCommandStart(parser, name, rangeCommand, start, end); + EXPECT_EQ(args, parser.getCommandArguments()); } -static void assertData(OsmlStreamParser &reader, const std::string &data, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset, - WhitespaceMode mode = WhitespaceMode::COLLAPSE) +static void assertCommand(OsmlStreamParser &parser, + const std::string &name, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - Variant text = reader.getText(mode); - ASSERT_TRUE(text.isString()); - EXPECT_EQ(data, text.asString()); + assertCommandStart(parser, name, false, Variant::mapType{}, start, end); +} + +static void assertCommandEnd(OsmlStreamParser &parser, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) +{ + ASSERT_EQ(OsmlStreamParser::State::COMMAND_END, parser.parse()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, text.getLocation().getStart()); - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, text.getLocation().getEnd()); - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); + } +} + +static void assertTextData(OsmlStreamParser &parser, const std::string &text, + SourceOffset dataStart = InvalidSourceOffset, + SourceOffset dataEnd = InvalidSourceOffset, + SourceOffset textStart = InvalidSourceOffset, + SourceOffset textEnd = InvalidSourceOffset, + WhitespaceMode mode = WhitespaceMode::COLLAPSE) +{ + ASSERT_EQ(OsmlStreamParser::State::DATA, parser.parse()); + + const TokenizedData &data = parser.getData(); + TokenizedDataReader dataReader = data.reader(); + + Token token; + ASSERT_TRUE(dataReader.read(token, TokenSet{}, mode)); + EXPECT_EQ(Tokens::Data, token.id); + EXPECT_EQ(text, token.content); + if (dataStart != InvalidSourceOffset) { + EXPECT_EQ(dataStart, data.getLocation().getStart()); + EXPECT_EQ(dataStart, parser.getLocation().getStart()); + } + if (dataEnd != InvalidSourceOffset) { + EXPECT_EQ(dataEnd, data.getLocation().getEnd()); + EXPECT_EQ(dataEnd, parser.getLocation().getEnd()); + } + if (textStart != InvalidSourceOffset) { + EXPECT_EQ(textStart, token.getLocation().getStart()); } + if (textEnd != InvalidSourceOffset) { + EXPECT_EQ(textEnd, token.getLocation().getEnd()); + } +} + +static void assertData(OsmlStreamParser &parser, const std::string &text, + SourceOffset textStart = InvalidSourceOffset, + SourceOffset textEnd = InvalidSourceOffset, + WhitespaceMode mode = WhitespaceMode::COLLAPSE) +{ + assertTextData(parser, text, InvalidSourceOffset, InvalidSourceOffset, textStart, textEnd, mode); +} + +static void assertEmptyData(OsmlStreamParser &parser) +{ + ASSERT_EQ(OsmlStreamParser::State::DATA, parser.parse()); + + const TokenizedData &data = parser.getData(); + TokenizedDataReader dataReader = data.reader(); + + Token token; + EXPECT_FALSE(dataReader.read(token, TokenSet{}, WhitespaceMode::TRIM)); } -static void assertFieldStart(OsmlStreamParser &reader, bool defaultField, + +static void assertFieldStart(OsmlStreamParser &parser, bool defaultField, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::FIELD_START, skipEmptyData(reader)); - EXPECT_EQ(defaultField, reader.inDefaultField()); + ASSERT_EQ(OsmlStreamParser::State::FIELD_START, parser.parse()); + EXPECT_EQ(defaultField, parser.inDefaultField()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } -static void assertFieldEnd(OsmlStreamParser &reader, +static void assertFieldEnd(OsmlStreamParser &parser, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::FIELD_END, skipEmptyData(reader)); + ASSERT_EQ(OsmlStreamParser::State::FIELD_END, parser.parse()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } -static void assertAnnotationStart(OsmlStreamParser &reader, +static void assertAnnotationStart(OsmlStreamParser &parser, const std::string &name, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_START, skipEmptyData(reader)); - EXPECT_EQ(name, reader.getCommandName().asString()); + ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_START, parser.parse()); + EXPECT_EQ(name, parser.getCommandName().asString()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getCommandName().getLocation().getStart()); - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getCommandName().getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getCommandName().getLocation().getEnd()); - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getCommandName().getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } -static void assertAnnotationStart(OsmlStreamParser &reader, +static void assertAnnotationStart(OsmlStreamParser &parser, const std::string &name, const Variant::mapType &args, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - assertAnnotationStart(reader, name, start, end); - EXPECT_EQ(args, reader.getCommandArguments()); + assertAnnotationStart(parser, name, start, end); + EXPECT_EQ(args, parser.getCommandArguments()); } -static void assertAnnotationEnd(OsmlStreamParser &reader, +static void assertAnnotationEnd(OsmlStreamParser &parser, const std::string &name, const std::string &elementName, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_END, skipEmptyData(reader)); - ASSERT_EQ(name, reader.getCommandName().asString()); + ASSERT_EQ(OsmlStreamParser::State::ANNOTATION_END, parser.parse()); + ASSERT_EQ(name, parser.getCommandName().asString()); if (!elementName.empty()) { - ASSERT_EQ(1U, reader.getCommandArguments().asMap().size()); - ASSERT_EQ(1U, reader.getCommandArguments().asMap().count("name")); + ASSERT_EQ(1U, parser.getCommandArguments().asMap().size()); + ASSERT_EQ(1U, parser.getCommandArguments().asMap().count("name")); - auto it = reader.getCommandArguments().asMap().find("name"); + auto it = parser.getCommandArguments().asMap().find("name"); ASSERT_EQ(elementName, it->second.asString()); } if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } -static void assertEnd(OsmlStreamParser &reader, +static void assertEnd(OsmlStreamParser &parser, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::END, skipEmptyData(reader)); + ASSERT_EQ(OsmlStreamParser::State::END, parser.parse()); if (start != InvalidSourceOffset) { - EXPECT_EQ(start, reader.getLocation().getStart()); + EXPECT_EQ(start, parser.getLocation().getStart()); } if (end != InvalidSourceOffset) { - EXPECT_EQ(end, reader.getLocation().getEnd()); + EXPECT_EQ(end, parser.getLocation().getEnd()); } } @@ -179,9 +230,9 @@ TEST(OsmlStreamParser, empty) const char *testString = ""; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertEnd(parser, 0, 0); } TEST(OsmlStreamParser, oneCharacter) @@ -189,45 +240,102 @@ TEST(OsmlStreamParser, oneCharacter) const char *testString = "a"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); + + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::COLLAPSE); + assertEnd(parser, 1, 1); +} + +TEST(OsmlStreamParser, whitespacePreserve) +{ + const char *testString = " hello \t world "; + // 0123456 78901234 + // 0 1 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + assertTextData(parser, " hello \t world ", 0, 15, 0, 15, + WhitespaceMode::PRESERVE); + assertEnd(parser, 15, 15); +} + +TEST(OsmlStreamParser, whitespaceTrim) +{ + const char *testString = " hello \t world "; + // 0123456 78901234 + // 0 1 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); - assertData(reader, "a", 0, 1); + assertTextData(parser, "hello \t world", 0, 15, 1, 14, + WhitespaceMode::TRIM); + assertEnd(parser, 15, 15); } -TEST(OsmlStreamParser, whitespaceElimination) +TEST(OsmlStreamParser, whitespaceCollapse) { const char *testString = " hello \t world "; // 0123456 78901234 // 0 1 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertData(reader, "hello world", 1, 14); + assertTextData(parser, "hello world", 0, 15, 1, 14, + WhitespaceMode::COLLAPSE); + assertEnd(parser, 15, 15); } -TEST(OsmlStreamParser, whitespaceEliminationWithLinebreak) +TEST(OsmlStreamParser, whitespaceCollapseLinebreak) { const char *testString = " hello \n world "; // 0123456 78901234 // 0 1 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); + + assertTextData(parser, "hello world", 0, 15, 1, 14, + WhitespaceMode::COLLAPSE); + assertEnd(parser, 15, 15); +} + +TEST(OsmlStreamParser, whitespaceCollapseProtected) +{ + const char *testString = " hello\\ \\ world "; + // 012345 67 89012345 + // 0 1 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + assertTextData(parser, "hello world", 0, 16, 1, 15, + WhitespaceMode::COLLAPSE); + assertEnd(parser, 16, 16); +} + +TEST(OsmlStreamParser, whitespaceCollapseProtected2) +{ + const char *testString = " hello \\ \\ world "; + // 012345 67 89012345 + // 0 1 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); - assertData(reader, "hello world", 1, 14); + assertTextData(parser, "hello world", 0, 17, 1, 16, + WhitespaceMode::COLLAPSE); + assertEnd(parser, 17, 17); } static void testEscapeSpecialCharacter(const std::string &c) { CharReader charReader(std::string("\\") + c); - OsmlStreamParser reader(charReader, logger); - EXPECT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - EXPECT_EQ(c, reader.getText().asString()); - - SourceLocation loc = reader.getText().getLocation(); - EXPECT_EQ(0U, loc.getStart()); - EXPECT_EQ(1U + c.size(), loc.getEnd()); + OsmlStreamParser parser(charReader, logger); + assertTextData(parser, c, 0, 2, 0, 2, WhitespaceMode::PRESERVE); + assertEnd(parser, 2, 2); } TEST(OsmlStreamParser, escapeSpecialCharacters) @@ -240,9 +348,11 @@ TEST(OsmlStreamParser, escapeSpecialCharacters) TEST(OsmlStreamParser, simpleSingleLineComment) { const char *testString = "% This is a single line comment"; + // 0123456789012345678901234567890 + // 0 1 2 3 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + OsmlStreamParser parser(charReader, logger); + assertEnd(parser, 31, 31); } TEST(OsmlStreamParser, singleLineComment) @@ -251,24 +361,11 @@ TEST(OsmlStreamParser, singleLineComment) // 01234567890123456789012345678901 23 // 0 1 2 3 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(1U, loc.getEnd()); - } - - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(33U, loc.getStart()); - ASSERT_EQ(34U, loc.getEnd()); - } + OsmlStreamParser parser(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); + assertTextData(parser, "b", 33, 34, 33, 34, WhitespaceMode::PRESERVE); + assertEnd(parser, 34, 34); } TEST(OsmlStreamParser, multilineComment) @@ -277,24 +374,27 @@ TEST(OsmlStreamParser, multilineComment) // 0123456789012 3 456789012345678901234567890 // 0 1 2 3 4 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(1U, loc.getEnd()); - } + OsmlStreamParser parser(charReader, logger); - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(40U, loc.getStart()); - ASSERT_EQ(41U, loc.getEnd()); - } + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); + assertTextData(parser, "b", 40, 41, 40, 41, WhitespaceMode::PRESERVE); + assertEnd(parser, 41, 41); +} - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); +TEST(OsmlStreamParser, unfinishedMultilineComment) +{ + const char *testString = "a%{ This is a\n\n multiline line comment"; + // 0123456789012 3 456789012345678901234567 + // 0 1 2 3 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + logger.reset(); + + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 38, 38); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, nestedMultilineComment) @@ -303,24 +403,11 @@ TEST(OsmlStreamParser, nestedMultilineComment) // 0123456789012 3 456789012345678901234567890 // 0 1 2 3 4 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("a", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(1U, loc.getEnd()); - } - - { - ASSERT_EQ(OsmlStreamParser::State::DATA, reader.parse()); - ASSERT_EQ("b", reader.getText().asString()); - SourceLocation loc = reader.getText().getLocation(); - ASSERT_EQ(40U, loc.getStart()); - ASSERT_EQ(41U, loc.getEnd()); - } + OsmlStreamParser parser(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); + assertTextData(parser, "b", 40, 41, 40, 41, WhitespaceMode::PRESERVE); + assertEnd(parser, 41, 41); } TEST(OsmlStreamParser, simpleCommand) @@ -328,45 +415,27 @@ TEST(OsmlStreamParser, simpleCommand) const char *testString = "\\test"; // 0 12345 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::COMMAND, reader.parse()); + OsmlStreamParser parser(charReader, logger); - Variant commandName = reader.getCommandName(); - ASSERT_EQ("test", commandName.asString()); - - SourceLocation loc = commandName.getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); - - ASSERT_EQ(0U, reader.getCommandArguments().asMap().size()); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertCommand(parser, "test", 0, 5); + assertEnd(parser); } TEST(OsmlStreamParser, simpleCommandWithName) { - const char *testString = "\\test#bla"; - // 0 12345678 + const char *testString = "\\test#foo"; + // 012345678 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::COMMAND, reader.parse()); - - Variant commandName = reader.getCommandName(); - ASSERT_EQ("test", commandName.asString()); - SourceLocation loc = commandName.getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); + OsmlStreamParser parser(charReader, logger); - Variant commandArguments = reader.getCommandArguments(); - ASSERT_TRUE(commandArguments.isMap()); - ASSERT_EQ(1U, commandArguments.asMap().size()); - ASSERT_EQ(1U, commandArguments.asMap().count("name")); - ASSERT_EQ("bla", commandArguments.asMap()["name"].asString()); + assertCommandStart(parser, "test", false, Variant::mapType{{"name", "foo"}}, + 0, 5); - loc = commandArguments.asMap()["name"].getLocation(); - ASSERT_EQ(5U, loc.getStart()); - ASSERT_EQ(9U, loc.getEnd()); + Variant::mapType args = parser.getCommandArguments().asMap(); + ASSERT_EQ(5U, args["name"].getLocation().getStart()); + ASSERT_EQ(9U, args["name"].getLocation().getEnd()); - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertEnd(parser); } TEST(OsmlStreamParser, simpleCommandWithArguments) @@ -375,38 +444,21 @@ TEST(OsmlStreamParser, simpleCommandWithArguments) // 0 123456789012345 678901 2 // 0 1 2 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::COMMAND, reader.parse()); + OsmlStreamParser parser(charReader, logger); - Variant commandName = reader.getCommandName(); - ASSERT_EQ("test", commandName.asString()); - SourceLocation loc = commandName.getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); + assertCommandStart(parser, "test", false, + Variant::mapType{{"a", 1}, {"b", 2}, {"c", "test"}}, 0, + 5); - Variant commandArguments = reader.getCommandArguments(); - ASSERT_TRUE(commandArguments.isMap()); - ASSERT_EQ(3U, commandArguments.asMap().size()); - ASSERT_EQ(1U, commandArguments.asMap().count("a")); - ASSERT_EQ(1U, commandArguments.asMap().count("b")); - ASSERT_EQ(1U, commandArguments.asMap().count("c")); - ASSERT_EQ(1, commandArguments.asMap()["a"].asInt()); - ASSERT_EQ(2, commandArguments.asMap()["b"].asInt()); - ASSERT_EQ("test", commandArguments.asMap()["c"].asString()); + Variant::mapType args = parser.getCommandArguments().asMap(); + ASSERT_EQ(8U, args["a"].getLocation().getStart()); + ASSERT_EQ(9U, args["a"].getLocation().getEnd()); + ASSERT_EQ(12U, args["b"].getLocation().getStart()); + ASSERT_EQ(13U, args["b"].getLocation().getEnd()); + ASSERT_EQ(16U, args["c"].getLocation().getStart()); + ASSERT_EQ(22U, args["c"].getLocation().getEnd()); - loc = commandArguments.asMap()["a"].getLocation(); - ASSERT_EQ(8U, loc.getStart()); - ASSERT_EQ(9U, loc.getEnd()); - - loc = commandArguments.asMap()["b"].getLocation(); - ASSERT_EQ(12U, loc.getStart()); - ASSERT_EQ(13U, loc.getEnd()); - - loc = commandArguments.asMap()["c"].getLocation(); - ASSERT_EQ(16U, loc.getStart()); - ASSERT_EQ(22U, loc.getEnd()); - - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + assertEnd(parser); } TEST(OsmlStreamParser, simpleCommandWithArgumentsAndName) @@ -415,44 +467,24 @@ TEST(OsmlStreamParser, simpleCommandWithArgumentsAndName) // 0 1234567890123456789 01234 56 // 0 1 2 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - ASSERT_EQ(OsmlStreamParser::State::COMMAND, reader.parse()); - - Variant commandName = reader.getCommandName(); - ASSERT_EQ("test", commandName.asString()); - SourceLocation loc = commandName.getLocation(); - ASSERT_EQ(0U, loc.getStart()); - ASSERT_EQ(5U, loc.getEnd()); - - Variant commandArguments = reader.getCommandArguments(); - ASSERT_TRUE(commandArguments.isMap()); - ASSERT_EQ(4U, commandArguments.asMap().size()); - ASSERT_EQ(1U, commandArguments.asMap().count("a")); - ASSERT_EQ(1U, commandArguments.asMap().count("b")); - ASSERT_EQ(1U, commandArguments.asMap().count("c")); - ASSERT_EQ(1U, commandArguments.asMap().count("name")); - ASSERT_EQ(1, commandArguments.asMap()["a"].asInt()); - ASSERT_EQ(2, commandArguments.asMap()["b"].asInt()); - ASSERT_EQ("test", commandArguments.asMap()["c"].asString()); - ASSERT_EQ("bla", commandArguments.asMap()["name"].asString()); - - loc = commandArguments.asMap()["a"].getLocation(); - ASSERT_EQ(12U, loc.getStart()); - ASSERT_EQ(13U, loc.getEnd()); - - loc = commandArguments.asMap()["b"].getLocation(); - ASSERT_EQ(16U, loc.getStart()); - ASSERT_EQ(17U, loc.getEnd()); - - loc = commandArguments.asMap()["c"].getLocation(); - ASSERT_EQ(20U, loc.getStart()); - ASSERT_EQ(26U, loc.getEnd()); - - loc = commandArguments.asMap()["name"].getLocation(); - ASSERT_EQ(5U, loc.getStart()); - ASSERT_EQ(9U, loc.getEnd()); - - ASSERT_EQ(OsmlStreamParser::State::END, reader.parse()); + OsmlStreamParser parser(charReader, logger); + + assertCommandStart( + parser, "test", false, + Variant::mapType{{"name", "bla"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 0, + 5); + + Variant::mapType args = parser.getCommandArguments().asMap(); + ASSERT_EQ(5U, args["name"].getLocation().getStart()); + ASSERT_EQ(9U, args["name"].getLocation().getEnd()); + ASSERT_EQ(12U, args["a"].getLocation().getStart()); + ASSERT_EQ(13U, args["a"].getLocation().getEnd()); + ASSERT_EQ(16U, args["b"].getLocation().getStart()); + ASSERT_EQ(17U, args["b"].getLocation().getEnd()); + ASSERT_EQ(20U, args["c"].getLocation().getStart()); + ASSERT_EQ(26U, args["c"].getLocation().getEnd()); + + assertEnd(parser); } TEST(OsmlStreamParser, fields) @@ -461,21 +493,21 @@ TEST(OsmlStreamParser, fields) // 01234567890123 // 0 1 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - assertData(reader, "a", 6, 7); - assertFieldEnd(reader, 7, 8); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertTextData(parser, "a", 6, 7, 6, 7, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 7, 8); - assertFieldStart(reader, false, 8, 9); - assertData(reader, "b", 9, 10); - assertFieldEnd(reader, 10, 11); + assertFieldStart(parser, false, 8, 9); + assertTextData(parser, "b", 9, 10, 9, 10, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 10, 11); - assertFieldStart(reader, false, 11, 12); - assertData(reader, "c", 12, 13); - assertFieldEnd(reader, 13, 14); - assertEnd(reader, 14, 14); + assertFieldStart(parser, false, 11, 12); + assertTextData(parser, "c", 12, 13, 12, 13, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 13, 14); + assertEnd(parser, 14, 14); } TEST(OsmlStreamParser, dataOutsideField) @@ -484,785 +516,781 @@ TEST(OsmlStreamParser, dataOutsideField) // 0123456789012 // 0 1 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - assertData(reader, "a", 6, 7); - assertFieldEnd(reader, 7, 8); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertTextData(parser, "a", 6, 7, 6, 7, WhitespaceMode::COLLAPSE); + assertFieldEnd(parser, 7, 8); - assertFieldStart(reader, false, 8, 9); - assertData(reader, "b", 9, 10); - assertFieldEnd(reader, 10, 11); + assertFieldStart(parser, false, 8, 9); + assertTextData(parser, "b", 9, 10, 9, 10, WhitespaceMode::COLLAPSE); + assertFieldEnd(parser, 10, 11); - assertData(reader, "c", 12, 13); - assertEnd(reader, 13, 13); + assertTextData(parser, "c", 11, 13, 12, 13, WhitespaceMode::COLLAPSE); + assertEnd(parser, 13, 13); } TEST(OsmlStreamParser, nestedCommand) { - const char *testString = "\\test{a}{\\test2{b} c} d"; - // 012345678 90123456789012 - // 0 1 2 - CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + const char *testString = "\\test{a}{\\test2{b} c} d"; + // 012345678 90123456789012 + // 0 1 2 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test", 0, 5); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertData(parser, "a", 6, 7); + assertFieldEnd(parser, 7, 8); - assertFieldStart(reader, false, 5, 6); - assertData(reader, "a", 6, 7); - assertFieldEnd(reader, 7, 8); - - assertFieldStart(reader, false, 8, 9); - { - assertCommand(reader, "test2", 9, 15); - assertFieldStart(reader, false, 15, 16); - assertData(reader, "b", 16, 17); - assertFieldEnd(reader, 17, 18); - } - assertData(reader, "c", 19, 20); - assertFieldEnd(reader, 20, 21); - assertData(reader, "d", 22, 23); - assertEnd(reader, 23, 23); + assertFieldStart(parser, false, 8, 9); + assertCommand(parser, "test2", 9, 15); + assertFieldStart(parser, false, 15, 16); + assertData(parser, "b", 16, 17); + assertFieldEnd(parser, 17, 18); + assertData(parser, "c", 19, 20); + assertFieldEnd(parser, 20, 21); + assertData(parser, "d", 22, 23); + assertEnd(parser, 23, 23); } + TEST(OsmlStreamParser, nestedCommandImmediateEnd) { - const char *testString = "\\test{\\test2{b}} d"; - // 012345 678901234567 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - { - assertCommand(reader, "test2", 6, 12); - assertFieldStart(reader, false, 12, 13); - assertData(reader, "b", 13, 14); - assertFieldEnd(reader, 14, 15); - } - assertFieldEnd(reader, 15, 16); - assertData(reader, "d", 17, 18); - assertEnd(reader, 18, 18); + const char *testString = "\\test{\\test2{b}} d"; + // 012345 678901234567 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + { + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertData(parser, "b", 13, 14); + assertFieldEnd(parser, 14, 15); + } + assertFieldEnd(parser, 15, 16); + assertData(parser, "d", 17, 18); + assertEnd(parser, 18, 18); } TEST(OsmlStreamParser, nestedCommandNoData) { - const char *testString = "\\test{\\test2}"; - // 012345 6789012 - CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + const char *testString = "\\test{\\test2}"; + // 012345 6789012 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - assertCommand(reader, "test2", 6, 12); - assertFieldEnd(reader, 12, 13); - assertEnd(reader, 13, 13); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldEnd(parser, 12, 13); + assertEnd(parser, 13, 13); } TEST(OsmlStreamParser, multipleCommands) { - const char *testString = "\\a \\b \\c \\d"; - // 012 345 678 90 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + const char *testString = "\\a \\b \\c \\d"; + // 012 345 678 90 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "a", 0, 2); - assertData(reader, " ", 2, 3, WhitespaceMode::PRESERVE); - assertCommand(reader, "b", 3, 5); - assertData(reader, " ", 5, 6, WhitespaceMode::PRESERVE); - assertCommand(reader, "c", 6, 8); - assertData(reader, " ", 8, 9, WhitespaceMode::PRESERVE); - assertCommand(reader, "d", 9, 11); - assertEnd(reader, 11, 11); + assertCommand(parser, "a", 0, 2); + assertEmptyData(parser); + assertCommand(parser, "b", 3, 5); + assertEmptyData(parser); + assertCommand(parser, "c", 6, 8); + assertEmptyData(parser); + assertCommand(parser, "d", 9, 11); + assertEnd(parser, 11, 11); } TEST(OsmlStreamParser, fieldsWithSpaces) { - const char *testString = "\\a {\\b \\c} \n\n {\\d}"; - // 0123 456 789012 3 456 789 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - - assertCommand(reader, "a", 0, 2); - assertData(reader, " ", 2, 3, WhitespaceMode::PRESERVE); - assertFieldStart(reader, false, 3, 4); - assertCommand(reader, "b", 4, 6); - assertData(reader, " ", 6, 7, WhitespaceMode::PRESERVE); - assertCommand(reader, "c", 7, 9); - assertFieldEnd(reader, 9, 10); - assertData(reader, " \n\n {", 10, 12, WhitespaceMode::PRESERVE); - assertFieldStart(reader, false, 16, 17); - assertCommand(reader, "d", 17, 19); - assertFieldEnd(reader, 19, 20); - assertEnd(reader, 20, 20); -} - -TEST(OsmlStreamParser, errorNoFieldToStart) -{ - const char *testString = "\\a b {"; - // 012345 - // 0 - CharReader charReader(testString); + const char *testString = "\\a {\\b \\c} \n\n {\\d}"; + // 0123 456 789012 3 456 789 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - OsmlStreamParser reader(charReader, logger); + assertCommand(parser, "a", 0, 2); + assertEmptyData(parser); + assertFieldStart(parser, false, 3, 4); + assertCommand(parser, "b", 4, 6); + assertEmptyData(parser); + assertCommand(parser, "c", 7, 9); + assertFieldEnd(parser, 9, 10); + assertEmptyData(parser); + assertFieldStart(parser, false, 16, 17); + assertCommand(parser, "d", 17, 19); + assertFieldEnd(parser, 19, 20); + assertEnd(parser, 20, 20); +} - logger.reset(); - assertCommand(reader, "a", 0, 2); - assertData(reader, "b", 3, 4); - ASSERT_FALSE(logger.hasError()); - assertEnd(reader, 6, 6); - ASSERT_TRUE(logger.hasError()); +TEST(OsmlStreamParser, errorEndButOpenField) +{ + const char *testString = "\\a b {"; + // 012345 + // 0 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + logger.reset(); + assertCommand(parser, "a", 0, 2); + assertData(parser, "b", 3, 4); + assertFieldStart(parser, false, 5, 6); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 6, 6); + ASSERT_TRUE(logger.hasError()); } + TEST(OsmlStreamParser, errorNoFieldToEnd) { - const char *testString = "\\a b }"; - // 012345 - // 0 - CharReader charReader(testString); + const char *testString = "\\a b }"; + // 012345 + // 0 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(reader, "a", 0, 2); - assertData(reader, "b", 3, 4); - ASSERT_FALSE(logger.hasError()); - assertEnd(reader, 6, 6); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "a", 0, 2); + assertData(parser, "b", 3, 4); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 6, 6); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorNoFieldEndNested) { - const char *testString = "\\test{\\test2{}}}"; - // 012345 6789012345 - // 0 1 - CharReader charReader(testString); + const char *testString = "\\test{\\test2{}}}"; + // 012345 6789012345 + // 0 1 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - assertCommand(reader, "test2", 6, 12); - assertFieldStart(reader, false, 12, 13); - assertFieldEnd(reader, 13, 14); - assertFieldEnd(reader, 14, 15); - ASSERT_FALSE(logger.hasError()); - assertEnd(reader, 16, 16); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertFieldEnd(parser, 13, 14); + assertFieldEnd(parser, 14, 15); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 16, 16); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorNoFieldEndNestedData) { - const char *testString = "\\test{\\test2{}}a}"; - // 012345 67890123456 - // 0 1 - CharReader charReader(testString); + const char *testString = "\\test{\\test2{}}a}"; + // 012345 67890123456 + // 0 1 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(reader, "test", 0, 5); - assertFieldStart(reader, false, 5, 6); - assertCommand(reader, "test2", 6, 12); - assertFieldStart(reader, false, 12, 13); - assertFieldEnd(reader, 13, 14); - assertFieldEnd(reader, 14, 15); - assertData(reader, "a", 15, 16); - ASSERT_FALSE(logger.hasError()); - assertEnd(reader, 17, 17); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertFieldEnd(parser, 13, 14); + assertFieldEnd(parser, 14, 15); + assertData(parser, "a", 15, 16); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 17, 17); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, beginEnd) { - const char *testString = "\\begin{book}\\end{book}"; - // 012345678901 2345678901 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}\\end{book}"; + // 012345678901 2345678901 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "book", 7, 11); - assertFieldStart(reader, true, 12, 13); - assertFieldEnd(reader, 17, 21); - assertEnd(reader, 22, 22); + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertCommandEnd(parser, 17, 21); + assertEnd(parser, 22, 22); } TEST(OsmlStreamParser, beginEndWithName) { - const char *testString = "\\begin{book#a}\\end{book}"; - // 01234567890123 4567890123 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book#a}\\end{book}"; + // 01234567890123 4567890123 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "book", {{"name", "a"}}, 7, 11); - assertFieldStart(reader, true, 14, 15); - assertFieldEnd(reader, 19, 23); - assertEnd(reader, 24, 24); + assertCommandStart(parser, "book", true, {{"name", "a"}}, 7, 11); + assertCommandEnd(parser, 19, 23); + assertEnd(parser, 24, 24); } TEST(OsmlStreamParser, beginEndWithNameAndArgs) { - const char *testString = "\\begin{book#a}[a=1,b=2,c=\"test\"]\\end{book}"; - // 0123456789012345678901234 56789 01 2345678901 - // 0 1 2 3 4 - CharReader charReader(testString); + const char *testString = "\\begin{book#a}[a=1,b=2,c=\"test\"]\\end{book}"; + // 0123456789012345678901234 56789 01 2345678901 + // 0 1 2 3 4 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "book", - {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); - assertFieldStart(reader, true, 32, 33); - assertFieldEnd(reader, 37, 41); - assertEnd(reader, 42, 42); + assertCommandStart(parser, "book", true, + {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); + assertCommandEnd(parser, 37, 41); + assertEnd(parser, 42, 42); } TEST(OsmlStreamParser, beginEndWithNameAndArgsMultipleFields) { - const char *testString = - "\\begin{book#a}[a=1,b=2,c=\"test\"]{a \\test}{b \\test{}}\\end{book}"; - // 0123456789012345678901234 56789 01234 567890123 45678901 2345678901 - // 0 1 2 3 4 5 6 - CharReader charReader(testString); - - OsmlStreamParser reader(charReader, logger); - - assertCommand(reader, "book", - {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); - assertFieldStart(reader, false, 32, 33); - assertData(reader, "a", 33, 34); - assertCommand(reader, "test", Variant::mapType{}, 35, 40); - assertFieldEnd(reader, 40, 41); - assertFieldStart(reader, false, 41, 42); - assertData(reader, "b", 42, 43); - assertCommand(reader, "test", Variant::mapType{}, 44, 49); - assertFieldStart(reader, false, 49, 50); - assertFieldEnd(reader, 50, 51); - assertFieldEnd(reader, 51, 52); - assertFieldStart(reader, true, 52, 53); - assertFieldEnd(reader, 57, 61); - assertEnd(reader, 62, 62); + const char *testString = + "\\begin{book#a}[a=1,b=2,c=\"test\"]{a \\test}{b \\test{}}\\end{book}"; + // 0123456789012345678901234 56789 01234 567890123 45678901 2345678901 + // 0 1 2 3 4 5 6 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + assertCommandStart(parser, "book", true, + {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); + assertFieldStart(parser, false, 32, 33); + assertData(parser, "a", 33, 34); + assertCommand(parser, "test", 35, 40); + assertFieldEnd(parser, 40, 41); + assertFieldStart(parser, false, 41, 42); + assertData(parser, "b", 42, 43); + assertCommand(parser, "test", 44, 49); + assertFieldStart(parser, false, 49, 50); + assertFieldEnd(parser, 50, 51); + assertFieldEnd(parser, 51, 52); + assertCommandEnd(parser, 57, 61); + assertEnd(parser, 62, 62); } TEST(OsmlStreamParser, beginEndWithData) { - const char *testString = "\\begin{book}a\\end{book}"; - // 0123456789012 3456789012 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}a\\end{book}"; + // 0123456789012 3456789012 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "book", 7, 11); - assertFieldStart(reader, true, 12, 13); - assertData(reader, "a", 12, 13); - assertFieldEnd(reader, 18, 22); - assertEnd(reader, 23, 23); + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertCommandEnd(parser, 18, 22); + assertEnd(parser, 23, 23); } - +/* TEST(OsmlStreamParser, beginEndNested) { - const char *testString = - "\\begin{a}{b} c \\begin{d}{e}{f} \\g{h} \\end{d}\\end{a}"; - // 012345678901234 5678901234567890 123456 7890123 4567890 - // 0 1 2 3 4 5 - CharReader charReader(testString); - - OsmlStreamParser reader(charReader, logger); - - assertCommand(reader, "a", 7, 8); - assertFieldStart(reader, false, 9, 10); - assertData(reader, "b", 10, 11); - assertFieldEnd(reader, 11, 12); - assertFieldStart(reader, true, 13, 14); - assertData(reader, "c", 13, 14); - assertCommand(reader, "d", 22, 23); - assertFieldStart(reader, false, 24, 25); - assertData(reader, "e", 25, 26); - assertFieldEnd(reader, 26, 27); - assertFieldStart(reader, false, 27, 28); - assertData(reader, "f", 28, 29); - assertFieldEnd(reader, 29, 30); - assertFieldStart(reader, true, 31, 32); - assertCommand(reader, "g", 31, 33); - assertFieldStart(reader, false, 33, 34); - assertData(reader, "h", 34, 35); - assertFieldEnd(reader, 35, 36); - assertFieldEnd(reader, 42, 43); - assertFieldEnd(reader, 49, 50); - assertEnd(reader, 51, 51); + const char *testString = + "\\begin{a}{b} c \\begin{d}{e}{f} \\g{h} \\end{d}\\end{a}"; + // 012345678901234 5678901234567890 123456 7890123 4567890 + // 0 1 2 3 4 5 + CharReader charReader(testString); + + OsmlStreamParser reader(charReader, logger); + + assertCommand(reader, "a", 7, 8); + assertFieldStart(reader, false, 9, 10); + assertData(reader, "b", 10, 11); + assertFieldEnd(reader, 11, 12); + assertFieldStart(reader, true, 13, 14); + assertData(reader, "c", 13, 14); + assertCommand(reader, "d", 22, 23); + assertFieldStart(reader, false, 24, 25); + assertData(reader, "e", 25, 26); + assertFieldEnd(reader, 26, 27); + assertFieldStart(reader, false, 27, 28); + assertData(reader, "f", 28, 29); + assertFieldEnd(reader, 29, 30); + assertFieldStart(reader, true, 31, 32); + assertCommand(reader, "g", 31, 33); + assertFieldStart(reader, false, 33, 34); + assertData(reader, "h", 34, 35); + assertFieldEnd(reader, 35, 36); + assertFieldEnd(reader, 42, 43); + assertFieldEnd(reader, 49, 50); + assertEnd(reader, 51, 51); } TEST(OsmlStreamParser, beginEndWithCommand) { - const char *testString = "\\begin{book}\\a{test}\\end{book}"; - // 012345678901 23456789 0123456789 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}\\a{test}\\end{book}"; + // 012345678901 23456789 0123456789 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertCommand(reader, "book", 7, 11); - assertFieldStart(reader, true, 12, 13); - assertCommand(reader, "a", 12, 14); - assertFieldStart(reader, false, 14, 15); - assertData(reader, "test", 15, 19); - assertFieldEnd(reader, 19, 20); - assertFieldEnd(reader, 25, 29); - assertEnd(reader, 30, 30); + assertCommand(reader, "book", 7, 11); + assertFieldStart(reader, true, 12, 13); + assertCommand(reader, "a", 12, 14); + assertFieldStart(reader, false, 14, 15); + assertData(reader, "test", 15, 19); + assertFieldEnd(reader, 19, 20); + assertFieldEnd(reader, 25, 29); + assertEnd(reader, 30, 30); } TEST(OsmlStreamParser, errorBeginNoBraceOpen) { - const char *testString = "\\begin a"; - // 01234567 - CharReader charReader(testString); + const char *testString = "\\begin a"; + // 01234567 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertData(reader, "a", 7, 8); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertData(reader, "a", 7, 8); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoIdentifier) { - const char *testString = "\\begin{!"; - CharReader charReader(testString); + const char *testString = "\\begin{!"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoBraceClose) { - const char *testString = "\\begin{a"; - CharReader charReader(testString); + const char *testString = "\\begin{a"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoName) { - const char *testString = "\\begin{a#}"; - CharReader charReader(testString); + const char *testString = "\\begin{a#}"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "a"); - ASSERT_TRUE(logger.hasError()); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertEnd(reader); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(reader, "a"); + ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertEnd(reader); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBraceOpen) { - const char *testString = "\\end a"; - // 012345 - CharReader charReader(testString); + const char *testString = "\\end a"; + // 012345 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertData(reader, "a", 5, 6); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertData(reader, "a", 5, 6); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoIdentifier) { - const char *testString = "\\end{!"; - CharReader charReader(testString); + const char *testString = "\\end{!"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBraceClose) { - const char *testString = "\\end{a"; - CharReader charReader(testString); + const char *testString = "\\end{a"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBegin) { - const char *testString = "\\end{a}"; - CharReader charReader(testString); + const char *testString = "\\end{a}"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginEndMismatch) { - const char *testString = "\\begin{a} \\begin{b} test \\end{a}"; - // 0123456789 012345678901234 5678901 - // 0 1 2 3 - CharReader charReader(testString); + const char *testString = "\\begin{a} \\begin{b} test \\end{a}"; + // 0123456789 012345678901234 5678901 + // 0 1 2 3 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - assertCommand(reader, "a", 7, 8); - assertFieldStart(reader, true, 10, 11); - assertCommand(reader, "b", 17, 18); - assertFieldStart(reader, true, 20, 24); - assertData(reader, "test", 20, 24); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(reader.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(reader, "a", 7, 8); + assertFieldStart(reader, true, 10, 11); + assertCommand(reader, "b", 17, 18); + assertFieldStart(reader, true, 20, 24); + assertData(reader, "test", 20, 24); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, commandWithNSSep) { - const char *testString = "\\test1:test2"; - // 012345678901 - CharReader charReader(testString); + const char *testString = "\\test1:test2"; + // 012345678901 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertCommand(reader, "test1:test2", 0, 12); - assertEnd(reader, 12, 12); + assertCommand(reader, "test1:test2", 0, 12); + assertEnd(reader, 12, 12); } TEST(OsmlStreamParser, beginEndWithNSSep) { - const char *testString = "\\begin{test1:test2}\\end{test1:test2}"; - // 0123456789012345678 90123456789012345 - // 0 1 2 3 - CharReader charReader(testString); + const char *testString = "\\begin{test1:test2}\\end{test1:test2}"; + // 0123456789012345678 90123456789012345 + // 0 1 2 3 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertCommand(reader, "test1:test2", 7, 18); - assertFieldStart(reader, true, 19, 20); - assertFieldEnd(reader, 24, 35); - assertEnd(reader, 36, 36); + assertCommand(reader, "test1:test2", 7, 18); + assertFieldStart(reader, true, 19, 20); + assertFieldEnd(reader, 24, 35); + assertEnd(reader, 36, 36); } TEST(OsmlStreamParser, errorBeginNSSep) { - const char *testString = "\\begin:test{blub}\\end{blub}"; - CharReader charReader(testString); + const char *testString = "\\begin:test{blub}\\end{blub}"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "blub"); - ASSERT_TRUE(logger.hasError()); - assertFieldStart(reader, true); - assertFieldEnd(reader); - assertEnd(reader); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(reader, "blub"); + ASSERT_TRUE(logger.hasError()); + assertFieldStart(reader, true); + assertFieldEnd(reader); + assertEnd(reader); } TEST(OsmlStreamParser, errorEndNSSep) { - const char *testString = "\\begin{blub}\\end:test{blub}"; - CharReader charReader(testString); + const char *testString = "\\begin{blub}\\end:test{blub}"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - assertCommand(reader, "blub"); - assertFieldStart(reader, true); - ASSERT_FALSE(logger.hasError()); - assertFieldEnd(reader); - ASSERT_TRUE(logger.hasError()); - assertEnd(reader); + logger.reset(); + assertCommand(reader, "blub"); + assertFieldStart(reader, true); + ASSERT_FALSE(logger.hasError()); + assertFieldEnd(reader); + ASSERT_TRUE(logger.hasError()); + assertEnd(reader); } TEST(OsmlStreamParser, errorEmptyNs) { - const char *testString = "\\test:"; - CharReader charReader(testString); + const char *testString = "\\test:"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "test"); - ASSERT_TRUE(logger.hasError()); - assertData(reader, ":"); - assertEnd(reader); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(reader, "test"); + ASSERT_TRUE(logger.hasError()); + assertData(reader, ":"); + assertEnd(reader); } TEST(OsmlStreamParser, errorRepeatedNs) { - const char *testString = "\\test::"; - CharReader charReader(testString); + const char *testString = "\\test::"; + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "test"); - ASSERT_TRUE(logger.hasError()); - assertData(reader, "::"); - assertEnd(reader); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(reader, "test"); + ASSERT_TRUE(logger.hasError()); + assertData(reader, "::"); + assertEnd(reader); } TEST(OsmlStreamParser, explicitDefaultField) { - const char *testString = "\\a{!b}c"; - // 01234567 - CharReader charReader(testString); + const char *testString = "\\a{!b}c"; + // 01234567 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertData(reader, "b", 4, 5); - assertFieldEnd(reader, 5, 6); - assertData(reader, "c", 6, 7); - assertEnd(reader, 7, 7); + assertCommand(reader, "a", 0, 2); + assertFieldStart(reader, true, 2, 4); + assertData(reader, "b", 4, 5); + assertFieldEnd(reader, 5, 6); + assertData(reader, "c", 6, 7); + assertEnd(reader, 7, 7); } TEST(OsmlStreamParser, explicitDefaultFieldWithCommand) { - const char *testString = "\\a{!\\b}c"; - // 0123 4567 - CharReader charReader(testString); + const char *testString = "\\a{!\\b}c"; + // 0123 4567 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertCommand(reader, "b", 4, 6); - assertFieldEnd(reader, 6, 7); - assertData(reader, "c", 7, 8); - assertEnd(reader, 8, 8); + assertCommand(reader, "a", 0, 2); + assertFieldStart(reader, true, 2, 4); + assertCommand(reader, "b", 4, 6); + assertFieldEnd(reader, 6, 7); + assertData(reader, "c", 7, 8); + assertEnd(reader, 8, 8); } TEST(OsmlStreamParser, errorFieldAfterExplicitDefaultField) { - const char *testString = "\\a{!\\b}{c}"; - // 0123 456789 - CharReader charReader(testString); + const char *testString = "\\a{!\\b}{c}"; + // 0123 456789 + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertCommand(reader, "b", 4, 6); - assertFieldEnd(reader, 6, 7); - ASSERT_FALSE(logger.hasError()); - assertData(reader, "c", 8, 9); - ASSERT_TRUE(logger.hasError()); - assertEnd(reader, 10, 10); + logger.reset(); + assertCommand(reader, "a", 0, 2); + assertFieldStart(reader, true, 2, 4); + assertCommand(reader, "b", 4, 6); + assertFieldEnd(reader, 6, 7); + ASSERT_FALSE(logger.hasError()); + assertData(reader, "c", 8, 9); + ASSERT_TRUE(logger.hasError()); + assertEnd(reader, 10, 10); } TEST(OsmlStreamParser, annotationStart) { - const char *testString = "<\\a"; - // 0 12 + const char *testString = "<\\a"; + // 0 12 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); - assertEnd(reader, 3, 3); + assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); + assertEnd(reader, 3, 3); } TEST(OsmlStreamParser, annotationStartWithName) { - const char *testString = "<\\annotationWithName#aName"; - // 0 1234567890123456789012345 - // 0 1 2 + const char *testString = "<\\annotationWithName#aName"; + // 0 1234567890123456789012345 + // 0 1 2 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationStart(reader, "annotationWithName", - Variant::mapType{{"name", "aName"}}, 0, 20); - assertEnd(reader, 26, 26); + assertAnnotationStart(reader, "annotationWithName", + Variant::mapType{{"name", "aName"}}, 0, 20); + assertEnd(reader, 26, 26); } TEST(OsmlStreamParser, annotationStartWithArguments) { - const char *testString = "<\\annotationWithName#aName[a=1,b=2]"; - // 0 1234567890123456789012345678901234 - // 0 1 2 3 + const char *testString = "<\\annotationWithName#aName[a=1,b=2]"; + // 0 1234567890123456789012345678901234 + // 0 1 2 3 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationStart( - reader, "annotationWithName", - Variant::mapType{{"name", "aName"}, {"a", 1}, {"b", 2}}, 0, 20); - assertEnd(reader, 35, 35); + assertAnnotationStart( + reader, "annotationWithName", + Variant::mapType{{"name", "aName"}, {"a", 1}, {"b", 2}}, 0, 20); + assertEnd(reader, 35, 35); } TEST(OsmlStreamParser, simpleAnnotationStartBeginEnd) { - const char *testString = "<\\begin{ab#name}[a=1,b=2] a \\end{ab}\\>"; - // 0 123456789012345678901234567 89012345 67 - // 0 1 2 3 + const char *testString = "<\\begin{ab#name}[a=1,b=2] a \\end{ab}\\>"; + // 0 123456789012345678901234567 89012345 67 + // 0 1 2 3 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationStart( - reader, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, - 10); - assertFieldStart(reader, true, 26, 27); - assertData(reader, "a", 26, 27); - assertFieldEnd(reader, 33, 35); - assertAnnotationEnd(reader, "", "", 36, 38); - assertEnd(reader, 38, 38); + assertAnnotationStart( + reader, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, + 10); + assertFieldStart(reader, true, 26, 27); + assertData(reader, "a", 26, 27); + assertFieldEnd(reader, 33, 35); + assertAnnotationEnd(reader, "", "", 36, 38); + assertEnd(reader, 38, 38); } TEST(OsmlStreamParser, annotationEnd) { - const char *testString = "\\a>"; - // 012 + const char *testString = "\\a>"; + // 012 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationEnd(reader, "a", "", 0, 2); - assertEnd(reader, 3, 3); + assertAnnotationEnd(reader, "a", "", 0, 2); + assertEnd(reader, 3, 3); } TEST(OsmlStreamParser, annotationEndWithName) { - const char *testString = "\\a#name>"; - // 01234567 + const char *testString = "\\a#name>"; + // 01234567 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationEnd(reader, "a", "name", 0, 2); - assertEnd(reader, 8, 8); + assertAnnotationEnd(reader, "a", "name", 0, 2); + assertEnd(reader, 8, 8); } TEST(OsmlStreamParser, annotationEndWithNameAsArgs) { - const char *testString = "\\a[name=name]>"; - // 01234567890123 + const char *testString = "\\a[name=name]>"; + // 01234567890123 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationEnd(reader, "a", "name", 0, 2); - assertEnd(reader, 14, 14); + assertAnnotationEnd(reader, "a", "name", 0, 2); + assertEnd(reader, 14, 14); } TEST(OsmlStreamParser, errorAnnotationEndWithArguments) { - const char *testString = "\\a[foo=bar]>"; - // 012345678901 - // 0 1 + const char *testString = "\\a[foo=bar]>"; + // 012345678901 + // 0 1 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "a", Variant::mapType{{"foo", "bar"}}, 0, 2); - ASSERT_TRUE(logger.hasError()); - assertData(reader, ">", 11, 12); - assertEnd(reader, 12, 12); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(reader, "a", Variant::mapType{{"foo", "bar"}}, 0, 2); + ASSERT_TRUE(logger.hasError()); + assertData(reader, ">", 11, 12); + assertEnd(reader, 12, 12); } TEST(OsmlStreamParser, closingAnnotation) { - const char *testString = "<\\a>"; - // 0 123 + const char *testString = "<\\a>"; + // 0 123 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); - assertData(reader, ">", 3, 4); - assertEnd(reader, 4, 4); + assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); + assertData(reader, ">", 3, 4); + assertEnd(reader, 4, 4); } TEST(OsmlStreamParser, annotationWithFields) { - const char *testString = "a <\\b{c}{d}{!e} f \\> g"; - // 012 345678901234567 8901 - // 0 1 2 + const char *testString = "a <\\b{c}{d}{!e} f \\> g"; + // 012 345678901234567 8901 + // 0 1 2 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - - assertData(reader, "a", 0, 1); - assertAnnotationStart(reader, "b", Variant::mapType{}, 2, 5); - assertFieldStart(reader, false, 5, 6); - assertData(reader, "c", 6, 7); - assertFieldEnd(reader, 7, 8); - assertFieldStart(reader, false, 8, 9); - assertData(reader, "d", 9, 10); - assertFieldEnd(reader, 10, 11); - assertFieldStart(reader, true, 11, 13); - assertData(reader, "e", 13, 14); - assertFieldEnd(reader, 14, 15); - assertData(reader, "f", 16, 17); - assertAnnotationEnd(reader, "", "", 18, 20); - assertData(reader, "g", 21, 22); - assertEnd(reader, 22, 22); + OsmlStreamParser reader(charReader, logger); + + assertData(reader, "a", 0, 1); + assertAnnotationStart(reader, "b", Variant::mapType{}, 2, 5); + assertFieldStart(reader, false, 5, 6); + assertData(reader, "c", 6, 7); + assertFieldEnd(reader, 7, 8); + assertFieldStart(reader, false, 8, 9); + assertData(reader, "d", 9, 10); + assertFieldEnd(reader, 10, 11); + assertFieldStart(reader, true, 11, 13); + assertData(reader, "e", 13, 14); + assertFieldEnd(reader, 14, 15); + assertData(reader, "f", 16, 17); + assertAnnotationEnd(reader, "", "", 18, 20); + assertData(reader, "g", 21, 22); + assertEnd(reader, 22, 22); } TEST(OsmlStreamParser, annotationStartEscape) { - const char *testString = "<\\%test"; - // 0 123456 - // 0 + const char *testString = "<\\%test"; + // 0 123456 + // 0 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser reader(charReader, logger); - assertData(reader, "<%test", 0, 7); - assertEnd(reader, 7, 7); + assertData(reader, "<%test", 0, 7); + assertEnd(reader, 7, 7); } +*/ } -- cgit v1.2.3 From b54760fbd5470032dc716dc870dc08b32dfba5ac Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sat, 28 Feb 2015 15:48:07 +0100 Subject: Test case for data being empty if a token is found --- test/core/parser/utils/TokenizerTest.cpp | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) (limited to 'test') diff --git a/test/core/parser/utils/TokenizerTest.cpp b/test/core/parser/utils/TokenizerTest.cpp index 785bd81..9f644c2 100644 --- a/test/core/parser/utils/TokenizerTest.cpp +++ b/test/core/parser/utils/TokenizerTest.cpp @@ -450,5 +450,32 @@ TEST(Tokenizer, nonPrimaryTokens) TokenizedData data; ASSERT_FALSE(tokenizer.read(reader, token, data)); } + + +TEST(Tokenizer, ambiguousTokens2) +{ + CharReader reader{"<\\"}; + + Tokenizer tokenizer; + + TokenId tBackslash = tokenizer.registerToken("\\"); + TokenId tAnnotationStart = tokenizer.registerToken("<\\"); + + TokenSet tokens = TokenSet{tBackslash, tAnnotationStart}; + Token token; + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ("<\\", token.content); + ASSERT_EQ(tAnnotationStart, token.id); + ASSERT_TRUE(data.empty()); + } + + { + TokenizedData data; + ASSERT_FALSE(tokenizer.read(reader, token, data)); + } +} + } -- cgit v1.2.3 From fa2a5bdf0152002de520fcc72e48686b9e2657b1 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sat, 28 Feb 2015 15:48:38 +0100 Subject: Adapted all unit tests, renamed COMMAND_END event to RANGE_END event to match ranged annotations --- src/formats/osml/OsmlStreamParser.cpp | 28 +- src/formats/osml/OsmlStreamParser.hpp | 11 +- test/formats/osml/OsmlStreamParserTest.cpp | 438 ++++++++++++++++++----------- 3 files changed, 299 insertions(+), 178 deletions(-) (limited to 'test') diff --git a/src/formats/osml/OsmlStreamParser.cpp b/src/formats/osml/OsmlStreamParser.cpp index 7e01a3c..e467dc5 100644 --- a/src/formats/osml/OsmlStreamParser.cpp +++ b/src/formats/osml/OsmlStreamParser.cpp @@ -127,7 +127,7 @@ private: /** * Set to true if this is a command with clear begin and end. */ - bool hasRange; + bool hasRange: 1; public: /** @@ -259,7 +259,7 @@ public: */ enum class State : uint8_t { COMMAND_START = 0, - COMMAND_END = 1, + RANGE_END = 1, FIELD_START = 2, FIELD_END = 3, ANNOTATION_START = 4, @@ -328,7 +328,7 @@ private: * * @return an internal State specifying whether an error occured (return * values State::REOVERABLE_ERROR or State::IRRECOVERABLE_ERROR) or a - * command was actually ended (return value State::COMMAND_END). + * command was actually ended (return value State::RANGE_END). */ State parseEndCommand(); @@ -569,7 +569,7 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parseEndCommand() // End the current command location = name.getLocation(); commands.pop(); - return State::COMMAND_END; + return State::RANGE_END; } Variant OsmlStreamParserImpl::parseCommandArguments(Variant commandArgName) @@ -808,14 +808,15 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() // If this was an annotation start token, add the parsed < to the // output + SourceOffset charStart = token.location.getStart(); + SourceOffset charEnd = reader.getPeekOffset(); if (type == OsmlTokens.AnnotationStart) { - data.append('<', token.location.getStart(), - token.location.getStart() + 1); + data.append('<', charStart, charStart + 1); + charStart = charStart + 1; } // Append the character to the output data, mark it as protected - data.append(c, token.location.getStart(), reader.getPeekOffset(), - true); + data.append(c, charStart, charEnd, true); reader.consumePeek(); continue; } else if (type == Tokens::Data) { @@ -880,11 +881,12 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() // Make sure all open commands and fields have been ended at the end of the // stream - while (commands.size() > 1) { + while (true) { + bool topLevelCommand = commands.size() == 1U; if (cmd().inField()) { // If the stream ended with an open range field, issue information // about the range field - if (cmd().inRangeField()) { + if (cmd().inRangeField() && !topLevelCommand) { // Inform about the still open command itself logger.error("Reached end of stream, but command \"" + getCommandName().asString() + @@ -901,7 +903,11 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() } } } - commands.pop(); + if (!topLevelCommand) { + commands.pop(); + } else { + break; + } } location = SourceLocation{reader.getSourceId(), reader.getOffset()}; diff --git a/src/formats/osml/OsmlStreamParser.hpp b/src/formats/osml/OsmlStreamParser.hpp index 1fee90b..10d5296 100644 --- a/src/formats/osml/OsmlStreamParser.hpp +++ b/src/formats/osml/OsmlStreamParser.hpp @@ -69,11 +69,11 @@ public: COMMAND_START = 0, /** - * State returned if a range command has just ended. This state is not - * returned for non-range commands (as the actual end of a command is - * context dependant). + * State returned if a range command or range annotation has just ended. + * This state is not returned for non-range commands (as the actual end + * of a command is context dependent). */ - COMMAND_END = 1, + RANGE_END = 1, /** * State returned if a new field started. The reader assures that the @@ -185,7 +185,8 @@ public: /** * Returns true if the currently started command is a range command, only - * valid if State::COMMAND_START was returned by the "parse" function. + * valid if State::COMMAND_START or State::ANNOTATION_START was returned by + * the "parse" function. * * @return true if the command is started is a range command, false * otherwise. diff --git a/test/formats/osml/OsmlStreamParserTest.cpp b/test/formats/osml/OsmlStreamParserTest.cpp index 8b64e51..3e7f4c1 100644 --- a/test/formats/osml/OsmlStreamParserTest.cpp +++ b/test/formats/osml/OsmlStreamParserTest.cpp @@ -70,11 +70,11 @@ static void assertCommand(OsmlStreamParser &parser, assertCommandStart(parser, name, false, Variant::mapType{}, start, end); } -static void assertCommandEnd(OsmlStreamParser &parser, +static void assertRangeEnd(OsmlStreamParser &parser, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { - ASSERT_EQ(OsmlStreamParser::State::COMMAND_END, parser.parse()); + ASSERT_EQ(OsmlStreamParser::State::RANGE_END, parser.parse()); if (start != InvalidSourceOffset) { EXPECT_EQ(start, parser.getLocation().getStart()); } @@ -510,6 +510,61 @@ TEST(OsmlStreamParser, fields) assertEnd(parser, 14, 14); } +TEST(OsmlStreamParser, fieldsWithoutCommand) +{ + const char *testString = "{a}{b}{c}"; + // 012345678 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + assertFieldStart(parser, false, 0, 1); + assertTextData(parser, "a", 1, 2, 1, 2, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 2, 3); + + assertFieldStart(parser, false, 3, 4); + assertTextData(parser, "b", 4, 5, 4, 5, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 5, 6); + + assertFieldStart(parser, false, 6, 7); + assertTextData(parser, "c", 7, 8, 7, 8, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 8, 9); + assertEnd(parser, 9, 9); +} + +TEST(OsmlStreamParser, nestedField) +{ + const char *testString = "{{a{b}}}"; + // 01234567 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + assertFieldStart(parser, false, 0, 1); + assertFieldStart(parser, false, 1, 2); + assertTextData(parser, "a", 2, 3, 2, 3, WhitespaceMode::PRESERVE); + assertFieldStart(parser, false, 3, 4); + assertTextData(parser, "b", 4, 5, 4, 5, WhitespaceMode::PRESERVE); + assertFieldEnd(parser, 5, 6); + assertFieldEnd(parser, 6, 7); + assertFieldEnd(parser, 7, 8); + assertEnd(parser, 8, 8); +} + +TEST(OsmlStreamParser, errorUnbalancedField) +{ + const char *testString = "{a"; + // 01 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + logger.reset(); + + assertFieldStart(parser, false, 0, 1); + assertTextData(parser, "a", 1, 2, 1, 2, WhitespaceMode::PRESERVE); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 2, 2); + ASSERT_TRUE(logger.hasError()); +} + TEST(OsmlStreamParser, dataOutsideField) { const char *testString = "\\test{a}{b} c"; @@ -720,7 +775,7 @@ TEST(OsmlStreamParser, beginEnd) OsmlStreamParser parser(charReader, logger); assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertCommandEnd(parser, 17, 21); + assertRangeEnd(parser, 17, 21); assertEnd(parser, 22, 22); } @@ -734,7 +789,7 @@ TEST(OsmlStreamParser, beginEndWithName) OsmlStreamParser parser(charReader, logger); assertCommandStart(parser, "book", true, {{"name", "a"}}, 7, 11); - assertCommandEnd(parser, 19, 23); + assertRangeEnd(parser, 19, 23); assertEnd(parser, 24, 24); } @@ -749,7 +804,7 @@ TEST(OsmlStreamParser, beginEndWithNameAndArgs) assertCommandStart(parser, "book", true, {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); - assertCommandEnd(parser, 37, 41); + assertRangeEnd(parser, 37, 41); assertEnd(parser, 42, 42); } @@ -775,7 +830,7 @@ TEST(OsmlStreamParser, beginEndWithNameAndArgsMultipleFields) assertFieldStart(parser, false, 49, 50); assertFieldEnd(parser, 50, 51); assertFieldEnd(parser, 51, 52); - assertCommandEnd(parser, 57, 61); + assertRangeEnd(parser, 57, 61); assertEnd(parser, 62, 62); } @@ -790,10 +845,10 @@ TEST(OsmlStreamParser, beginEndWithData) assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); assertData(parser, "a", 12, 13); - assertCommandEnd(parser, 18, 22); + assertRangeEnd(parser, 18, 22); assertEnd(parser, 23, 23); } -/* + TEST(OsmlStreamParser, beginEndNested) { const char *testString = @@ -802,29 +857,32 @@ TEST(OsmlStreamParser, beginEndNested) // 0 1 2 3 4 5 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - - assertCommand(reader, "a", 7, 8); - assertFieldStart(reader, false, 9, 10); - assertData(reader, "b", 10, 11); - assertFieldEnd(reader, 11, 12); - assertFieldStart(reader, true, 13, 14); - assertData(reader, "c", 13, 14); - assertCommand(reader, "d", 22, 23); - assertFieldStart(reader, false, 24, 25); - assertData(reader, "e", 25, 26); - assertFieldEnd(reader, 26, 27); - assertFieldStart(reader, false, 27, 28); - assertData(reader, "f", 28, 29); - assertFieldEnd(reader, 29, 30); - assertFieldStart(reader, true, 31, 32); - assertCommand(reader, "g", 31, 33); - assertFieldStart(reader, false, 33, 34); - assertData(reader, "h", 34, 35); - assertFieldEnd(reader, 35, 36); - assertFieldEnd(reader, 42, 43); - assertFieldEnd(reader, 49, 50); - assertEnd(reader, 51, 51); + OsmlStreamParser parser(charReader, logger); + + assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); + assertFieldStart(parser, false, 9, 10); + assertData(parser, "b", 10, 11); + assertFieldEnd(parser, 11, 12); + + assertData(parser, "c", 13, 14); + + assertCommandStart(parser, "d", true, Variant::mapType{}, 22, 23); + assertFieldStart(parser, false, 24, 25); + assertData(parser, "e", 25, 26); + assertFieldEnd(parser, 26, 27); + assertFieldStart(parser, false, 27, 28); + assertData(parser, "f", 28, 29); + assertFieldEnd(parser, 29, 30); + + assertEmptyData(parser); + assertCommand(parser, "g", 31, 33); + assertFieldStart(parser, false, 33, 34); + assertData(parser, "h", 34, 35); + assertFieldEnd(parser, 35, 36); + assertEmptyData(parser); + assertRangeEnd(parser, 42, 43); + assertRangeEnd(parser, 49, 50); + assertEnd(parser, 51, 51); } TEST(OsmlStreamParser, beginEndWithCommand) @@ -834,16 +892,75 @@ TEST(OsmlStreamParser, beginEndWithCommand) // 0 1 2 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); + + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertCommand(parser, "a", 12, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "test", 15, 19); + assertFieldEnd(parser, 19, 20); + assertRangeEnd(parser, 25, 29); + assertEnd(parser, 30, 30); +} + +TEST(OsmlStreamParser, beginEndNestedFields) +{ + const char *testString = "\\begin{book}a{{b{c}}}\\end{book}"; + // 012345678901234567890 1234567890 + // 0 1 2 3 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + logger.reset(); + + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertFieldStart(parser, false, 13, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "b", 15, 16); + assertFieldStart(parser, false, 16, 17); + assertData(parser, "c", 17, 18); + assertFieldEnd(parser, 18, 19); + assertFieldEnd(parser, 19, 20); + assertFieldEnd(parser, 20, 21); + assertRangeEnd(parser, 26, 30); + assertEnd(parser, 31, 31); +} + +TEST(OsmlStreamParser, errorBeginEndUnbalancedNestedFields) +{ + const char *testString = "\\begin{book}a{{b{c}}\\end{book}"; + // 012345678901234567890 123456789 + // 0 1 2 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + logger.reset(); + + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertFieldStart(parser, false, 13, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "b", 15, 16); + assertFieldStart(parser, false, 16, 17); + assertData(parser, "c", 17, 18); + assertFieldEnd(parser, 18, 19); + assertFieldEnd(parser, 19, 20); + ASSERT_THROW(assertRangeEnd(parser, 25, 29), LoggableException); +} + +TEST(OsmlStreamParser, errorBeginEndUnbalancedFields) +{ + const char *testString = "{a"; + // 01 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "book", 7, 11); - assertFieldStart(reader, true, 12, 13); - assertCommand(reader, "a", 12, 14); - assertFieldStart(reader, false, 14, 15); - assertData(reader, "test", 15, 19); - assertFieldEnd(reader, 19, 20); - assertFieldEnd(reader, 25, 29); - assertEnd(reader, 30, 30); + logger.reset(); + + assertFieldStart(parser, false, 0, 1); + assertTextData(parser, "a", 1, 2, 1, 2, WhitespaceMode::PRESERVE); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 2, 2); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoBraceOpen) @@ -852,12 +969,13 @@ TEST(OsmlStreamParser, errorBeginNoBraceOpen) // 01234567 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertData(reader, "a", 7, 8); + assertData(parser, "a", 7, 8); ASSERT_TRUE(logger.hasError()); + assertEnd(parser, 8, 8); } TEST(OsmlStreamParser, errorBeginNoIdentifier) @@ -865,7 +983,7 @@ TEST(OsmlStreamParser, errorBeginNoIdentifier) const char *testString = "\\begin{!"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); @@ -878,7 +996,7 @@ TEST(OsmlStreamParser, errorBeginNoBraceClose) const char *testString = "\\begin{a"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); @@ -891,15 +1009,15 @@ TEST(OsmlStreamParser, errorBeginNoName) const char *testString = "\\begin{a#}"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "a"); + assertCommandStart(parser, "a", true); ASSERT_TRUE(logger.hasError()); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertEnd(reader); + assertEnd(parser); ASSERT_TRUE(logger.hasError()); } @@ -909,11 +1027,11 @@ TEST(OsmlStreamParser, errorEndNoBraceOpen) // 012345 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertData(reader, "a", 5, 6); + assertData(parser, "a", 5, 6); ASSERT_TRUE(logger.hasError()); } @@ -922,7 +1040,7 @@ TEST(OsmlStreamParser, errorEndNoIdentifier) const char *testString = "\\end{!"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); @@ -935,7 +1053,7 @@ TEST(OsmlStreamParser, errorEndNoBraceClose) const char *testString = "\\end{a"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); @@ -948,7 +1066,7 @@ TEST(OsmlStreamParser, errorEndNoBegin) const char *testString = "\\end{a}"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); @@ -963,14 +1081,13 @@ TEST(OsmlStreamParser, errorBeginEndMismatch) // 0 1 2 3 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); - assertCommand(reader, "a", 7, 8); - assertFieldStart(reader, true, 10, 11); - assertCommand(reader, "b", 17, 18); - assertFieldStart(reader, true, 20, 24); - assertData(reader, "test", 20, 24); + assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); + assertEmptyData(parser); + assertCommandStart(parser, "b", true, Variant::mapType{}, 17, 18); + assertData(parser, "test", 20, 24); ASSERT_FALSE(logger.hasError()); ASSERT_THROW(parser.parse(), LoggableException); ASSERT_TRUE(logger.hasError()); @@ -982,10 +1099,10 @@ TEST(OsmlStreamParser, commandWithNSSep) // 012345678901 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test1:test2", 0, 12); - assertEnd(reader, 12, 12); + assertCommand(parser, "test1:test2", 0, 12); + assertEnd(parser, 12, 12); } TEST(OsmlStreamParser, beginEndWithNSSep) @@ -995,12 +1112,11 @@ TEST(OsmlStreamParser, beginEndWithNSSep) // 0 1 2 3 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "test1:test2", 7, 18); - assertFieldStart(reader, true, 19, 20); - assertFieldEnd(reader, 24, 35); - assertEnd(reader, 36, 36); + assertCommandStart(parser, "test1:test2", true, Variant::mapType{}, 7, 18); + assertRangeEnd(parser, 24, 35); + assertEnd(parser, 36, 36); } TEST(OsmlStreamParser, errorBeginNSSep) @@ -1008,15 +1124,14 @@ TEST(OsmlStreamParser, errorBeginNSSep) const char *testString = "\\begin:test{blub}\\end{blub}"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "blub"); + assertCommandStart(parser, "blub", true, Variant::mapType{}); ASSERT_TRUE(logger.hasError()); - assertFieldStart(reader, true); - assertFieldEnd(reader); - assertEnd(reader); + assertRangeEnd(parser); + assertEnd(parser); } TEST(OsmlStreamParser, errorEndNSSep) @@ -1024,15 +1139,14 @@ TEST(OsmlStreamParser, errorEndNSSep) const char *testString = "\\begin{blub}\\end:test{blub}"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); - assertCommand(reader, "blub"); - assertFieldStart(reader, true); + assertCommandStart(parser, "blub", true, Variant::mapType{}); ASSERT_FALSE(logger.hasError()); - assertFieldEnd(reader); + assertRangeEnd(parser); ASSERT_TRUE(logger.hasError()); - assertEnd(reader); + assertEnd(parser); } TEST(OsmlStreamParser, errorEmptyNs) @@ -1040,14 +1154,14 @@ TEST(OsmlStreamParser, errorEmptyNs) const char *testString = "\\test:"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "test"); + assertCommand(parser, "test"); ASSERT_TRUE(logger.hasError()); - assertData(reader, ":"); - assertEnd(reader); + assertData(parser, ":"); + assertEnd(parser); } TEST(OsmlStreamParser, errorRepeatedNs) @@ -1055,14 +1169,14 @@ TEST(OsmlStreamParser, errorRepeatedNs) const char *testString = "\\test::"; CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "test"); + assertCommand(parser, "test"); ASSERT_TRUE(logger.hasError()); - assertData(reader, "::"); - assertEnd(reader); + assertData(parser, "::"); + assertEnd(parser); } TEST(OsmlStreamParser, explicitDefaultField) @@ -1071,14 +1185,14 @@ TEST(OsmlStreamParser, explicitDefaultField) // 01234567 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertData(reader, "b", 4, 5); - assertFieldEnd(reader, 5, 6); - assertData(reader, "c", 6, 7); - assertEnd(reader, 7, 7); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertData(parser, "b", 4, 5); + assertFieldEnd(parser, 5, 6); + assertData(parser, "c", 6, 7); + assertEnd(parser, 7, 7); } TEST(OsmlStreamParser, explicitDefaultFieldWithCommand) @@ -1087,33 +1201,33 @@ TEST(OsmlStreamParser, explicitDefaultFieldWithCommand) // 0123 4567 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertCommand(reader, "b", 4, 6); - assertFieldEnd(reader, 6, 7); - assertData(reader, "c", 7, 8); - assertEnd(reader, 8, 8); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertCommand(parser, "b", 4, 6); + assertFieldEnd(parser, 6, 7); + assertData(parser, "c", 7, 8); + assertEnd(parser, 8, 8); } -TEST(OsmlStreamParser, errorFieldAfterExplicitDefaultField) +TEST(OsmlStreamParser, fieldAfterExplicitDefaultField) { const char *testString = "\\a{!\\b}{c}"; // 0123 456789 CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); - assertCommand(reader, "a", 0, 2); - assertFieldStart(reader, true, 2, 4); - assertCommand(reader, "b", 4, 6); - assertFieldEnd(reader, 6, 7); - ASSERT_FALSE(logger.hasError()); - assertData(reader, "c", 8, 9); - ASSERT_TRUE(logger.hasError()); - assertEnd(reader, 10, 10); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertCommand(parser, "b", 4, 6); + assertFieldEnd(parser, 6, 7); + assertFieldStart(parser, false, 7, 8); + assertData(parser, "c", 8, 9); + assertFieldEnd(parser, 9, 10); + assertEnd(parser, 10, 10); } TEST(OsmlStreamParser, annotationStart) @@ -1123,10 +1237,10 @@ TEST(OsmlStreamParser, annotationStart) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); - assertEnd(reader, 3, 3); + assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); + assertEnd(parser, 3, 3); } TEST(OsmlStreamParser, annotationStartWithName) @@ -1137,11 +1251,11 @@ TEST(OsmlStreamParser, annotationStartWithName) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(reader, "annotationWithName", + assertAnnotationStart(parser, "annotationWithName", Variant::mapType{{"name", "aName"}}, 0, 20); - assertEnd(reader, 26, 26); + assertEnd(parser, 26, 26); } TEST(OsmlStreamParser, annotationStartWithArguments) @@ -1152,12 +1266,12 @@ TEST(OsmlStreamParser, annotationStartWithArguments) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); assertAnnotationStart( - reader, "annotationWithName", + parser, "annotationWithName", Variant::mapType{{"name", "aName"}, {"a", 1}, {"b", 2}}, 0, 20); - assertEnd(reader, 35, 35); + assertEnd(parser, 35, 35); } TEST(OsmlStreamParser, simpleAnnotationStartBeginEnd) @@ -1168,16 +1282,16 @@ TEST(OsmlStreamParser, simpleAnnotationStartBeginEnd) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); assertAnnotationStart( - reader, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, + parser, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, 10); - assertFieldStart(reader, true, 26, 27); - assertData(reader, "a", 26, 27); - assertFieldEnd(reader, 33, 35); - assertAnnotationEnd(reader, "", "", 36, 38); - assertEnd(reader, 38, 38); + ASSERT_TRUE(parser.inRangeCommand()); + assertData(parser, "a", 26, 27); + assertRangeEnd(parser, 33, 35); + assertAnnotationEnd(parser, "", "", 36, 38); + assertEnd(parser, 38, 38); } TEST(OsmlStreamParser, annotationEnd) @@ -1187,10 +1301,10 @@ TEST(OsmlStreamParser, annotationEnd) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(reader, "a", "", 0, 2); - assertEnd(reader, 3, 3); + assertAnnotationEnd(parser, "a", "", 0, 2); + assertEnd(parser, 3, 3); } TEST(OsmlStreamParser, annotationEndWithName) @@ -1200,10 +1314,10 @@ TEST(OsmlStreamParser, annotationEndWithName) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(reader, "a", "name", 0, 2); - assertEnd(reader, 8, 8); + assertAnnotationEnd(parser, "a", "name", 0, 2); + assertEnd(parser, 8, 8); } TEST(OsmlStreamParser, annotationEndWithNameAsArgs) @@ -1213,10 +1327,10 @@ TEST(OsmlStreamParser, annotationEndWithNameAsArgs) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(reader, "a", "name", 0, 2); - assertEnd(reader, 14, 14); + assertAnnotationEnd(parser, "a", "name", 0, 2); + assertEnd(parser, 14, 14); } TEST(OsmlStreamParser, errorAnnotationEndWithArguments) @@ -1227,14 +1341,14 @@ TEST(OsmlStreamParser, errorAnnotationEndWithArguments) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); logger.reset(); ASSERT_FALSE(logger.hasError()); - assertCommand(reader, "a", Variant::mapType{{"foo", "bar"}}, 0, 2); + assertCommandStart(parser, "a", false, Variant::mapType{{"foo", "bar"}}, 0, 2); ASSERT_TRUE(logger.hasError()); - assertData(reader, ">", 11, 12); - assertEnd(reader, 12, 12); + assertData(parser, ">", 11, 12); + assertEnd(parser, 12, 12); } TEST(OsmlStreamParser, closingAnnotation) @@ -1244,11 +1358,11 @@ TEST(OsmlStreamParser, closingAnnotation) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(reader, "a", Variant::mapType{}, 0, 3); - assertData(reader, ">", 3, 4); - assertEnd(reader, 4, 4); + assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); + assertData(parser, ">", 3, 4); + assertEnd(parser, 4, 4); } TEST(OsmlStreamParser, annotationWithFields) @@ -1259,23 +1373,23 @@ TEST(OsmlStreamParser, annotationWithFields) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); - - assertData(reader, "a", 0, 1); - assertAnnotationStart(reader, "b", Variant::mapType{}, 2, 5); - assertFieldStart(reader, false, 5, 6); - assertData(reader, "c", 6, 7); - assertFieldEnd(reader, 7, 8); - assertFieldStart(reader, false, 8, 9); - assertData(reader, "d", 9, 10); - assertFieldEnd(reader, 10, 11); - assertFieldStart(reader, true, 11, 13); - assertData(reader, "e", 13, 14); - assertFieldEnd(reader, 14, 15); - assertData(reader, "f", 16, 17); - assertAnnotationEnd(reader, "", "", 18, 20); - assertData(reader, "g", 21, 22); - assertEnd(reader, 22, 22); + OsmlStreamParser parser(charReader, logger); + + assertData(parser, "a", 0, 1); + assertAnnotationStart(parser, "b", Variant::mapType{}, 2, 5); + assertFieldStart(parser, false, 5, 6); + assertData(parser, "c", 6, 7); + assertFieldEnd(parser, 7, 8); + assertFieldStart(parser, false, 8, 9); + assertData(parser, "d", 9, 10); + assertFieldEnd(parser, 10, 11); + assertFieldStart(parser, true, 11, 13); + assertData(parser, "e", 13, 14); + assertFieldEnd(parser, 14, 15); + assertData(parser, "f", 16, 17); + assertAnnotationEnd(parser, "", "", 18, 20); + assertData(parser, "g", 21, 22); + assertEnd(parser, 22, 22); } TEST(OsmlStreamParser, annotationStartEscape) @@ -1286,11 +1400,11 @@ TEST(OsmlStreamParser, annotationStartEscape) CharReader charReader(testString); - OsmlStreamParser reader(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertData(reader, "<%test", 0, 7); - assertEnd(reader, 7, 7); + assertData(parser, "<%test", 0, 7); + assertEnd(parser, 7, 7); } -*/ + } -- cgit v1.2.3 From cb6cacdc7eade9d4290767bafb7ccf4e935d0fbf Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 1 Mar 2015 13:49:26 +0100 Subject: allowing to store gaps in SourceOffsetVector and fixed bug with trim not resetting offsets correctly when the new length is zero --- src/core/parser/utils/SourceOffsetVector.hpp | 64 ++++++++++++++++------- test/core/parser/utils/SourceOffsetVectorTest.cpp | 2 +- 2 files changed, 47 insertions(+), 19 deletions(-) (limited to 'test') diff --git a/src/core/parser/utils/SourceOffsetVector.hpp b/src/core/parser/utils/SourceOffsetVector.hpp index 67bacef..f322a88 100644 --- a/src/core/parser/utils/SourceOffsetVector.hpp +++ b/src/core/parser/utils/SourceOffsetVector.hpp @@ -33,6 +33,7 @@ #include #include #include +#include #include @@ -43,6 +44,9 @@ namespace ousia { * a delta compression. */ class SourceOffsetVector { +public: + using OffsPair = std::pair; + private: /** * Type used for representing the length of a character. @@ -81,10 +85,13 @@ private: */ std::vector offsets; + /** + * Map used to store discontinuities in the character offsets. + */ + std::unordered_map gaps; + /** * Last position given as "end" position in the storeOffset() method. - * Used to adapt the length of the previous element in case start and end - * positions do not match. */ SourceOffset lastEnd; @@ -105,19 +112,22 @@ public: // Make sure (end - start) is smaller than MAX_LEN assert(end - start < MAX_LEN); - // Adapt the length of the previous character in case there is a gap - if (!lens.empty() && start > lastEnd) { - lens.back() += start - lastEnd; - } - lastEnd = end; - // Store an absolute offset every OFFSET_INTERVAL elements if ((lens.size() & OFFSET_INTERVAL_MASK) == 0) { offsets.push_back(start); } - // Store the length - lens.push_back(end - start); + // Adapt the length of the previous character in case there is a gap + if (!lens.empty() && start > lastEnd) { + // There is a discontinuity, store the given offsets in the "gaps" + // map + gaps[lens.size()] = OffsPair(start, end); + lens.push_back(MAX_LEN); + } else { + // Store the length + lens.push_back(end - start); + } + lastEnd = end; } /** @@ -127,14 +137,13 @@ public: * read. * @return a pair containing start and end source offset. */ - std::pair loadOffset(size_t idx) const + OffsPair loadOffset(size_t idx) const { // Special treatment for the last character const size_t count = lens.size(); if (idx > 0 && idx == count) { auto offs = loadOffset(count - 1); - return std::pair(offs.second, - offs.second); + return OffsPair(offs.second, offs.second); } // Calculate the start index in the lens vector and in the offsets @@ -146,12 +155,26 @@ public: assert(idx < count); assert(offsetIdx < offsets.size()); + // If the length of the last character is MAX_LEN, the position is + // stored in the "gaps" list + if (lens[idx] == MAX_LEN) { + auto it = gaps.find(idx); + assert(it != gaps.end()); + return it->second; + } + // Sum over the length starting with the start offset SourceOffset start = offsets[offsetIdx]; for (size_t i = sumStartIdx; i < idx; i++) { - start += lens[i]; + if (lens[i] == MAX_LEN) { + auto it = gaps.find(i); + assert(it != gaps.end()); + start = it->second.first; + } else { + start += lens[i]; + } } - return std::pair(start, start + lens[idx]); + return OffsPair(start, start + lens[idx]); } /** @@ -166,13 +189,16 @@ public: * @param length is the number of characters to which the TokenizedData * instance should be trimmed. */ - void trim(size_t length) { + void trim(size_t length) + { if (length < size()) { lens.resize(length); - offsets.resize((length >> LOG2_OFFSET_INTERVAL) + 1); if (length > 0) { + offsets.resize((length >> LOG2_OFFSET_INTERVAL) + 1); lastEnd = loadOffset(length - 1).second; } else { + offsets.clear(); + gaps.clear(); lastEnd = 0; } } @@ -182,9 +208,11 @@ public: * Resets the SourceOffsetVector to the state it had when it was * constructed. */ - void clear() { + void clear() + { lens.clear(); offsets.clear(); + gaps.clear(); lastEnd = 0; } }; diff --git a/test/core/parser/utils/SourceOffsetVectorTest.cpp b/test/core/parser/utils/SourceOffsetVectorTest.cpp index 25a4163..26254f9 100644 --- a/test/core/parser/utils/SourceOffsetVectorTest.cpp +++ b/test/core/parser/utils/SourceOffsetVectorTest.cpp @@ -51,7 +51,7 @@ TEST(SourceOffsetVector, gaps) for (size_t i = 0; i < 999; i++) { auto elem = vec.loadOffset(i); EXPECT_EQ(i * 3 + 5, elem.first); - EXPECT_EQ((i + 1) * 3 + 5, elem.second); + EXPECT_EQ(i * 3 + 7, elem.second); } auto elem = vec.loadOffset(999); EXPECT_EQ(999U * 3 + 5, elem.first); -- cgit v1.2.3 From 31c83c05d257c9a7a336f12342c401f97d380674 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 1 Mar 2015 13:50:15 +0100 Subject: Prefer longer non-primary tokens --- src/core/parser/utils/Tokenizer.cpp | 45 +++++----- test/core/parser/utils/TokenizerTest.cpp | 148 ++++++++++++++++++++++++++----- 2 files changed, 150 insertions(+), 43 deletions(-) (limited to 'test') diff --git a/src/core/parser/utils/Tokenizer.cpp b/src/core/parser/utils/Tokenizer.cpp index 94d9cb0..8d540a6 100644 --- a/src/core/parser/utils/Tokenizer.cpp +++ b/src/core/parser/utils/Tokenizer.cpp @@ -188,7 +188,7 @@ bool Tokenizer::next(CharReader &reader, Token &token, TokenizedData &data) const size_t dataStartOffset = data.size(); // If we do not have a match yet, start a new lookup from the root - if (!bestMatch.hasMatch()) { + if (!bestMatch.hasMatch() || !bestMatch.primary) { lookups.emplace_back(root, charStart, dataStartOffset); } @@ -201,36 +201,35 @@ bool Tokenizer::next(CharReader &reader, Token &token, TokenizedData &data) continue; } - // If the matched token is primary, check whether it is better than - // the current best match, if yes, replace the best match. In any - // case just continue - if (match.primary) { - if (match.size() > bestMatch.size()) { - bestMatch = match; - } - continue; + // Replace the best match with longest token + if (match.size() > bestMatch.size()) { + bestMatch = match; } - // Otherwise -- if the matched token is a non-primary token (and no - // primary token has been found until now) -- mark the match in the - // TokenizedData - if (!bestMatch.hasMatch()) { + // If the matched token is a non-primary token -- mark the match in + // the TokenizedData list + if (!match.primary) { data.mark(match.token.id, data.size() - match.size() + 1, match.size()); } } - // We have found a token and there are no more states to advance or the - // text handler has found something -- abort to return the new token - if (bestMatch.hasMatch()) { - if ((nextLookups.empty() || data.size() > initialDataSize)) { + + // If a token has been found and the token is a primary token, check + // whether we have to abort, otherwise if we have a non-primary match, + // reset it once it can no longer be advanced + if (bestMatch.hasMatch() && nextLookups.empty()) { + if (bestMatch.primary) { break; + } else { + bestMatch = TokenMatch{}; } - } else { - // Record all incomming characters - data.append(c, charStart, charEnd); } + // Record all incomming characters + data.append(c, charStart, charEnd); + + // Swap the lookups and the nextLookups list lookups = std::move(nextLookups); nextLookups.clear(); @@ -241,17 +240,17 @@ bool Tokenizer::next(CharReader &reader, Token &token, TokenizedData &data) // If we found data, emit a corresponding data token if (data.size() > initialDataSize && - (!bestMatch.hasMatch() || + (!bestMatch.hasMatch() || !bestMatch.primary || bestMatch.dataStartOffset > initialDataSize)) { // If we have a "bestMatch" wich starts after text data has started, // trim the TokenizedData to this offset - if (bestMatch.dataStartOffset > initialDataSize) { + if (bestMatch.dataStartOffset > initialDataSize && bestMatch.primary) { data.trim(bestMatch.dataStartOffset); } // Create a token containing the data location bestMatch.token = Token{data.getLocation()}; - } else if (bestMatch.hasMatch() && + } else if (bestMatch.hasMatch() && bestMatch.primary && bestMatch.dataStartOffset == initialDataSize) { data.trim(initialDataSize); } diff --git a/test/core/parser/utils/TokenizerTest.cpp b/test/core/parser/utils/TokenizerTest.cpp index 9f644c2..45fc77a 100644 --- a/test/core/parser/utils/TokenizerTest.cpp +++ b/test/core/parser/utils/TokenizerTest.cpp @@ -26,6 +26,60 @@ namespace ousia { +static void assertPrimaryToken(CharReader &reader, Tokenizer &tokenizer, + TokenId id, const std::string &text, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset, + SourceId sourceId = InvalidSourceId) +{ + Token token; + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + EXPECT_EQ(id, token.id); + EXPECT_EQ(text, token.content); + if (start != InvalidSourceOffset) { + EXPECT_EQ(start, token.getLocation().getStart()); + } + if (end != InvalidSourceOffset) { + EXPECT_EQ(end, token.getLocation().getEnd()); + } + EXPECT_EQ(sourceId, token.getLocation().getSourceId()); +} + +static void expectData(const std::string &expected, SourceOffset tokenStart, + SourceOffset tokenEnd, SourceOffset textStart, + SourceOffset textEnd, const Token &token, + TokenizedData &data, + WhitespaceMode mode = WhitespaceMode::PRESERVE) +{ + ASSERT_EQ(Tokens::Data, token.id); + + Token textToken; + TokenizedDataReader reader = data.reader(); + ASSERT_TRUE(reader.read(textToken, TokenSet{}, mode)); + + EXPECT_EQ(expected, textToken.content); + EXPECT_EQ(tokenStart, token.location.getStart()); + EXPECT_EQ(tokenEnd, token.location.getEnd()); + EXPECT_EQ(textStart, textToken.getLocation().getStart()); + EXPECT_EQ(textEnd, textToken.getLocation().getEnd()); + EXPECT_TRUE(reader.atEnd()); +} + +static void assertDataToken(CharReader &reader, Tokenizer &tokenizer, + const std::string &expected, + SourceOffset tokenStart, SourceOffset tokenEnd, + SourceOffset textStart, SourceOffset textEnd, + WhitespaceMode mode = WhitespaceMode::PRESERVE) +{ + Token token; + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + + expectData(expected, tokenStart, tokenEnd, textStart, textEnd, token, data, + mode); +} + TEST(Tokenizer, tokenRegistration) { Tokenizer tokenizer; @@ -53,25 +107,6 @@ TEST(Tokenizer, tokenRegistration) ASSERT_EQ("d", tokenizer.lookupToken(1U).string); } -void expectData(const std::string &expected, SourceOffset tokenStart, - SourceOffset tokenEnd, SourceOffset textStart, - SourceOffset textEnd, const Token &token, TokenizedData &data, - WhitespaceMode mode = WhitespaceMode::PRESERVE) -{ - ASSERT_EQ(Tokens::Data, token.id); - - Token textToken; - TokenizedDataReader reader = data.reader(); - ASSERT_TRUE(reader.read(textToken, TokenSet{}, mode)); - - EXPECT_EQ(expected, textToken.content); - EXPECT_EQ(tokenStart, token.location.getStart()); - EXPECT_EQ(tokenEnd, token.location.getEnd()); - EXPECT_EQ(textStart, textToken.getLocation().getStart()); - EXPECT_EQ(textEnd, textToken.getLocation().getEnd()); - EXPECT_TRUE(reader.atEnd()); -} - TEST(Tokenizer, textTokenPreserveWhitespace) { { @@ -451,6 +486,80 @@ TEST(Tokenizer, nonPrimaryTokens) ASSERT_FALSE(tokenizer.read(reader, token, data)); } +TEST(Tokenizer, primaryNonPrimaryTokenInteraction) +{ + CharReader reader{"<><<<>>"}; + // 01234567890123456789012 3456789012345 + // 0 1 2 3 + + Tokenizer tokenizer; + + TokenId tP1 = tokenizer.registerToken("<", true); + TokenId tP2 = tokenizer.registerToken(">", true); + TokenId tP3 = tokenizer.registerToken("\\>", true); + TokenId tN1 = tokenizer.registerToken("<<", false); + TokenId tN2 = tokenizer.registerToken(">>", false); + + TokenSet tokens = TokenSet{tN1, tN2}; + + Token token, textToken; + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertToken(dataReader, tN1, "<<", tokens, WhitespaceMode::TRIM, 0, 2); + assertText(dataReader, "test1", tokens, WhitespaceMode::TRIM, 2, 7); + assertToken(dataReader, tN2, ">>", tokens, WhitespaceMode::TRIM, 7, 9); + assertEnd(dataReader); + } + + assertPrimaryToken(reader, tokenizer, tP1, "<", 9, 10); + assertDataToken(reader, tokenizer, "test2", 10, 15, 10, 15); + assertPrimaryToken(reader, tokenizer, tP2, ">", 15, 16); + + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertToken(dataReader, tN1, "<<", tokens, WhitespaceMode::TRIM, 16, 18); + assertText(dataReader, "test3", tokens, WhitespaceMode::TRIM, 18, 23); + assertEnd(dataReader); + } + + assertPrimaryToken(reader, tokenizer, tP3, "\\>", 23, 25); + + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertToken(dataReader, tN1, "<<", tokens, WhitespaceMode::TRIM, 25, 27); + assertEnd(dataReader); + } + + assertPrimaryToken(reader, tokenizer, tP1, "<", 27, 28); + + { + TokenizedData data; + ASSERT_TRUE(tokenizer.read(reader, token, data)); + ASSERT_EQ(Tokens::Data, token.id); + + TokenizedDataReader dataReader = data.reader(); + assertText(dataReader, "test4", tokens, WhitespaceMode::TRIM, 28, 33); + assertToken(dataReader, tN2, ">>", tokens, WhitespaceMode::TRIM, 33, 35); + assertEnd(dataReader); + } + + assertPrimaryToken(reader, tokenizer, tP2, ">", 35, 36); + + TokenizedData data; + ASSERT_FALSE(tokenizer.read(reader, token, data)); +} TEST(Tokenizer, ambiguousTokens2) { @@ -476,6 +585,5 @@ TEST(Tokenizer, ambiguousTokens2) ASSERT_FALSE(tokenizer.read(reader, token, data)); } } - } -- cgit v1.2.3 From 4e199ad0d5c5d94955839da2a52967b4f0f34a43 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 1 Mar 2015 13:52:34 +0100 Subject: Implemented registration of user-defined tokens, fixed comment handling (do not issue multiple data events if a comment occurs, just skip the comment data like in TeX) --- src/formats/osml/OsmlStreamParser.cpp | 50 +++++++++++++++++++++++------- src/formats/osml/OsmlStreamParser.hpp | 11 +++++-- test/formats/osml/OsmlStreamParserTest.cpp | 13 +++----- 3 files changed, 52 insertions(+), 22 deletions(-) (limited to 'test') diff --git a/src/formats/osml/OsmlStreamParser.cpp b/src/formats/osml/OsmlStreamParser.cpp index e467dc5..823075a 100644 --- a/src/formats/osml/OsmlStreamParser.cpp +++ b/src/formats/osml/OsmlStreamParser.cpp @@ -16,6 +16,10 @@ along with this program. If not, see . */ +#include +#include +#include + #include #include #include @@ -27,9 +31,6 @@ #include "OsmlStreamParser.hpp" -#include -#include - namespace ousia { namespace { @@ -127,7 +128,7 @@ private: /** * Set to true if this is a command with clear begin and end. */ - bool hasRange: 1; + bool hasRange; public: /** @@ -407,6 +408,9 @@ public: State parse(); + TokenId registerToken(const std::string &token); + void unregisterToken(TokenId token); + const TokenizedData &getData() const { return data; } const Variant &getCommandName() const { return cmd().getName(); } const Variant &getCommandArguments() const { return cmd().getArguments(); } @@ -700,10 +704,11 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parseCommand( void OsmlStreamParserImpl::parseBlockComment() { Token token; + TokenizedData commentData; size_t depth = 1; - while (tokenizer.read(reader, token, data)) { + while (tokenizer.read(reader, token, commentData)) { // Throw the comment data away - data.clear(); + commentData.clear(); if (token.id == OsmlTokens.BlockCommentEnd) { depth--; @@ -822,6 +827,14 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() } else if (type == Tokens::Data) { reader.consumePeek(); continue; + } else if (type == OsmlTokens.LineComment) { + reader.consumePeek(); + parseLineComment(); + continue; + } else if (type == OsmlTokens.BlockCommentStart) { + reader.consumePeek(); + parseBlockComment(); + continue; } // A non-text token was reached, make sure all pending data commands @@ -836,11 +849,7 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() // Synchronize the location with the current token location location = token.location; - if (token.id == OsmlTokens.LineComment) { - parseLineComment(); - } else if (token.id == OsmlTokens.BlockCommentStart) { - parseBlockComment(); - } else if (token.id == OsmlTokens.FieldStart) { + if (token.id == OsmlTokens.FieldStart) { cmd().pushField(false, token.location); return State::FIELD_START; } else if (token.id == OsmlTokens.FieldEnd) { @@ -914,6 +923,16 @@ OsmlStreamParserImpl::State OsmlStreamParserImpl::parse() return State::END; } +TokenId OsmlStreamParserImpl::registerToken(const std::string &token) +{ + return tokenizer.registerToken(token, false); +} + +void OsmlStreamParserImpl::unregisterToken(TokenId token) +{ + assert(tokenizer.unregisterToken(token)); +} + /* Class OsmlStreamParser */ OsmlStreamParser::OsmlStreamParser(CharReader &reader, Logger &logger) @@ -955,4 +974,13 @@ bool OsmlStreamParser::inDefaultField() const { return impl->inDefaultField(); } bool OsmlStreamParser::inRangeCommand() const { return impl->inRangeCommand(); } +TokenId OsmlStreamParser::registerToken(const std::string &token) +{ + return impl->registerToken(token); +} + +void OsmlStreamParser::unregisterToken(TokenId token) +{ + impl->unregisterToken(token); +} } diff --git a/src/formats/osml/OsmlStreamParser.hpp b/src/formats/osml/OsmlStreamParser.hpp index 10d5296..b7e64f7 100644 --- a/src/formats/osml/OsmlStreamParser.hpp +++ b/src/formats/osml/OsmlStreamParser.hpp @@ -32,6 +32,8 @@ #include #include +#include + namespace ousia { // Forward declarations @@ -50,10 +52,10 @@ class Variant; * syntactically valid and tries to recorver from most errors. If an error is * irrecoverable (this is the case for errors with wrong nesting of commands or * fields, as this would lead to too many consecutive errors) a - * LoggableException is thrown. The OsmlStreamParser can be compared to a SAX - * parser for XML. + * LoggableException is thrown. In short, the OsmlStreamParser can be described + * as a SAX parser for OSML. */ -class OsmlStreamParser { +class OsmlStreamParser: public parser_stack::ParserCallbacks { public: /** * Enum used to indicate which state the OsmlStreamParser class is in @@ -204,6 +206,9 @@ public: * "{!" syntax). */ bool inDefaultField() const; + + TokenId registerToken(const std::string &token) override; + void unregisterToken(TokenId token) override; }; } diff --git a/test/formats/osml/OsmlStreamParserTest.cpp b/test/formats/osml/OsmlStreamParserTest.cpp index 3e7f4c1..0ea087f 100644 --- a/test/formats/osml/OsmlStreamParserTest.cpp +++ b/test/formats/osml/OsmlStreamParserTest.cpp @@ -363,8 +363,7 @@ TEST(OsmlStreamParser, singleLineComment) CharReader charReader(testString); OsmlStreamParser parser(charReader, logger); - assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); - assertTextData(parser, "b", 33, 34, 33, 34, WhitespaceMode::PRESERVE); + assertTextData(parser, "ab", 0, 34, 0, 34, WhitespaceMode::PRESERVE); assertEnd(parser, 34, 34); } @@ -376,8 +375,7 @@ TEST(OsmlStreamParser, multilineComment) CharReader charReader(testString); OsmlStreamParser parser(charReader, logger); - assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); - assertTextData(parser, "b", 40, 41, 40, 41, WhitespaceMode::PRESERVE); + assertTextData(parser, "ab", 0, 41, 0, 41, WhitespaceMode::PRESERVE); assertEnd(parser, 41, 41); } @@ -391,10 +389,10 @@ TEST(OsmlStreamParser, unfinishedMultilineComment) logger.reset(); - assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); ASSERT_FALSE(logger.hasError()); - assertEnd(parser, 38, 38); + assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); ASSERT_TRUE(logger.hasError()); + assertEnd(parser, 38, 38); } TEST(OsmlStreamParser, nestedMultilineComment) @@ -405,8 +403,7 @@ TEST(OsmlStreamParser, nestedMultilineComment) CharReader charReader(testString); OsmlStreamParser parser(charReader, logger); - assertTextData(parser, "a", 0, 1, 0, 1, WhitespaceMode::PRESERVE); - assertTextData(parser, "b", 40, 41, 40, 41, WhitespaceMode::PRESERVE); + assertTextData(parser, "ab", 0, 41, 0, 41, WhitespaceMode::PRESERVE); assertEnd(parser, 41, 41); } -- cgit v1.2.3 From 689348baf70d00e5ff1c8eec3959afc56071994e Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 1 Mar 2015 14:25:39 +0100 Subject: Added user defined tokens test --- test/formats/osml/OsmlStreamParserTest.cpp | 1104 ++++++++++++++-------------- 1 file changed, 568 insertions(+), 536 deletions(-) (limited to 'test') diff --git a/test/formats/osml/OsmlStreamParserTest.cpp b/test/formats/osml/OsmlStreamParserTest.cpp index 0ea087f..d47f529 100644 --- a/test/formats/osml/OsmlStreamParserTest.cpp +++ b/test/formats/osml/OsmlStreamParserTest.cpp @@ -24,17 +24,17 @@ #include #include #include - #include +#include + namespace ousia { static TerminalLogger logger(std::cerr, true); // static ConcreteLogger logger; static void assertCommandStart(OsmlStreamParser &parser, - const std::string &name, - bool rangeCommand, + const std::string &name, bool rangeCommand, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) { @@ -52,8 +52,7 @@ static void assertCommandStart(OsmlStreamParser &parser, } static void assertCommandStart(OsmlStreamParser &parser, - const std::string &name, - bool rangeCommand, + const std::string &name, bool rangeCommand, const Variant::mapType &args, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) @@ -62,17 +61,16 @@ static void assertCommandStart(OsmlStreamParser &parser, EXPECT_EQ(args, parser.getCommandArguments()); } -static void assertCommand(OsmlStreamParser &parser, - const std::string &name, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset) +static void assertCommand(OsmlStreamParser &parser, const std::string &name, + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) { assertCommandStart(parser, name, false, Variant::mapType{}, start, end); } static void assertRangeEnd(OsmlStreamParser &parser, - SourceOffset start = InvalidSourceOffset, - SourceOffset end = InvalidSourceOffset) + SourceOffset start = InvalidSourceOffset, + SourceOffset end = InvalidSourceOffset) { ASSERT_EQ(OsmlStreamParser::State::RANGE_END, parser.parse()); if (start != InvalidSourceOffset) { @@ -116,11 +114,12 @@ static void assertTextData(OsmlStreamParser &parser, const std::string &text, } static void assertData(OsmlStreamParser &parser, const std::string &text, - SourceOffset textStart = InvalidSourceOffset, - SourceOffset textEnd = InvalidSourceOffset, - WhitespaceMode mode = WhitespaceMode::COLLAPSE) + SourceOffset textStart = InvalidSourceOffset, + SourceOffset textEnd = InvalidSourceOffset, + WhitespaceMode mode = WhitespaceMode::COLLAPSE) { - assertTextData(parser, text, InvalidSourceOffset, InvalidSourceOffset, textStart, textEnd, mode); + assertTextData(parser, text, InvalidSourceOffset, InvalidSourceOffset, + textStart, textEnd, mode); } static void assertEmptyData(OsmlStreamParser &parser) @@ -134,7 +133,6 @@ static void assertEmptyData(OsmlStreamParser &parser) EXPECT_FALSE(dataReader.read(token, TokenSet{}, WhitespaceMode::TRIM)); } - static void assertFieldStart(OsmlStreamParser &parser, bool defaultField, SourceOffset start = InvalidSourceOffset, SourceOffset end = InvalidSourceOffset) @@ -585,363 +583,363 @@ TEST(OsmlStreamParser, dataOutsideField) TEST(OsmlStreamParser, nestedCommand) { - const char *testString = "\\test{a}{\\test2{b} c} d"; - // 012345678 90123456789012 - // 0 1 2 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + const char *testString = "\\test{a}{\\test2{b} c} d"; + // 012345678 90123456789012 + // 0 1 2 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "test", 0, 5); - assertFieldStart(parser, false, 5, 6); - assertData(parser, "a", 6, 7); - assertFieldEnd(parser, 7, 8); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertData(parser, "a", 6, 7); + assertFieldEnd(parser, 7, 8); - assertFieldStart(parser, false, 8, 9); - assertCommand(parser, "test2", 9, 15); - assertFieldStart(parser, false, 15, 16); - assertData(parser, "b", 16, 17); - assertFieldEnd(parser, 17, 18); - assertData(parser, "c", 19, 20); - assertFieldEnd(parser, 20, 21); - assertData(parser, "d", 22, 23); - assertEnd(parser, 23, 23); + assertFieldStart(parser, false, 8, 9); + assertCommand(parser, "test2", 9, 15); + assertFieldStart(parser, false, 15, 16); + assertData(parser, "b", 16, 17); + assertFieldEnd(parser, 17, 18); + assertData(parser, "c", 19, 20); + assertFieldEnd(parser, 20, 21); + assertData(parser, "d", 22, 23); + assertEnd(parser, 23, 23); } - TEST(OsmlStreamParser, nestedCommandImmediateEnd) { - const char *testString = "\\test{\\test2{b}} d"; - // 012345 678901234567 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); - - assertCommand(parser, "test", 0, 5); - assertFieldStart(parser, false, 5, 6); - { - assertCommand(parser, "test2", 6, 12); - assertFieldStart(parser, false, 12, 13); - assertData(parser, "b", 13, 14); - assertFieldEnd(parser, 14, 15); - } - assertFieldEnd(parser, 15, 16); - assertData(parser, "d", 17, 18); - assertEnd(parser, 18, 18); + const char *testString = "\\test{\\test2{b}} d"; + // 012345 678901234567 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + { + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertData(parser, "b", 13, 14); + assertFieldEnd(parser, 14, 15); + } + assertFieldEnd(parser, 15, 16); + assertData(parser, "d", 17, 18); + assertEnd(parser, 18, 18); } TEST(OsmlStreamParser, nestedCommandNoData) { - const char *testString = "\\test{\\test2}"; - // 012345 6789012 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + const char *testString = "\\test{\\test2}"; + // 012345 6789012 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "test", 0, 5); - assertFieldStart(parser, false, 5, 6); - assertCommand(parser, "test2", 6, 12); - assertFieldEnd(parser, 12, 13); - assertEnd(parser, 13, 13); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldEnd(parser, 12, 13); + assertEnd(parser, 13, 13); } TEST(OsmlStreamParser, multipleCommands) { - const char *testString = "\\a \\b \\c \\d"; - // 012 345 678 90 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + const char *testString = "\\a \\b \\c \\d"; + // 012 345 678 90 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "a", 0, 2); - assertEmptyData(parser); - assertCommand(parser, "b", 3, 5); - assertEmptyData(parser); - assertCommand(parser, "c", 6, 8); - assertEmptyData(parser); - assertCommand(parser, "d", 9, 11); - assertEnd(parser, 11, 11); + assertCommand(parser, "a", 0, 2); + assertEmptyData(parser); + assertCommand(parser, "b", 3, 5); + assertEmptyData(parser); + assertCommand(parser, "c", 6, 8); + assertEmptyData(parser); + assertCommand(parser, "d", 9, 11); + assertEnd(parser, 11, 11); } TEST(OsmlStreamParser, fieldsWithSpaces) { - const char *testString = "\\a {\\b \\c} \n\n {\\d}"; - // 0123 456 789012 3 456 789 - // 0 1 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); - - assertCommand(parser, "a", 0, 2); - assertEmptyData(parser); - assertFieldStart(parser, false, 3, 4); - assertCommand(parser, "b", 4, 6); - assertEmptyData(parser); - assertCommand(parser, "c", 7, 9); - assertFieldEnd(parser, 9, 10); - assertEmptyData(parser); - assertFieldStart(parser, false, 16, 17); - assertCommand(parser, "d", 17, 19); - assertFieldEnd(parser, 19, 20); - assertEnd(parser, 20, 20); + const char *testString = "\\a {\\b \\c} \n\n {\\d}"; + // 0123 456 789012 3 456 789 + // 0 1 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + + assertCommand(parser, "a", 0, 2); + assertEmptyData(parser); + assertFieldStart(parser, false, 3, 4); + assertCommand(parser, "b", 4, 6); + assertEmptyData(parser); + assertCommand(parser, "c", 7, 9); + assertFieldEnd(parser, 9, 10); + assertEmptyData(parser); + assertFieldStart(parser, false, 16, 17); + assertCommand(parser, "d", 17, 19); + assertFieldEnd(parser, 19, 20); + assertEnd(parser, 20, 20); } TEST(OsmlStreamParser, errorEndButOpenField) { - const char *testString = "\\a b {"; - // 012345 - // 0 - CharReader charReader(testString); + const char *testString = "\\a b {"; + // 012345 + // 0 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(parser, "a", 0, 2); - assertData(parser, "b", 3, 4); - assertFieldStart(parser, false, 5, 6); - ASSERT_FALSE(logger.hasError()); - assertEnd(parser, 6, 6); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "a", 0, 2); + assertData(parser, "b", 3, 4); + assertFieldStart(parser, false, 5, 6); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 6, 6); + ASSERT_TRUE(logger.hasError()); } - TEST(OsmlStreamParser, errorNoFieldToEnd) { - const char *testString = "\\a b }"; - // 012345 - // 0 - CharReader charReader(testString); + const char *testString = "\\a b }"; + // 012345 + // 0 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(parser, "a", 0, 2); - assertData(parser, "b", 3, 4); - ASSERT_FALSE(logger.hasError()); - assertEnd(parser, 6, 6); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "a", 0, 2); + assertData(parser, "b", 3, 4); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 6, 6); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorNoFieldEndNested) { - const char *testString = "\\test{\\test2{}}}"; - // 012345 6789012345 - // 0 1 - CharReader charReader(testString); + const char *testString = "\\test{\\test2{}}}"; + // 012345 6789012345 + // 0 1 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(parser, "test", 0, 5); - assertFieldStart(parser, false, 5, 6); - assertCommand(parser, "test2", 6, 12); - assertFieldStart(parser, false, 12, 13); - assertFieldEnd(parser, 13, 14); - assertFieldEnd(parser, 14, 15); - ASSERT_FALSE(logger.hasError()); - assertEnd(parser, 16, 16); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertFieldEnd(parser, 13, 14); + assertFieldEnd(parser, 14, 15); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 16, 16); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorNoFieldEndNestedData) { - const char *testString = "\\test{\\test2{}}a}"; - // 012345 67890123456 - // 0 1 - CharReader charReader(testString); + const char *testString = "\\test{\\test2{}}a}"; + // 012345 67890123456 + // 0 1 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(parser, "test", 0, 5); - assertFieldStart(parser, false, 5, 6); - assertCommand(parser, "test2", 6, 12); - assertFieldStart(parser, false, 12, 13); - assertFieldEnd(parser, 13, 14); - assertFieldEnd(parser, 14, 15); - assertData(parser, "a", 15, 16); - ASSERT_FALSE(logger.hasError()); - assertEnd(parser, 17, 17); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommand(parser, "test", 0, 5); + assertFieldStart(parser, false, 5, 6); + assertCommand(parser, "test2", 6, 12); + assertFieldStart(parser, false, 12, 13); + assertFieldEnd(parser, 13, 14); + assertFieldEnd(parser, 14, 15); + assertData(parser, "a", 15, 16); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser, 17, 17); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, beginEnd) { - const char *testString = "\\begin{book}\\end{book}"; - // 012345678901 2345678901 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}\\end{book}"; + // 012345678901 2345678901 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertRangeEnd(parser, 17, 21); - assertEnd(parser, 22, 22); + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertRangeEnd(parser, 17, 21); + assertEnd(parser, 22, 22); } TEST(OsmlStreamParser, beginEndWithName) { - const char *testString = "\\begin{book#a}\\end{book}"; - // 01234567890123 4567890123 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book#a}\\end{book}"; + // 01234567890123 4567890123 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "book", true, {{"name", "a"}}, 7, 11); - assertRangeEnd(parser, 19, 23); - assertEnd(parser, 24, 24); + assertCommandStart(parser, "book", true, {{"name", "a"}}, 7, 11); + assertRangeEnd(parser, 19, 23); + assertEnd(parser, 24, 24); } TEST(OsmlStreamParser, beginEndWithNameAndArgs) { - const char *testString = "\\begin{book#a}[a=1,b=2,c=\"test\"]\\end{book}"; - // 0123456789012345678901234 56789 01 2345678901 - // 0 1 2 3 4 - CharReader charReader(testString); + const char *testString = "\\begin{book#a}[a=1,b=2,c=\"test\"]\\end{book}"; + // 0123456789012345678901234 56789 01 2345678901 + // 0 1 2 3 4 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "book", true, - {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); - assertRangeEnd(parser, 37, 41); - assertEnd(parser, 42, 42); + assertCommandStart(parser, "book", true, + {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, + 11); + assertRangeEnd(parser, 37, 41); + assertEnd(parser, 42, 42); } TEST(OsmlStreamParser, beginEndWithNameAndArgsMultipleFields) { - const char *testString = - "\\begin{book#a}[a=1,b=2,c=\"test\"]{a \\test}{b \\test{}}\\end{book}"; - // 0123456789012345678901234 56789 01234 567890123 45678901 2345678901 - // 0 1 2 3 4 5 6 - CharReader charReader(testString); - - OsmlStreamParser parser(charReader, logger); - - assertCommandStart(parser, "book", true, - {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, 11); - assertFieldStart(parser, false, 32, 33); - assertData(parser, "a", 33, 34); - assertCommand(parser, "test", 35, 40); - assertFieldEnd(parser, 40, 41); - assertFieldStart(parser, false, 41, 42); - assertData(parser, "b", 42, 43); - assertCommand(parser, "test", 44, 49); - assertFieldStart(parser, false, 49, 50); - assertFieldEnd(parser, 50, 51); - assertFieldEnd(parser, 51, 52); - assertRangeEnd(parser, 57, 61); - assertEnd(parser, 62, 62); + const char *testString = + "\\begin{book#a}[a=1,b=2,c=\"test\"]{a \\test}{b \\test{}}\\end{book}"; + // 0123456789012345678901234 56789 01234 567890123 45678901 2345678901 + // 0 1 2 3 4 5 6 + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + assertCommandStart(parser, "book", true, + {{"name", "a"}, {"a", 1}, {"b", 2}, {"c", "test"}}, 7, + 11); + assertFieldStart(parser, false, 32, 33); + assertData(parser, "a", 33, 34); + assertCommand(parser, "test", 35, 40); + assertFieldEnd(parser, 40, 41); + assertFieldStart(parser, false, 41, 42); + assertData(parser, "b", 42, 43); + assertCommand(parser, "test", 44, 49); + assertFieldStart(parser, false, 49, 50); + assertFieldEnd(parser, 50, 51); + assertFieldEnd(parser, 51, 52); + assertRangeEnd(parser, 57, 61); + assertEnd(parser, 62, 62); } TEST(OsmlStreamParser, beginEndWithData) { - const char *testString = "\\begin{book}a\\end{book}"; - // 0123456789012 3456789012 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}a\\end{book}"; + // 0123456789012 3456789012 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertData(parser, "a", 12, 13); - assertRangeEnd(parser, 18, 22); - assertEnd(parser, 23, 23); + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertRangeEnd(parser, 18, 22); + assertEnd(parser, 23, 23); } TEST(OsmlStreamParser, beginEndNested) { - const char *testString = - "\\begin{a}{b} c \\begin{d}{e}{f} \\g{h} \\end{d}\\end{a}"; - // 012345678901234 5678901234567890 123456 7890123 4567890 - // 0 1 2 3 4 5 - CharReader charReader(testString); + const char *testString = + "\\begin{a}{b} c \\begin{d}{e}{f} \\g{h} \\end{d}\\end{a}"; + // 012345678901234 5678901234567890 123456 7890123 4567890 + // 0 1 2 3 4 5 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); - assertFieldStart(parser, false, 9, 10); - assertData(parser, "b", 10, 11); - assertFieldEnd(parser, 11, 12); + assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); + assertFieldStart(parser, false, 9, 10); + assertData(parser, "b", 10, 11); + assertFieldEnd(parser, 11, 12); - assertData(parser, "c", 13, 14); + assertData(parser, "c", 13, 14); - assertCommandStart(parser, "d", true, Variant::mapType{}, 22, 23); - assertFieldStart(parser, false, 24, 25); - assertData(parser, "e", 25, 26); - assertFieldEnd(parser, 26, 27); - assertFieldStart(parser, false, 27, 28); - assertData(parser, "f", 28, 29); - assertFieldEnd(parser, 29, 30); + assertCommandStart(parser, "d", true, Variant::mapType{}, 22, 23); + assertFieldStart(parser, false, 24, 25); + assertData(parser, "e", 25, 26); + assertFieldEnd(parser, 26, 27); + assertFieldStart(parser, false, 27, 28); + assertData(parser, "f", 28, 29); + assertFieldEnd(parser, 29, 30); - assertEmptyData(parser); - assertCommand(parser, "g", 31, 33); - assertFieldStart(parser, false, 33, 34); - assertData(parser, "h", 34, 35); - assertFieldEnd(parser, 35, 36); - assertEmptyData(parser); - assertRangeEnd(parser, 42, 43); - assertRangeEnd(parser, 49, 50); - assertEnd(parser, 51, 51); + assertEmptyData(parser); + assertCommand(parser, "g", 31, 33); + assertFieldStart(parser, false, 33, 34); + assertData(parser, "h", 34, 35); + assertFieldEnd(parser, 35, 36); + assertEmptyData(parser); + assertRangeEnd(parser, 42, 43); + assertRangeEnd(parser, 49, 50); + assertEnd(parser, 51, 51); } TEST(OsmlStreamParser, beginEndWithCommand) { - const char *testString = "\\begin{book}\\a{test}\\end{book}"; - // 012345678901 23456789 0123456789 - // 0 1 2 - CharReader charReader(testString); + const char *testString = "\\begin{book}\\a{test}\\end{book}"; + // 012345678901 23456789 0123456789 + // 0 1 2 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertCommand(parser, "a", 12, 14); - assertFieldStart(parser, false, 14, 15); - assertData(parser, "test", 15, 19); - assertFieldEnd(parser, 19, 20); - assertRangeEnd(parser, 25, 29); - assertEnd(parser, 30, 30); + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertCommand(parser, "a", 12, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "test", 15, 19); + assertFieldEnd(parser, 19, 20); + assertRangeEnd(parser, 25, 29); + assertEnd(parser, 30, 30); } TEST(OsmlStreamParser, beginEndNestedFields) { - const char *testString = "\\begin{book}a{{b{c}}}\\end{book}"; - // 012345678901234567890 1234567890 - // 0 1 2 3 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); - logger.reset(); - - assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertData(parser, "a", 12, 13); - assertFieldStart(parser, false, 13, 14); - assertFieldStart(parser, false, 14, 15); - assertData(parser, "b", 15, 16); - assertFieldStart(parser, false, 16, 17); - assertData(parser, "c", 17, 18); - assertFieldEnd(parser, 18, 19); - assertFieldEnd(parser, 19, 20); - assertFieldEnd(parser, 20, 21); - assertRangeEnd(parser, 26, 30); - assertEnd(parser, 31, 31); + const char *testString = "\\begin{book}a{{b{c}}}\\end{book}"; + // 012345678901234567890 1234567890 + // 0 1 2 3 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + logger.reset(); + + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertFieldStart(parser, false, 13, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "b", 15, 16); + assertFieldStart(parser, false, 16, 17); + assertData(parser, "c", 17, 18); + assertFieldEnd(parser, 18, 19); + assertFieldEnd(parser, 19, 20); + assertFieldEnd(parser, 20, 21); + assertRangeEnd(parser, 26, 30); + assertEnd(parser, 31, 31); } TEST(OsmlStreamParser, errorBeginEndUnbalancedNestedFields) { - const char *testString = "\\begin{book}a{{b{c}}\\end{book}"; - // 012345678901234567890 123456789 - // 0 1 2 - CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); - logger.reset(); - - assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); - assertData(parser, "a", 12, 13); - assertFieldStart(parser, false, 13, 14); - assertFieldStart(parser, false, 14, 15); - assertData(parser, "b", 15, 16); - assertFieldStart(parser, false, 16, 17); - assertData(parser, "c", 17, 18); - assertFieldEnd(parser, 18, 19); - assertFieldEnd(parser, 19, 20); - ASSERT_THROW(assertRangeEnd(parser, 25, 29), LoggableException); + const char *testString = "\\begin{book}a{{b{c}}\\end{book}"; + // 012345678901234567890 123456789 + // 0 1 2 + CharReader charReader(testString); + OsmlStreamParser parser(charReader, logger); + logger.reset(); + + assertCommandStart(parser, "book", true, Variant::mapType{}, 7, 11); + assertData(parser, "a", 12, 13); + assertFieldStart(parser, false, 13, 14); + assertFieldStart(parser, false, 14, 15); + assertData(parser, "b", 15, 16); + assertFieldStart(parser, false, 16, 17); + assertData(parser, "c", 17, 18); + assertFieldEnd(parser, 18, 19); + assertFieldEnd(parser, 19, 20); + ASSERT_THROW(assertRangeEnd(parser, 25, 29), LoggableException); } TEST(OsmlStreamParser, errorBeginEndUnbalancedFields) @@ -962,446 +960,480 @@ TEST(OsmlStreamParser, errorBeginEndUnbalancedFields) TEST(OsmlStreamParser, errorBeginNoBraceOpen) { - const char *testString = "\\begin a"; - // 01234567 - CharReader charReader(testString); + const char *testString = "\\begin a"; + // 01234567 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertData(parser, "a", 7, 8); - ASSERT_TRUE(logger.hasError()); - assertEnd(parser, 8, 8); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertData(parser, "a", 7, 8); + ASSERT_TRUE(logger.hasError()); + assertEnd(parser, 8, 8); } TEST(OsmlStreamParser, errorBeginNoIdentifier) { - const char *testString = "\\begin{!"; - CharReader charReader(testString); + const char *testString = "\\begin{!"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoBraceClose) { - const char *testString = "\\begin{a"; - CharReader charReader(testString); + const char *testString = "\\begin{a"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginNoName) { - const char *testString = "\\begin{a#}"; - CharReader charReader(testString); + const char *testString = "\\begin{a#}"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommandStart(parser, "a", true); - ASSERT_TRUE(logger.hasError()); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertEnd(parser); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommandStart(parser, "a", true); + ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertEnd(parser); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBraceOpen) { - const char *testString = "\\end a"; - // 012345 - CharReader charReader(testString); + const char *testString = "\\end a"; + // 012345 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertData(parser, "a", 5, 6); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertData(parser, "a", 5, 6); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoIdentifier) { - const char *testString = "\\end{!"; - CharReader charReader(testString); + const char *testString = "\\end{!"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBraceClose) { - const char *testString = "\\end{a"; - CharReader charReader(testString); + const char *testString = "\\end{a"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorEndNoBegin) { - const char *testString = "\\end{a}"; - CharReader charReader(testString); + const char *testString = "\\end{a}"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, errorBeginEndMismatch) { - const char *testString = "\\begin{a} \\begin{b} test \\end{a}"; - // 0123456789 012345678901234 5678901 - // 0 1 2 3 - CharReader charReader(testString); + const char *testString = "\\begin{a} \\begin{b} test \\end{a}"; + // 0123456789 012345678901234 5678901 + // 0 1 2 3 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); - assertEmptyData(parser); - assertCommandStart(parser, "b", true, Variant::mapType{}, 17, 18); - assertData(parser, "test", 20, 24); - ASSERT_FALSE(logger.hasError()); - ASSERT_THROW(parser.parse(), LoggableException); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + assertCommandStart(parser, "a", true, Variant::mapType{}, 7, 8); + assertEmptyData(parser); + assertCommandStart(parser, "b", true, Variant::mapType{}, 17, 18); + assertData(parser, "test", 20, 24); + ASSERT_FALSE(logger.hasError()); + ASSERT_THROW(parser.parse(), LoggableException); + ASSERT_TRUE(logger.hasError()); } TEST(OsmlStreamParser, commandWithNSSep) { - const char *testString = "\\test1:test2"; - // 012345678901 - CharReader charReader(testString); + const char *testString = "\\test1:test2"; + // 012345678901 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "test1:test2", 0, 12); - assertEnd(parser, 12, 12); + assertCommand(parser, "test1:test2", 0, 12); + assertEnd(parser, 12, 12); } TEST(OsmlStreamParser, beginEndWithNSSep) { - const char *testString = "\\begin{test1:test2}\\end{test1:test2}"; - // 0123456789012345678 90123456789012345 - // 0 1 2 3 - CharReader charReader(testString); + const char *testString = "\\begin{test1:test2}\\end{test1:test2}"; + // 0123456789012345678 90123456789012345 + // 0 1 2 3 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommandStart(parser, "test1:test2", true, Variant::mapType{}, 7, 18); - assertRangeEnd(parser, 24, 35); - assertEnd(parser, 36, 36); + assertCommandStart(parser, "test1:test2", true, Variant::mapType{}, 7, 18); + assertRangeEnd(parser, 24, 35); + assertEnd(parser, 36, 36); } TEST(OsmlStreamParser, errorBeginNSSep) { - const char *testString = "\\begin:test{blub}\\end{blub}"; - CharReader charReader(testString); + const char *testString = "\\begin:test{blub}\\end{blub}"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommandStart(parser, "blub", true, Variant::mapType{}); - ASSERT_TRUE(logger.hasError()); - assertRangeEnd(parser); - assertEnd(parser); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommandStart(parser, "blub", true, Variant::mapType{}); + ASSERT_TRUE(logger.hasError()); + assertRangeEnd(parser); + assertEnd(parser); } TEST(OsmlStreamParser, errorEndNSSep) { - const char *testString = "\\begin{blub}\\end:test{blub}"; - CharReader charReader(testString); + const char *testString = "\\begin{blub}\\end:test{blub}"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommandStart(parser, "blub", true, Variant::mapType{}); - ASSERT_FALSE(logger.hasError()); - assertRangeEnd(parser); - ASSERT_TRUE(logger.hasError()); - assertEnd(parser); + logger.reset(); + assertCommandStart(parser, "blub", true, Variant::mapType{}); + ASSERT_FALSE(logger.hasError()); + assertRangeEnd(parser); + ASSERT_TRUE(logger.hasError()); + assertEnd(parser); } TEST(OsmlStreamParser, errorEmptyNs) { - const char *testString = "\\test:"; - CharReader charReader(testString); + const char *testString = "\\test:"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(parser, "test"); - ASSERT_TRUE(logger.hasError()); - assertData(parser, ":"); - assertEnd(parser); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(parser, "test"); + ASSERT_TRUE(logger.hasError()); + assertData(parser, ":"); + assertEnd(parser); } TEST(OsmlStreamParser, errorRepeatedNs) { - const char *testString = "\\test::"; - CharReader charReader(testString); + const char *testString = "\\test::"; + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommand(parser, "test"); - ASSERT_TRUE(logger.hasError()); - assertData(parser, "::"); - assertEnd(parser); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommand(parser, "test"); + ASSERT_TRUE(logger.hasError()); + assertData(parser, "::"); + assertEnd(parser); } TEST(OsmlStreamParser, explicitDefaultField) { - const char *testString = "\\a{!b}c"; - // 01234567 - CharReader charReader(testString); + const char *testString = "\\a{!b}c"; + // 01234567 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "a", 0, 2); - assertFieldStart(parser, true, 2, 4); - assertData(parser, "b", 4, 5); - assertFieldEnd(parser, 5, 6); - assertData(parser, "c", 6, 7); - assertEnd(parser, 7, 7); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertData(parser, "b", 4, 5); + assertFieldEnd(parser, 5, 6); + assertData(parser, "c", 6, 7); + assertEnd(parser, 7, 7); } TEST(OsmlStreamParser, explicitDefaultFieldWithCommand) { - const char *testString = "\\a{!\\b}c"; - // 0123 4567 - CharReader charReader(testString); + const char *testString = "\\a{!\\b}c"; + // 0123 4567 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertCommand(parser, "a", 0, 2); - assertFieldStart(parser, true, 2, 4); - assertCommand(parser, "b", 4, 6); - assertFieldEnd(parser, 6, 7); - assertData(parser, "c", 7, 8); - assertEnd(parser, 8, 8); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertCommand(parser, "b", 4, 6); + assertFieldEnd(parser, 6, 7); + assertData(parser, "c", 7, 8); + assertEnd(parser, 8, 8); } TEST(OsmlStreamParser, fieldAfterExplicitDefaultField) { - const char *testString = "\\a{!\\b}{c}"; - // 0123 456789 - CharReader charReader(testString); + const char *testString = "\\a{!\\b}{c}"; + // 0123 456789 + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - assertCommand(parser, "a", 0, 2); - assertFieldStart(parser, true, 2, 4); - assertCommand(parser, "b", 4, 6); - assertFieldEnd(parser, 6, 7); - assertFieldStart(parser, false, 7, 8); - assertData(parser, "c", 8, 9); - assertFieldEnd(parser, 9, 10); - assertEnd(parser, 10, 10); + logger.reset(); + assertCommand(parser, "a", 0, 2); + assertFieldStart(parser, true, 2, 4); + assertCommand(parser, "b", 4, 6); + assertFieldEnd(parser, 6, 7); + assertFieldStart(parser, false, 7, 8); + assertData(parser, "c", 8, 9); + assertFieldEnd(parser, 9, 10); + assertEnd(parser, 10, 10); } TEST(OsmlStreamParser, annotationStart) { - const char *testString = "<\\a"; - // 0 12 + const char *testString = "<\\a"; + // 0 12 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); - assertEnd(parser, 3, 3); + assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); + assertEnd(parser, 3, 3); } TEST(OsmlStreamParser, annotationStartWithName) { - const char *testString = "<\\annotationWithName#aName"; - // 0 1234567890123456789012345 - // 0 1 2 + const char *testString = "<\\annotationWithName#aName"; + // 0 1234567890123456789012345 + // 0 1 2 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(parser, "annotationWithName", - Variant::mapType{{"name", "aName"}}, 0, 20); - assertEnd(parser, 26, 26); + assertAnnotationStart(parser, "annotationWithName", + Variant::mapType{{"name", "aName"}}, 0, 20); + assertEnd(parser, 26, 26); } TEST(OsmlStreamParser, annotationStartWithArguments) { - const char *testString = "<\\annotationWithName#aName[a=1,b=2]"; - // 0 1234567890123456789012345678901234 - // 0 1 2 3 + const char *testString = "<\\annotationWithName#aName[a=1,b=2]"; + // 0 1234567890123456789012345678901234 + // 0 1 2 3 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart( - parser, "annotationWithName", - Variant::mapType{{"name", "aName"}, {"a", 1}, {"b", 2}}, 0, 20); - assertEnd(parser, 35, 35); + assertAnnotationStart( + parser, "annotationWithName", + Variant::mapType{{"name", "aName"}, {"a", 1}, {"b", 2}}, 0, 20); + assertEnd(parser, 35, 35); } TEST(OsmlStreamParser, simpleAnnotationStartBeginEnd) { - const char *testString = "<\\begin{ab#name}[a=1,b=2] a \\end{ab}\\>"; - // 0 123456789012345678901234567 89012345 67 - // 0 1 2 3 + const char *testString = "<\\begin{ab#name}[a=1,b=2] a \\end{ab}\\>"; + // 0 123456789012345678901234567 89012345 67 + // 0 1 2 3 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart( - parser, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, - 10); - ASSERT_TRUE(parser.inRangeCommand()); - assertData(parser, "a", 26, 27); - assertRangeEnd(parser, 33, 35); - assertAnnotationEnd(parser, "", "", 36, 38); - assertEnd(parser, 38, 38); + assertAnnotationStart( + parser, "ab", Variant::mapType{{"name", "name"}, {"a", 1}, {"b", 2}}, 8, + 10); + ASSERT_TRUE(parser.inRangeCommand()); + assertData(parser, "a", 26, 27); + assertRangeEnd(parser, 33, 35); + assertAnnotationEnd(parser, "", "", 36, 38); + assertEnd(parser, 38, 38); } TEST(OsmlStreamParser, annotationEnd) { - const char *testString = "\\a>"; - // 012 + const char *testString = "\\a>"; + // 012 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(parser, "a", "", 0, 2); - assertEnd(parser, 3, 3); + assertAnnotationEnd(parser, "a", "", 0, 2); + assertEnd(parser, 3, 3); } TEST(OsmlStreamParser, annotationEndWithName) { - const char *testString = "\\a#name>"; - // 01234567 + const char *testString = "\\a#name>"; + // 01234567 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(parser, "a", "name", 0, 2); - assertEnd(parser, 8, 8); + assertAnnotationEnd(parser, "a", "name", 0, 2); + assertEnd(parser, 8, 8); } TEST(OsmlStreamParser, annotationEndWithNameAsArgs) { - const char *testString = "\\a[name=name]>"; - // 01234567890123 + const char *testString = "\\a[name=name]>"; + // 01234567890123 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationEnd(parser, "a", "name", 0, 2); - assertEnd(parser, 14, 14); + assertAnnotationEnd(parser, "a", "name", 0, 2); + assertEnd(parser, 14, 14); } TEST(OsmlStreamParser, errorAnnotationEndWithArguments) { - const char *testString = "\\a[foo=bar]>"; - // 012345678901 - // 0 1 + const char *testString = "\\a[foo=bar]>"; + // 012345678901 + // 0 1 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - logger.reset(); - ASSERT_FALSE(logger.hasError()); - assertCommandStart(parser, "a", false, Variant::mapType{{"foo", "bar"}}, 0, 2); - ASSERT_TRUE(logger.hasError()); - assertData(parser, ">", 11, 12); - assertEnd(parser, 12, 12); + logger.reset(); + ASSERT_FALSE(logger.hasError()); + assertCommandStart(parser, "a", false, Variant::mapType{{"foo", "bar"}}, 0, + 2); + ASSERT_TRUE(logger.hasError()); + assertData(parser, ">", 11, 12); + assertEnd(parser, 12, 12); } TEST(OsmlStreamParser, closingAnnotation) { - const char *testString = "<\\a>"; - // 0 123 + const char *testString = "<\\a>"; + // 0 123 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); - assertData(parser, ">", 3, 4); - assertEnd(parser, 4, 4); + assertAnnotationStart(parser, "a", Variant::mapType{}, 0, 3); + assertData(parser, ">", 3, 4); + assertEnd(parser, 4, 4); } TEST(OsmlStreamParser, annotationWithFields) { - const char *testString = "a <\\b{c}{d}{!e} f \\> g"; - // 012 345678901234567 8901 - // 0 1 2 + const char *testString = "a <\\b{c}{d}{!e} f \\> g"; + // 012 345678901234567 8901 + // 0 1 2 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertData(parser, "a", 0, 1); - assertAnnotationStart(parser, "b", Variant::mapType{}, 2, 5); - assertFieldStart(parser, false, 5, 6); - assertData(parser, "c", 6, 7); - assertFieldEnd(parser, 7, 8); - assertFieldStart(parser, false, 8, 9); - assertData(parser, "d", 9, 10); - assertFieldEnd(parser, 10, 11); - assertFieldStart(parser, true, 11, 13); - assertData(parser, "e", 13, 14); - assertFieldEnd(parser, 14, 15); - assertData(parser, "f", 16, 17); - assertAnnotationEnd(parser, "", "", 18, 20); - assertData(parser, "g", 21, 22); - assertEnd(parser, 22, 22); + assertData(parser, "a", 0, 1); + assertAnnotationStart(parser, "b", Variant::mapType{}, 2, 5); + assertFieldStart(parser, false, 5, 6); + assertData(parser, "c", 6, 7); + assertFieldEnd(parser, 7, 8); + assertFieldStart(parser, false, 8, 9); + assertData(parser, "d", 9, 10); + assertFieldEnd(parser, 10, 11); + assertFieldStart(parser, true, 11, 13); + assertData(parser, "e", 13, 14); + assertFieldEnd(parser, 14, 15); + assertData(parser, "f", 16, 17); + assertAnnotationEnd(parser, "", "", 18, 20); + assertData(parser, "g", 21, 22); + assertEnd(parser, 22, 22); } TEST(OsmlStreamParser, annotationStartEscape) { - const char *testString = "<\\%test"; - // 0 123456 - // 0 + const char *testString = "<\\%test"; + // 0 123456 + // 0 - CharReader charReader(testString); + CharReader charReader(testString); - OsmlStreamParser parser(charReader, logger); + OsmlStreamParser parser(charReader, logger); - assertData(parser, "<%test", 0, 7); - assertEnd(parser, 7, 7); + assertData(parser, "<%test", 0, 7); + assertEnd(parser, 7, 7); } +TEST(OsmlStreamParser, userDefinedTokens) +{ + const char *testString = "<>, the *old man* said."; + // 0123456789012345678901234567890123456789 + // 0 1 2 3 + + CharReader charReader(testString); + + OsmlStreamParser parser(charReader, logger); + + TokenId tSpeechStart = parser.registerToken("<<"); + TokenId tSpeechEnd = parser.registerToken(">>"); + TokenId tStar = parser.registerToken("*"); + + ASSERT_TRUE(tSpeechStart != Tokens::Empty); + ASSERT_TRUE(tSpeechEnd != Tokens::Empty); + ASSERT_TRUE(tStar != Tokens::Empty); + + TokenSet tokens{tSpeechStart, tSpeechEnd, tStar}; + + ASSERT_EQ(OsmlStreamParser::State::DATA, parser.parse()); + TokenizedDataReader reader = parser.getData().reader(); + + assertToken(reader, tSpeechStart, "<<", tokens, WhitespaceMode::PRESERVE, 0, 2); + assertText(reader, "My dear fellows", tokens, WhitespaceMode::PRESERVE, 2, 17); + assertToken(reader, tSpeechEnd, ">>", tokens, WhitespaceMode::PRESERVE, 17, 19); + assertText(reader, ", the ", tokens, WhitespaceMode::PRESERVE, 19, 25); + assertToken(reader, tStar, "*", tokens, WhitespaceMode::PRESERVE, 25, 26); + assertText(reader, "old man", tokens, WhitespaceMode::PRESERVE, 26, 33); + assertToken(reader, tStar, "*", tokens, WhitespaceMode::PRESERVE, 33, 34); + assertText(reader, " said.", tokens, WhitespaceMode::PRESERVE, 34, 40); + assertEnd(reader); +} } -- cgit v1.2.3 From e2fd79ac8c85ac6191f6ed895fa5cdff091f7551 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 1 Mar 2015 16:28:35 +0100 Subject: Implemented TokenRegistry class and corresponding simple test case --- CMakeLists.txt | 2 + src/core/parser/stack/TokenRegistry.cpp | 72 +++++++++++++++++++++++ src/core/parser/stack/TokenRegistry.hpp | 87 ++++++++++++++++++++++++++++ test/core/parser/stack/TokenRegistryTest.cpp | 78 +++++++++++++++++++++++++ 4 files changed, 239 insertions(+) create mode 100644 src/core/parser/stack/TokenRegistry.cpp create mode 100644 src/core/parser/stack/TokenRegistry.hpp create mode 100644 test/core/parser/stack/TokenRegistryTest.cpp (limited to 'test') diff --git a/CMakeLists.txt b/CMakeLists.txt index 4e2d7f7..6e021fd 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -188,6 +188,7 @@ ADD_LIBRARY(ousia_core # src/core/parser/stack/ImportIncludeHandler src/core/parser/stack/State # src/core/parser/stack/Stack + src/core/parser/stack/TokenRegistry # src/core/parser/stack/TypesystemHandler src/core/parser/utils/SourceOffsetVector src/core/parser/utils/TokenizedData @@ -325,6 +326,7 @@ IF(TEST) test/core/parser/ParserScopeTest # test/core/parser/stack/StackTest test/core/parser/stack/StateTest + test/core/parser/stack/TokenRegistryTest test/core/parser/utils/SourceOffsetVectorTest test/core/parser/utils/TokenizedDataTest test/core/parser/utils/TokenizerTest diff --git a/src/core/parser/stack/TokenRegistry.cpp b/src/core/parser/stack/TokenRegistry.cpp new file mode 100644 index 0000000..21ae109 --- /dev/null +++ b/src/core/parser/stack/TokenRegistry.cpp @@ -0,0 +1,72 @@ +/* + Ousía + Copyright (C) 2014, 2015 Benjamin Paaßen, Andreas Stöckel + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ + +#include "TokenRegistry.hpp" + +namespace ousia { +namespace parser_stack { + +TokenId TokenRegistry::registerToken(const std::string &token) +{ + // Check whether the given token is already registered + auto it = tokens.find(token); + if (it != tokens.end()) { + // Increment the reference count + size_t &refCount = it->second.second; + refCount++; + + // Return the token id + return it->second.first; + } + + // Register the token in the parser + TokenId id = parser.registerToken(token); + tokens[token] = std::pair(id, 1); + tokenIds[id] = token; + return id; +} + +void TokenRegistry::unregisterToken(TokenId id) +{ + // Lookup the token corresponding to the given token id + auto tokenIt = tokenIds.find(id); + if (tokenIt != tokenIds.end()) { + const std::string &token = tokenIt->second; + // Lookup the reference count for the corresponding token + auto idIt = tokens.find(token); + if (idIt != tokens.end()) { + // Decrement the reference count, abort if the refCount is larger + // than zero + size_t &refCount = idIt->second.second; + refCount--; + if (refCount > 0) { + return; + } + + // Unregister the token from the parser + parser.unregisterToken(id); + + // Unregister the token from the internal tokens map + tokens.erase(token); + } + // Unregister the token from the internal id map + tokenIds.erase(id); + } +} +} +} diff --git a/src/core/parser/stack/TokenRegistry.hpp b/src/core/parser/stack/TokenRegistry.hpp new file mode 100644 index 0000000..21c36b5 --- /dev/null +++ b/src/core/parser/stack/TokenRegistry.hpp @@ -0,0 +1,87 @@ +/* + Ousía + Copyright (C) 2014, 2015 Benjamin Paaßen, Andreas Stöckel + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ + +/** + * @file TokenRegistry.hpp + * + * Contains the TokenRegistry class used for registering all possible tokens + * during the parsing process. + * + * @author Andreas Stöckel (astoecke@techfak.uni-bielefeld.de) + */ + +#ifndef _OUSIA_PARSER_STACK_TOKEN_REGISTRY_HPP_ +#define _OUSIA_PARSER_STACK_TOKEN_REGISTRY_HPP_ + +#include +#include + +#include "Callbacks.hpp" + +namespace ousia { +namespace parser_stack { + +/** + * The TokenRegistry class is used for registering all possible tokens during + * the Parsing process. The TokenRegistry class acts as an adapter between the + * parser which allocates TokenId for each unique token and the Handler classes + * which may register tokens multiple times and expect the same TokenId to be + * returned for the same token. + */ +class TokenRegistry : public ParserCallbacks { +private: + /** + * Reference at the ParserCallback instance the tokens are relayed to. + */ + ParserCallbacks &parser; + + /** + * Store containing all TokenId instances for all registered tokens. The map + * maps from the token strings to the corresponding TokenId and a reference + * count. + */ + std::unordered_map> tokens; + + /** + * Reverse map containing the string corresponding to a TokenId. + */ + std::unordered_map tokenIds; + +public: + /** + * Constructor of the TokenRegistry class. + * + * @param parser is the underlying parser implementing the ParserCallbacks + * interface to which all calls are relayed. + */ + TokenRegistry(ParserCallbacks &parser) : parser(parser) {} + + /* No copy construction */ + TokenRegistry(const TokenRegistry &) = delete; + + /* No assignment */ + TokenRegistry &operator=(const TokenRegistry &) = delete; + + TokenId registerToken(const std::string &token) override; + void unregisterToken(TokenId id) override; +}; +} +} + +#endif /* _OUSIA_PARSER_STACK_TOKEN_REGISTRY_HPP_ */ + diff --git a/test/core/parser/stack/TokenRegistryTest.cpp b/test/core/parser/stack/TokenRegistryTest.cpp new file mode 100644 index 0000000..390851e --- /dev/null +++ b/test/core/parser/stack/TokenRegistryTest.cpp @@ -0,0 +1,78 @@ +/* + Ousía + Copyright (C) 2014, 2015 Benjamin Paaßen, Andreas Stöckel + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ + +#include + +#include + +namespace ousia { +namespace parser_stack { + +class ParserCallbacksProxy : public ParserCallbacks { +public: + size_t registerTokenCount = 0; + size_t unregisterTokenCount = 0; + + TokenId registerToken(const std::string &token) override + { + registerTokenCount++; + return registerTokenCount; + } + + void unregisterToken(TokenId id) override { unregisterTokenCount++; } +}; + +TEST(TokenRegistry, simple) +{ + ParserCallbacksProxy parser; + TokenRegistry registry(parser); + + ASSERT_EQ(0U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + ASSERT_EQ(1U, registry.registerToken("test")); + ASSERT_EQ(1U, registry.registerToken("test")); + ASSERT_EQ(2U, registry.registerToken("test2")); + ASSERT_EQ(2U, registry.registerToken("test2")); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(2); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(2); + ASSERT_EQ(2U, parser.registerTokenCount); + ASSERT_EQ(2U, parser.unregisterTokenCount); +} + +} +} + -- cgit v1.2.3 From 3bdc30e0798d6b356782da430e93b72b4303e963 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Mon, 2 Mar 2015 00:32:34 +0100 Subject: Adapted OsxmlParser to new Stack interface, enabled Osxml code in CMakeLists again --- CMakeLists.txt | 22 ++++++++-------- src/formats/osxml/OsxmlEventParser.cpp | 6 ++--- src/formats/osxml/OsxmlEventParser.hpp | 11 +++----- src/formats/osxml/OsxmlParser.cpp | 13 ++++----- test/formats/osxml/OsxmlEventParserTest.cpp | 41 ++++++++++++++++------------- 5 files changed, 48 insertions(+), 45 deletions(-) (limited to 'test') diff --git a/CMakeLists.txt b/CMakeLists.txt index 6e021fd..2a09b54 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -225,7 +225,7 @@ TARGET_LINK_LIBRARIES(ousia_osml ADD_LIBRARY(ousia_osxml src/formats/osxml/OsxmlAttributeLocator src/formats/osxml/OsxmlEventParser -# src/formats/osxml/OsxmlParser + src/formats/osxml/OsxmlParser ) TARGET_LINK_LIBRARIES(ousia_osxml @@ -397,17 +397,17 @@ IF(TEST) ousia_filesystem ) -# ADD_EXECUTABLE(ousia_test_osxml -# test/formats/osxml/OsxmlEventParserTest + ADD_EXECUTABLE(ousia_test_osxml + test/formats/osxml/OsxmlEventParserTest # test/formats/osxml/OsxmlParserTest -# ) + ) -# TARGET_LINK_LIBRARIES(ousia_test_osxml -# ${GTEST_LIBRARIES} -# ousia_core -# ousia_osxml -# ousia_filesystem -# ) + TARGET_LINK_LIBRARIES(ousia_test_osxml + ${GTEST_LIBRARIES} + ousia_core + ousia_osxml + ousia_filesystem + ) ADD_EXECUTABLE(ousia_test_xml test/plugins/xml/XmlOutputTest @@ -426,7 +426,7 @@ IF(TEST) ADD_TEST(ousia_test_html ousia_test_html) # ADD_TEST(ousia_test_mozjs ousia_test_mozjs) ADD_TEST(ousia_test_osml ousia_test_osml) -# ADD_TEST(ousia_test_osxml ousia_test_osxml) + ADD_TEST(ousia_test_osxml ousia_test_osxml) ADD_TEST(ousia_test_xml ousia_test_xml) ENDIF() diff --git a/src/formats/osxml/OsxmlEventParser.cpp b/src/formats/osxml/OsxmlEventParser.cpp index 855f80d..83c16f0 100644 --- a/src/formats/osxml/OsxmlEventParser.cpp +++ b/src/formats/osxml/OsxmlEventParser.cpp @@ -323,7 +323,7 @@ static void xmlStartElementHandler(void *ref, const XML_Char *name, // Just issue a "commandStart" event in any other case Variant nameVar = Variant::fromString(nameStr); nameVar.setLocation(nameLoc); - parser->getEvents().command(nameVar, args); + parser->getEvents().commandStart(nameVar, args); } } @@ -358,8 +358,8 @@ static void xmlEndElementHandler(void *ref, const XML_Char *name) return; } - // Issue the "fieldEnd" event - parser->getEvents().fieldEnd(); + // Issue the "rangeEnd" event + parser->getEvents().rangeEnd(); } static void xmlCharacterDataHandler(void *ref, const XML_Char *s, int len) diff --git a/src/formats/osxml/OsxmlEventParser.hpp b/src/formats/osxml/OsxmlEventParser.hpp index e3fd5d4..7a8c96d 100644 --- a/src/formats/osxml/OsxmlEventParser.hpp +++ b/src/formats/osxml/OsxmlEventParser.hpp @@ -59,7 +59,8 @@ public: * @param args is a map containing the arguments that were given to the * command. */ - virtual void command(const Variant &name, const Variant::mapType &args) = 0; + virtual void commandStart(const Variant &name, + const Variant::mapType &args) = 0; /** * Called whenever an annotation starts. Note that this implicitly always @@ -88,13 +89,9 @@ public: const Variant &elementName) = 0; /** - * Called whenever the default field which was implicitly started by - * commandStart or annotationStart ends. Note that this does not end the - * range of an annotation, but the default field of the annotation. To - * signal the end of the annotation this, the annotationEnd method will be - * invoked. + * Called whenever the command or annotation tags end. */ - virtual void fieldEnd() = 0; + virtual void rangeEnd() = 0; /** * Called whenever string data is found. diff --git a/src/formats/osxml/OsxmlParser.cpp b/src/formats/osxml/OsxmlParser.cpp index c216855..924d11b 100644 --- a/src/formats/osxml/OsxmlParser.cpp +++ b/src/formats/osxml/OsxmlParser.cpp @@ -16,6 +16,8 @@ along with this program. If not, see . */ +#include +#include #include #include #include @@ -63,17 +65,16 @@ public: */ void parse() { parser.parse(); } - void command(const Variant &name, const Variant::mapType &args) override + void commandStart(const Variant &name, + const Variant::mapType &args) override { - stack.command(name, args); - stack.fieldStart(true); + stack.commandStart(name, args, true); } void annotationStart(const Variant &name, const Variant::mapType &args) override { - stack.annotationStart(name, args); - stack.fieldStart(true); + stack.annotationStart(name, args, true); } void annotationEnd(const Variant &className, @@ -82,7 +83,7 @@ public: stack.annotationEnd(className, elementName); } - void fieldEnd() override { stack.fieldEnd(); } + void rangeEnd() override { stack.rangeEnd(); } void data(const Variant &data) override { stack.data(data); } }; diff --git a/test/formats/osxml/OsxmlEventParserTest.cpp b/test/formats/osxml/OsxmlEventParserTest.cpp index 6942166..b24a43d 100644 --- a/test/formats/osxml/OsxmlEventParserTest.cpp +++ b/test/formats/osxml/OsxmlEventParserTest.cpp @@ -32,10 +32,10 @@ static TerminalLogger logger(std::cerr, true); namespace { enum class OsxmlEvent { - COMMAND, + COMMAND_START, ANNOTATION_START, ANNOTATION_END, - FIELD_END, + RANGE_END, DATA }; @@ -43,9 +43,10 @@ class TestOsxmlEventListener : public OsxmlEvents { public: std::vector> events; - void command(const Variant &name, const Variant::mapType &args) override + void commandStart(const Variant &name, + const Variant::mapType &args) override { - events.emplace_back(OsxmlEvent::COMMAND, + events.emplace_back(OsxmlEvent::COMMAND_START, Variant::arrayType{name, args}); } @@ -63,9 +64,9 @@ public: Variant::arrayType{className, elementName}); } - void fieldEnd() override + void rangeEnd() override { - events.emplace_back(OsxmlEvent::FIELD_END, Variant::arrayType{}); + events.emplace_back(OsxmlEvent::RANGE_END, Variant::arrayType{}); } void data(const Variant &data) override @@ -92,11 +93,11 @@ TEST(OsxmlEventParser, simpleCommandWithArgs) // 0 1 2 3 std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, + {OsxmlEvent::COMMAND_START, Variant::arrayType{ "a", Variant::mapType{ {"name", "test"}, {"a", 1}, {"b", 2}, {"c", "blub"}}}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; + {OsxmlEvent::RANGE_END, Variant::arrayType{}}}; auto events = parseXml(testString); ASSERT_EQ(expectedEvents, events); @@ -132,10 +133,12 @@ TEST(OsxmlEventParser, magicTopLevelTag) const char *testString = ""; std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, Variant::arrayType{{"a", Variant::mapType{}}}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}, - {OsxmlEvent::COMMAND, Variant::arrayType{{"b", Variant::mapType{}}}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; + {OsxmlEvent::COMMAND_START, + Variant::arrayType{{"a", Variant::mapType{}}}}, + {OsxmlEvent::RANGE_END, Variant::arrayType{}}, + {OsxmlEvent::COMMAND_START, + Variant::arrayType{{"b", Variant::mapType{}}}}, + {OsxmlEvent::RANGE_END, Variant::arrayType{}}}; auto events = parseXml(testString); ASSERT_EQ(expectedEvents, events); @@ -146,11 +149,12 @@ TEST(OsxmlEventParser, magicTopLevelTagInside) const char *testString = ""; std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, Variant::arrayType{{"a", Variant::mapType{}}}}, - {OsxmlEvent::COMMAND, + {OsxmlEvent::COMMAND_START, + Variant::arrayType{{"a", Variant::mapType{}}}}, + {OsxmlEvent::COMMAND_START, Variant::arrayType{{"ousia", Variant::mapType{}}}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; + {OsxmlEvent::RANGE_END, Variant::arrayType{}}, + {OsxmlEvent::RANGE_END, Variant::arrayType{}}}; auto events = parseXml(testString); ASSERT_EQ(expectedEvents, events); @@ -163,9 +167,10 @@ TEST(OsxmlEventParser, commandWithData) // 0 1 2 std::vector> expectedEvents{ - {OsxmlEvent::COMMAND, Variant::arrayType{"a", Variant::mapType{}}}, + {OsxmlEvent::COMMAND_START, + Variant::arrayType{"a", Variant::mapType{}}}, {OsxmlEvent::DATA, Variant::arrayType{" hello \n world "}}, - {OsxmlEvent::FIELD_END, Variant::arrayType{}}}; + {OsxmlEvent::RANGE_END, Variant::arrayType{}}}; auto events = parseXml(testString); ASSERT_EQ(expectedEvents, events); -- cgit v1.2.3 From 596fdab71b8bd116e20e33647d68f1d7a567696e Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Mon, 2 Mar 2015 00:34:15 +0100 Subject: Wrote isUserDefinedToken function which checks whether a token is a valid user defined token and added unit tests --- src/core/common/Utils.cpp | 24 ++++++++++++++++++++++++ src/core/common/Utils.hpp | 19 +++++++++++++++++++ test/core/common/UtilsTest.cpp | 31 ++++++++++++++++++++++++++++++- 3 files changed, 73 insertions(+), 1 deletion(-) (limited to 'test') diff --git a/src/core/common/Utils.cpp b/src/core/common/Utils.cpp index 85d2c28..219b437 100644 --- a/src/core/common/Utils.cpp +++ b/src/core/common/Utils.cpp @@ -118,5 +118,29 @@ bool Utils::endsWith(const std::string &s, const std::string &suffix) return suffix.size() <= s.size() && s.substr(s.size() - suffix.size(), suffix.size()) == suffix; } + +bool Utils::isUserDefinedToken(const std::string &token) +{ + // Make sure the token meets is neither empty, nor starts or ends with an + // alphanumeric character + const size_t len = token.size(); + if (len == 0 || isAlphanumeric(token[0]) || isAlphanumeric(token[len - 1])) { + return false; + } + + // Make sure the token is not any special OSML token + if (token == "\\" || token == "%" || token == "%{" || token == "}%" || + token == "{!" || token == "<\\" || token == "\\>") { + return false; + } + + // Make sure the token contains other characters but { and } + for (char c: token) { + if (c != '{' && c != '}') { + return true; + } + } + return false; +} } diff --git a/src/core/common/Utils.hpp b/src/core/common/Utils.hpp index 82a8f8c..25a4de5 100644 --- a/src/core/common/Utils.hpp +++ b/src/core/common/Utils.hpp @@ -102,6 +102,25 @@ public: */ static bool isNamespacedIdentifier(const std::string &name); + /** + * Returns true if the given characters form a valid user-defined token. + * This function returns true under the following circumstances: + *
    + *
  • The given token is not empty
    • + *
  • The given token starts and ends with a non-alphanumeric character + *
    • + *
  • The token is none of the following character sequences (which are + * special in OSML): + *
      + *
    • '{', '}' or any combined repetition of these characters
      • + *
    • '\', '{!', '<\', '\>'
      • + *
    • '%', '%{', '}%'
      • + *
    + *
    • + *
+ */ + static bool isUserDefinedToken(const std::string &token); + /** * Returns true if the given character is a linebreak character. */ diff --git a/test/core/common/UtilsTest.cpp b/test/core/common/UtilsTest.cpp index 4bf1587..54890ee 100644 --- a/test/core/common/UtilsTest.cpp +++ b/test/core/common/UtilsTest.cpp @@ -131,4 +131,33 @@ TEST(Utils, collapse) ASSERT_EQ("long test", Utils::collapse(" long test ")); } -} \ No newline at end of file +TEST(Utils, isUserDefinedToken) +{ + EXPECT_FALSE(Utils::isUserDefinedToken("")); + EXPECT_FALSE(Utils::isUserDefinedToken("a")); + EXPECT_TRUE(Utils::isUserDefinedToken(":")); + EXPECT_TRUE(Utils::isUserDefinedToken("::")); + EXPECT_TRUE(Utils::isUserDefinedToken("!?")); + EXPECT_TRUE(Utils::isUserDefinedToken(".")); + EXPECT_TRUE(Utils::isUserDefinedToken("<<")); + EXPECT_TRUE(Utils::isUserDefinedToken(">>")); + EXPECT_TRUE(Utils::isUserDefinedToken("''")); + EXPECT_TRUE(Utils::isUserDefinedToken("``")); + EXPECT_TRUE(Utils::isUserDefinedToken("´´")); + EXPECT_TRUE(Utils::isUserDefinedToken("´")); + EXPECT_TRUE(Utils::isUserDefinedToken("`")); + EXPECT_TRUE(Utils::isUserDefinedToken("<")); + EXPECT_TRUE(Utils::isUserDefinedToken(">")); + EXPECT_FALSE(Utils::isUserDefinedToken("a:")); + EXPECT_FALSE(Utils::isUserDefinedToken("a:a")); + EXPECT_FALSE(Utils::isUserDefinedToken(":a")); + EXPECT_FALSE(Utils::isUserDefinedToken("{")); + EXPECT_FALSE(Utils::isUserDefinedToken("{{")); + EXPECT_FALSE(Utils::isUserDefinedToken("}}")); + EXPECT_FALSE(Utils::isUserDefinedToken("{{}{}")); + EXPECT_FALSE(Utils::isUserDefinedToken("<\\")); + EXPECT_FALSE(Utils::isUserDefinedToken("\\>")); + EXPECT_FALSE(Utils::isUserDefinedToken("{!")); +} + +} -- cgit v1.2.3 From 231f426708babe0964495ac28a54f0f2835c084a Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Mon, 2 Mar 2015 00:35:36 +0100 Subject: Unregistering all registered tokens when TokenRegistry is destroyed --- src/core/parser/stack/TokenRegistry.cpp | 8 ++++ src/core/parser/stack/TokenRegistry.hpp | 47 +++++++++++++++---- test/core/parser/stack/TokenRegistryTest.cpp | 68 +++++++++++++++------------- 3 files changed, 81 insertions(+), 42 deletions(-) (limited to 'test') diff --git a/src/core/parser/stack/TokenRegistry.cpp b/src/core/parser/stack/TokenRegistry.cpp index 21ae109..c135b98 100644 --- a/src/core/parser/stack/TokenRegistry.cpp +++ b/src/core/parser/stack/TokenRegistry.cpp @@ -16,11 +16,19 @@ along with this program. If not, see . */ +#include "Callbacks.hpp" #include "TokenRegistry.hpp" namespace ousia { namespace parser_stack { +TokenRegistry::~TokenRegistry() +{ + for (const auto &tid: tokenIds) { + parser.unregisterToken(tid.first); + } +} + TokenId TokenRegistry::registerToken(const std::string &token) { // Check whether the given token is already registered diff --git a/src/core/parser/stack/TokenRegistry.hpp b/src/core/parser/stack/TokenRegistry.hpp index 21c36b5..545db39 100644 --- a/src/core/parser/stack/TokenRegistry.hpp +++ b/src/core/parser/stack/TokenRegistry.hpp @@ -19,7 +19,7 @@ /** * @file TokenRegistry.hpp * - * Contains the TokenRegistry class used for registering all possible tokens + * Contains the TokenRegistry class used for registering all user defined tokens * during the parsing process. * * @author Andreas Stöckel (astoecke@techfak.uni-bielefeld.de) @@ -31,19 +31,22 @@ #include #include -#include "Callbacks.hpp" +#include namespace ousia { namespace parser_stack { +// Forward declarations +class ParserCallbacks; + /** - * The TokenRegistry class is used for registering all possible tokens during - * the Parsing process. The TokenRegistry class acts as an adapter between the - * parser which allocates TokenId for each unique token and the Handler classes - * which may register tokens multiple times and expect the same TokenId to be - * returned for the same token. + * The TokenRegistry class is used for registering all user defined tokens + * during the Parsing process. The TokenRegistry class acts as an adapter + * between the parser which allocates a TokenId for each unique token and the + * Handler classes which may register the same token multiple times and expect + * the same TokenId to be returned for the same token. */ -class TokenRegistry : public ParserCallbacks { +class TokenRegistry { private: /** * Reference at the ParserCallback instance the tokens are relayed to. @@ -71,14 +74,38 @@ public: */ TokenRegistry(ParserCallbacks &parser) : parser(parser) {} + /** + * Destructor of the TokenRegistry class, removes all registered tokens from + * the parser. + */ + ~TokenRegistry(); + /* No copy construction */ TokenRegistry(const TokenRegistry &) = delete; /* No assignment */ TokenRegistry &operator=(const TokenRegistry &) = delete; - TokenId registerToken(const std::string &token) override; - void unregisterToken(TokenId id) override; + /** + * Registers the given string token in the underlying parser and returns the + * TokenId of that token. If the same token string is given multiple times, + * the same TokenId is returned. The token is only registered once in the + * parser. + * + * @param token is the token that should be registered. + * @return the TokenId associated with this token. + */ + TokenId registerToken(const std::string &token); + + /** + * Unregisters the token with the given TokenId from the parser. Note that + * the token will only be unregistered if unregisterToken() has been called + * as many times as registerToken() for the same token. + * + * @param id is the id of the token returned by registerToken() that should + * be unregistered. + */ + void unregisterToken(TokenId id); }; } } diff --git a/test/core/parser/stack/TokenRegistryTest.cpp b/test/core/parser/stack/TokenRegistryTest.cpp index 390851e..20d6cd0 100644 --- a/test/core/parser/stack/TokenRegistryTest.cpp +++ b/test/core/parser/stack/TokenRegistryTest.cpp @@ -18,6 +18,7 @@ #include +#include #include namespace ousia { @@ -40,39 +41,42 @@ public: TEST(TokenRegistry, simple) { ParserCallbacksProxy parser; - TokenRegistry registry(parser); - - ASSERT_EQ(0U, parser.registerTokenCount); - ASSERT_EQ(0U, parser.unregisterTokenCount); - - ASSERT_EQ(1U, registry.registerToken("test")); - ASSERT_EQ(1U, registry.registerToken("test")); - ASSERT_EQ(2U, registry.registerToken("test2")); - ASSERT_EQ(2U, registry.registerToken("test2")); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(0U, parser.unregisterTokenCount); - - registry.unregisterToken(1); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(0U, parser.unregisterTokenCount); - - registry.unregisterToken(1); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(1U, parser.unregisterTokenCount); - - registry.unregisterToken(1); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(1U, parser.unregisterTokenCount); - - registry.unregisterToken(2); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(1U, parser.unregisterTokenCount); - - registry.unregisterToken(2); - ASSERT_EQ(2U, parser.registerTokenCount); - ASSERT_EQ(2U, parser.unregisterTokenCount); + { + TokenRegistry registry(parser); + + ASSERT_EQ(0U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + ASSERT_EQ(1U, registry.registerToken("test")); + ASSERT_EQ(1U, registry.registerToken("test")); + ASSERT_EQ(2U, registry.registerToken("test2")); + ASSERT_EQ(2U, registry.registerToken("test2")); + ASSERT_EQ(3U, registry.registerToken("test3")); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(0U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(1); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(2); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(1U, parser.unregisterTokenCount); + + registry.unregisterToken(2); + ASSERT_EQ(3U, parser.registerTokenCount); + ASSERT_EQ(2U, parser.unregisterTokenCount); + } + ASSERT_EQ(3U, parser.unregisterTokenCount); } - } } -- cgit v1.2.3 From 11ee669f29e426effaf4a1e0d82baa978219e92f Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Mon, 2 Mar 2015 10:35:22 +0100 Subject: OsxmlEventParser also fills a TokenizedData instance now --- src/formats/osxml/OsxmlEventParser.cpp | 87 +++++++++-------------------- src/formats/osxml/OsxmlEventParser.hpp | 10 ++-- src/formats/osxml/OsxmlParser.cpp | 2 +- test/formats/osxml/OsxmlEventParserTest.cpp | 11 +++- 4 files changed, 41 insertions(+), 69 deletions(-) (limited to 'test') diff --git a/src/formats/osxml/OsxmlEventParser.cpp b/src/formats/osxml/OsxmlEventParser.cpp index 83c16f0..79a8dbe 100644 --- a/src/formats/osxml/OsxmlEventParser.cpp +++ b/src/formats/osxml/OsxmlEventParser.cpp @@ -25,6 +25,7 @@ #include #include #include +#include #include "OsxmlAttributeLocator.hpp" #include "OsxmlEventParser.hpp" @@ -38,6 +39,11 @@ namespace ousia { */ class OsxmlEventParserData { public: + /** + * Current character data buffer. + */ + TokenizedData data; + /** * Contains the current depth of the parsing process. */ @@ -51,24 +57,13 @@ public: ssize_t annotationEndTagDepth; /** - * Current character data buffer. - */ - std::vector textBuf; - - /** - * Current character data start. - */ - size_t textStart; - - /** - * Current character data end. - */ - size_t textEnd; - - /** - * Default constructor. + * Constructor taking the sourceId of the file from which the XML is being + * parsed. + * + * @param sourceId is the source if of the XML file from which the data is + * currently being parsed. */ - OsxmlEventParserData(); + OsxmlEventParserData(SourceId sourceId); /** * Increments the depth. @@ -91,14 +86,6 @@ public: * @return true if character data is available. */ bool hasText(); - - /** - * Returns a Variant containing the character data and its location. - * - * @return a string variant containing the text data and the character - * location. - */ - Variant getText(SourceId sourceId); }; /* Class GuardedExpatXmlParser */ @@ -156,7 +143,7 @@ public: static const std::string TOP_LEVEL_TAG{"ousia"}; /** - * Prefix used to indicate the start of an annoation (note the trailing colon) + * Prefix used to indicate the start of an annoation (note the trailing colon). */ static const std::string ANNOTATION_START_PREFIX{"a:start:"}; @@ -203,8 +190,9 @@ static void xmlStartElementHandler(void *ref, const XML_Char *name, // If there is any text data in the buffer, issue that first if (parser->getData().hasText()) { - parser->getEvents().data( - parser->getData().getText(parser->getReader().getSourceId())); + TokenizedData &data = parser->getData().data; + parser->getEvents().data(data); + data.clear(); } // Read the argument locations -- this is only a stupid and slow hack, @@ -348,8 +336,9 @@ static void xmlEndElementHandler(void *ref, const XML_Char *name) // If there is any text data in the buffer, issue that first if (parser->getData().hasText()) { - parser->getEvents().data( - parser->getData().getText(parser->getReader().getSourceId())); + TokenizedData &data = parser->getData().data; + parser->getEvents().data(data); + data.clear(); } // Abort if the special ousia tag ends here @@ -381,18 +370,8 @@ static void xmlCharacterDataHandler(void *ref, const XML_Char *s, int len) // Synchronize the logger position SourceLocation loc = xmlSyncLoggerPosition(p, ulen); - // Fetch some variables for convenience - OsxmlEventParserData &data = parser->getData(); - std::vector &textBuf = data.textBuf; - - // Update start and end position - if (textBuf.empty()) { - data.textStart = loc.getStart(); - } - data.textEnd = loc.getEnd(); - - // Insert the data into the text buffer - textBuf.insert(textBuf.end(), &s[0], &s[ulen]); + // Append the data to the buffer + parser->getData().data.append(std::string(s, ulen), loc.getStart()); } /* Class OsxmlEvents */ @@ -401,8 +380,8 @@ OsxmlEvents::~OsxmlEvents() {} /* Class OsxmlEventParser */ -OsxmlEventParserData::OsxmlEventParserData() - : depth(0), annotationEndTagDepth(-1), textStart(0), textEnd(0) +OsxmlEventParserData::OsxmlEventParserData(SourceId sourceId) + : data(sourceId), depth(0), annotationEndTagDepth(-1) { } @@ -423,23 +402,7 @@ bool OsxmlEventParserData::inAnnotationEndTag() return (annotationEndTagDepth > 0) && (depth >= annotationEndTagDepth); } -bool OsxmlEventParserData::hasText() { return !textBuf.empty(); } - -Variant OsxmlEventParserData::getText(SourceId sourceId) -{ - // Create a variant containing the string data and the location - Variant var = - Variant::fromString(std::string{textBuf.data(), textBuf.size()}); - var.setLocation({sourceId, textStart, textEnd}); - - // Reset the text buffers - textBuf.clear(); - textStart = 0; - textEnd = 0; - - // Return the variant - return var; -} +bool OsxmlEventParserData::hasText() { return !data.empty(); } /* Class OsxmlEventParser */ @@ -448,7 +411,7 @@ OsxmlEventParser::OsxmlEventParser(CharReader &reader, OsxmlEvents &events, : reader(reader), events(events), logger(logger), - data(new OsxmlEventParserData()) + data(new OsxmlEventParserData(reader.getSourceId())) { } diff --git a/src/formats/osxml/OsxmlEventParser.hpp b/src/formats/osxml/OsxmlEventParser.hpp index 7a8c96d..4c5a485 100644 --- a/src/formats/osxml/OsxmlEventParser.hpp +++ b/src/formats/osxml/OsxmlEventParser.hpp @@ -96,10 +96,10 @@ public: /** * Called whenever string data is found. * - * @param data is a Variant containing the string data that was found in the - * XML file. + * @param data is a TokenizedData instance containing the string data that + * was found in the XML file. */ - virtual void data(const Variant &data) = 0; + virtual void data(const TokenizedData &data) = 0; }; /** @@ -179,7 +179,9 @@ public: OsxmlEvents &getEvents() const; /** - * Returns a reference at the internal data. + * Used internally to fetch a reference at the internal data. + * + * @return a reference at the internal OsxmlEventParserData structure. */ OsxmlEventParserData &getData() const; }; diff --git a/src/formats/osxml/OsxmlParser.cpp b/src/formats/osxml/OsxmlParser.cpp index 924d11b..afe0dc6 100644 --- a/src/formats/osxml/OsxmlParser.cpp +++ b/src/formats/osxml/OsxmlParser.cpp @@ -85,7 +85,7 @@ public: void rangeEnd() override { stack.rangeEnd(); } - void data(const Variant &data) override { stack.data(data); } + void data(const TokenizedData &data) override { stack.data(data); } }; /* Class OsxmlParser */ diff --git a/test/formats/osxml/OsxmlEventParserTest.cpp b/test/formats/osxml/OsxmlEventParserTest.cpp index b24a43d..d4e9443 100644 --- a/test/formats/osxml/OsxmlEventParserTest.cpp +++ b/test/formats/osxml/OsxmlEventParserTest.cpp @@ -69,9 +69,16 @@ public: events.emplace_back(OsxmlEvent::RANGE_END, Variant::arrayType{}); } - void data(const Variant &data) override + void data(const TokenizedData &data) override { - events.emplace_back(OsxmlEvent::DATA, Variant::arrayType{data}); + Token token; + Variant text; + TokenizedDataReader reader = data.reader(); + reader.read(token, TokenSet{}, WhitespaceMode::PRESERVE); + EXPECT_EQ(Tokens::Data, token.id); + text = Variant::fromString(token.content); + text.setLocation(token.getLocation()); + events.emplace_back(OsxmlEvent::DATA, Variant::arrayType{text}); } }; -- cgit v1.2.3 From 522580cfdfc9e6dc3448240448c29533e68f240f Mon Sep 17 00:00:00 2001 From: Benjamin Paassen Date: Mon, 2 Mar 2015 15:52:34 +0100 Subject: added check for witespace characters in Utils::isUserDefinedToken --- src/core/common/Utils.cpp | 15 +++++++++++---- src/core/common/Utils.hpp | 1 + test/core/common/UtilsTest.cpp | 2 ++ 3 files changed, 14 insertions(+), 4 deletions(-) (limited to 'test') diff --git a/src/core/common/Utils.cpp b/src/core/common/Utils.cpp index 219b437..a87ff6d 100644 --- a/src/core/common/Utils.cpp +++ b/src/core/common/Utils.cpp @@ -124,7 +124,8 @@ bool Utils::isUserDefinedToken(const std::string &token) // Make sure the token meets is neither empty, nor starts or ends with an // alphanumeric character const size_t len = token.size(); - if (len == 0 || isAlphanumeric(token[0]) || isAlphanumeric(token[len - 1])) { + if (len == 0 || isAlphanumeric(token[0]) || + isAlphanumeric(token[len - 1])) { return false; } @@ -134,13 +135,19 @@ bool Utils::isUserDefinedToken(const std::string &token) return false; } + // Make sure the token does not contain any whitespaces. + for (char c : token) { + if (isWhitespace(c)) { + return false; + } + } + // Make sure the token contains other characters but { and } - for (char c: token) { + for (char c : token) { if (c != '{' && c != '}') { return true; } } return false; } -} - +} \ No newline at end of file diff --git a/src/core/common/Utils.hpp b/src/core/common/Utils.hpp index 25a4de5..d9e26da 100644 --- a/src/core/common/Utils.hpp +++ b/src/core/common/Utils.hpp @@ -117,6 +117,7 @@ public: *
  • '%', '%{', '}%'
    • * * + *
  • The token does not contain any whitespaces.
    • * */ static bool isUserDefinedToken(const std::string &token); diff --git a/test/core/common/UtilsTest.cpp b/test/core/common/UtilsTest.cpp index 54890ee..2aaa430 100644 --- a/test/core/common/UtilsTest.cpp +++ b/test/core/common/UtilsTest.cpp @@ -148,6 +148,7 @@ TEST(Utils, isUserDefinedToken) EXPECT_TRUE(Utils::isUserDefinedToken("`")); EXPECT_TRUE(Utils::isUserDefinedToken("<")); EXPECT_TRUE(Utils::isUserDefinedToken(">")); + EXPECT_TRUE(Utils::isUserDefinedToken("<+>")); EXPECT_FALSE(Utils::isUserDefinedToken("a:")); EXPECT_FALSE(Utils::isUserDefinedToken("a:a")); EXPECT_FALSE(Utils::isUserDefinedToken(":a")); @@ -158,6 +159,7 @@ TEST(Utils, isUserDefinedToken) EXPECT_FALSE(Utils::isUserDefinedToken("<\\")); EXPECT_FALSE(Utils::isUserDefinedToken("\\>")); EXPECT_FALSE(Utils::isUserDefinedToken("{!")); + EXPECT_FALSE(Utils::isUserDefinedToken("< + >")); } } -- cgit v1.2.3 From ee943c5e9b60cf577ff236a694df180db89b0972 Mon Sep 17 00:00:00 2001 From: Benjamin Paassen Date: Mon, 2 Mar 2015 15:53:20 +0100 Subject: integrated syntax tokens in Domain. --- src/core/model/Domain.cpp | 193 +++++++++++++++++++++++--- src/core/model/Domain.hpp | 297 +++++++++++++++++++++++++++++++++++++---- test/core/model/DomainTest.cpp | 165 ++++++++++++++++++++++- 3 files changed, 607 insertions(+), 48 deletions(-) (limited to 'test') diff --git a/src/core/model/Domain.cpp b/src/core/model/Domain.cpp index 8255401..587a382 100644 --- a/src/core/model/Domain.cpp +++ b/src/core/model/Domain.cpp @@ -20,8 +20,9 @@ #include #include -#include #include +#include +#include #include "Domain.hpp" @@ -169,52 +170,60 @@ static NodeVector pathTo(const Node *start, Logger &logger, return shortest; } +struct CollectState { + Node *n; + size_t depth; + + CollectState(Node *n, size_t depth) : n(n), depth(depth) {} +}; + template static NodeVector collect(const Node *start, F match) { // result NodeVector res; // queue for breadth-first search of graph. - std::queue> q; + std::queue q; // put the initial node on the stack. - q.push(const_cast(start)); + q.push(CollectState(const_cast(start), 0)); // set of visited nodes. std::unordered_set visited; while (!q.empty()) { - Rooted n = q.front(); + CollectState state = q.front(); q.pop(); // do not proceed if this node was already visited. - if (!visited.insert(n.get()).second) { + if (!visited.insert(state.n).second) { continue; } - if (n->isa(&RttiTypes::StructuredClass)) { - Rooted strct = n.cast(); + if (state.n->isa(&RttiTypes::Descriptor)) { + Rooted strct{static_cast(state.n)}; // look through all fields. NodeVector fields = strct->getFieldDescriptors(); for (auto fd : fields) { // note matches. - if (match(fd)) { + if (match(fd, state.depth)) { res.push_back(fd); } // only continue in the TREE field. if (fd->getFieldType() == FieldDescriptor::FieldType::TREE) { - q.push(fd); + q.push(CollectState(fd.get(), state.depth)); } } } else { // otherwise this is a FieldDescriptor. - Rooted field = n.cast(); + Rooted field{ + static_cast(state.n)}; // and we proceed by visiting all permitted children. for (auto c : field->getChildrenWithSubclasses()) { // note matches. - if (match(c)) { + if (match(c, state.depth)) { res.push_back(c); } // We only continue our search via transparent children. if (c->isTransparent()) { - q.push(c); + q.push(CollectState(c.get(), state.depth + 1)); } } } @@ -222,28 +231,59 @@ static NodeVector collect(const Node *start, F match) return res; } +static std::vector collectPermittedTokens( + const Node *start, Handle domain) +{ + // gather SyntaxDescriptors for structure children first. + std::vector res; + collect(start, [&res](Handle n, size_t depth) { + SyntaxDescriptor stx; + if (n->isa(&RttiTypes::FieldDescriptor)) { + stx = n.cast()->getSyntaxDescriptor(depth); + } else { + stx = n.cast()->getSyntaxDescriptor(depth); + } + // do not add trivial SyntaxDescriptors. + if (!stx.isEmpty()) { + res.push_back(stx); + } + return false; + }); + // gather SyntaxDescriptors for AnnotationClasses. + for (auto a : domain->getAnnotationClasses()) { + SyntaxDescriptor stx = a->getSyntaxDescriptor(); + if (!stx.isEmpty()) { + res.push_back(stx); + } + } + return res; +} + /* Class FieldDescriptor */ FieldDescriptor::FieldDescriptor(Manager &mgr, Handle primitiveType, Handle parent, FieldType fieldType, - std::string name, bool optional) + std::string name, bool optional, + WhitespaceMode whitespaceMode) : Node(mgr, std::move(name), parent), children(this), fieldType(fieldType), primitiveType(acquire(primitiveType)), optional(optional), - primitive(true) + primitive(true), + whitespaceMode(whitespaceMode) { } FieldDescriptor::FieldDescriptor(Manager &mgr, Handle parent, FieldType fieldType, std::string name, - bool optional) + bool optional, WhitespaceMode whitespaceMode) : Node(mgr, std::move(name), parent), children(this), fieldType(fieldType), optional(optional), - primitive(false) + primitive(false), + whitespaceMode(whitespaceMode) { } @@ -272,6 +312,25 @@ bool FieldDescriptor::doValidate(Logger &logger) const } else { valid = valid & validateName(logger); } + // check start and end token. + if (!startToken.special && !startToken.token.empty() && + !Utils::isUserDefinedToken(startToken.token)) { + // TODO: Correct error message. + logger.error(std::string("Field \"") + getName() + + "\" has an invalid custom start token: " + + startToken.token, + *this); + valid = false; + } + if (!endToken.special && !endToken.token.empty() && + !Utils::isUserDefinedToken(endToken.token)) { + // TODO: Correct error message. + logger.error(std::string("Field \"") + getName() + + "\" has an invalid custom end token: " + + endToken.token, + *this); + valid = false; + } // check consistency of FieldType with the rest of the FieldDescriptor. if (primitive) { @@ -325,7 +384,7 @@ bool FieldDescriptor::doValidate(Logger &logger) const } static void gatherSubclasses( - std::unordered_set& visited, + std::unordered_set &visited, NodeVector &res, Handle strct) { // this check is to prevent cycles. @@ -334,7 +393,7 @@ static void gatherSubclasses( } for (auto sub : strct->getSubclasses()) { // this check is to prevent cycles. - if(visited.count(sub.get())){ + if (visited.count(sub.get())) { continue; } res.push_back(sub); @@ -381,7 +440,7 @@ NodeVector FieldDescriptor::pathTo(Handle field, NodeVector FieldDescriptor::getDefaultFields() const { // TODO: In principle a cast would be nicer here, but for now we copy. - NodeVector nodes = collect(this, [](Handle n) { + NodeVector nodes = collect(this, [](Handle n, size_t depth) { if (!n->isa(&RttiTypes::FieldDescriptor)) { return false; } @@ -396,6 +455,16 @@ NodeVector FieldDescriptor::getDefaultFields() const return res; } +std::vector FieldDescriptor::getPermittedTokens() const +{ + if (getParent() == nullptr || + getParent().cast()->getParent() == nullptr) { + return std::vector(); + } + return collectPermittedTokens( + this, getParent().cast()->getParent().cast()); +} + /* Class Descriptor */ void Descriptor::doResolve(ResolutionState &state) @@ -443,6 +512,25 @@ bool Descriptor::doValidate(Logger &logger) const } valid = valid & attributesDescriptor->validate(logger); } + + // check start and end token. + if (!startToken.special && !startToken.token.empty() && + !Utils::isUserDefinedToken(startToken.token)) { + logger.error(std::string("Descriptor \"") + getName() + + "\" has an invalid custom start token: " + + startToken.token, + *this); + valid = false; + } + if (!endToken.special && !endToken.token.empty() && + !Utils::isUserDefinedToken(endToken.token)) { + logger.error(std::string("Descriptor \"") + getName() + + "\" has an invalid custom end token: " + + endToken.token, + *this); + valid = false; + } + // check that only one FieldDescriptor is of type TREE. auto fds = Descriptor::getFieldDescriptors(); bool hasTREE = false; @@ -483,7 +571,7 @@ std::pair, bool> Descriptor::pathTo( NodeVector Descriptor::getDefaultFields() const { // TODO: In principle a cast would be nicer here, but for now we copy. - NodeVector nodes = collect(this, [](Handle n) { + NodeVector nodes = collect(this, [](Handle n, size_t depth) { if (!n->isa(&RttiTypes::FieldDescriptor)) { return false; } @@ -501,7 +589,7 @@ NodeVector Descriptor::getDefaultFields() const NodeVector Descriptor::getPermittedChildren() const { // TODO: In principle a cast would be nicer here, but for now we copy. - NodeVector nodes = collect(this, [](Handle n) { + NodeVector nodes = collect(this, [](Handle n, size_t depth) { return n->isa(&RttiTypes::StructuredClass); }); NodeVector res; @@ -669,6 +757,14 @@ std::pair, bool> Descriptor::createFieldDescriptor( return std::make_pair(fd, sorted); } +std::vector Descriptor::getPermittedTokens() const +{ + if (getParent() == nullptr) { + return std::vector(); + } + return collectPermittedTokens(this, getParent().cast()); +} + /* Class StructuredClass */ StructuredClass::StructuredClass(Manager &mgr, std::string name, @@ -709,6 +805,16 @@ bool StructuredClass::doValidate(Logger &logger) const logger.error(cardinality.toString() + " is not a cardinality!", *this); valid = false; } + + // check short token. + if (!shortToken.special && !shortToken.token.empty() && + !Utils::isUserDefinedToken(shortToken.token)) { + logger.error(std::string("Descriptor \"") + getName() + + "\" has an invalid custom short form token: " + + shortToken.token, + *this); + valid = false; + } // check the validity of this superclass. if (superclass != nullptr) { valid = valid & superclass->validate(logger); @@ -961,6 +1067,51 @@ Rooted Domain::createAnnotationClass(std::string name) new AnnotationClass(getManager(), std::move(name), this)}; } +static void gatherTokenDescriptors( + Handle desc, std::vector &res, + std::unordered_set &visited) +{ + // add the TokenDescriptors for the Descriptor itself. + if (!desc->getStartToken().isEmpty()) { + res.push_back(desc->getStartTokenPointer()); + } + if (!desc->getEndToken().isEmpty()) { + res.push_back(desc->getEndTokenPointer()); + } + // add the TokenDescriptors for its FieldDescriptors. + for (auto fd : desc->getFieldDescriptors()) { + if (!visited.insert(fd.get()).second) { + continue; + } + if (!fd->getStartToken().isEmpty()) { + res.push_back(fd->getStartTokenPointer()); + } + if (!fd->getEndToken().isEmpty()) { + res.push_back(fd->getEndTokenPointer()); + } + } +} + +std::vector Domain::getAllTokenDescriptors() const +{ + std::vector res; + // note all fields that are already visited because FieldReferences might + // lead to doubled fields. + std::unordered_set visited; + // add the TokenDescriptors for the StructuredClasses (and their fields). + for (auto s : structuredClasses) { + if (!s->getShortToken().isEmpty()) { + res.push_back(s->getShortTokenPointer()); + } + gatherTokenDescriptors(s, res, visited); + } + // add the TokenDescriptors for the AnnotationClasses (and their fields). + for (auto a : annotationClasses) { + gatherTokenDescriptors(a, res, visited); + } + return res; +} + /* Type registrations */ namespace RttiTypes { diff --git a/src/core/model/Domain.hpp b/src/core/model/Domain.hpp index 7e10d91..e984ed9 100644 --- a/src/core/model/Domain.hpp +++ b/src/core/model/Domain.hpp @@ -167,11 +167,13 @@ #ifndef _OUSIA_MODEL_DOMAIN_HPP_ #define _OUSIA_MODEL_DOMAIN_HPP_ +#include #include #include #include "Node.hpp" #include "RootNode.hpp" +#include "Syntax.hpp" #include "Typesystem.hpp" namespace ousia { @@ -225,6 +227,9 @@ private: Owned primitiveType; bool optional; bool primitive; + TokenDescriptor startToken; + TokenDescriptor endToken; + WhitespaceMode whitespaceMode; protected: bool doValidate(Logger &logger) const override; @@ -233,39 +238,46 @@ public: /** * This is the constructor for primitive fields. * - * @param mgr is the global Manager instance. - * @param parent is a handle of the Descriptor node that has this - * FieldDescriptor. - * @param primitiveType is a handle to some Type in some Typesystem of which - * one instance is allowed to fill this field. - * @param name is the name of this field. - * @param optional should be set to 'false' is this field needs to be - * filled in order for an instance of the parent - * Descriptor to be valid. + * @param mgr is the global Manager instance. + * @param parent is a handle of the Descriptor node that has this + * FieldDescriptor. + * @param primitiveType is a handle to some Type in some Typesystem of + *which + * one instance is allowed to fill this field. + * @param name is the name of this field. + * @param optional should be set to 'false' is this field needs to be + * filled in order for an instance of the parent + * Descriptor to be valid. + * @param whitespaceMode the WhitespaceMode to be used when an instance of + * this FieldDescriptor is parsed. */ FieldDescriptor(Manager &mgr, Handle primitiveType, Handle parent, FieldType fieldType = FieldType::TREE, - std::string name = "", bool optional = false); + std::string name = "", bool optional = false, + WhitespaceMode whitespaceMode = WhitespaceMode::COLLAPSE); /** * This is the constructor for non-primitive fields. You have to provide * children here later on. * - * @param mgr is the global Manager instance. - * @param parent is a handle of the Descriptor node that has this - * FieldDescriptor. - * @param fieldType is the FieldType of this FieldDescriptor, either - * TREE for the main or default structure or SUBTREE - * for supporting structures. - * @param name is the name of this field. - * @param optional should be set to 'false' is this field needs to be - * filled in order for an instance of the parent - * Descriptor to be valid. + * @param mgr is the global Manager instance. + * @param parent is a handle of the Descriptor node that has this + * FieldDescriptor. + * @param fieldType is the FieldType of this FieldDescriptor, either + * TREE for the main or default structure or SUBTREE + * for supporting structures. + * @param name is the name of this field. + * @param optional should be set to 'false' is this field needs to be + * filled in order for an instance of the parent + * Descriptor to be valid. + * @param whitespaceMode the WhitespaceMode to be used when an instance of + * this FieldDescriptor is parsed. */ FieldDescriptor(Manager &mgr, Handle parent = nullptr, FieldType fieldType = FieldType::TREE, - std::string name = "", bool optional = false); + std::string name = "", bool optional = false, + WhitespaceMode whitespaceMode = WhitespaceMode::COLLAPSE); /** * Returns a const reference to the NodeVector of StructuredClasses whose @@ -437,6 +449,109 @@ public: * children of an instance of this Descriptor. */ NodeVector getDefaultFields() const; + + /** + * Returns a pointer to the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * Note that this does not invalidate the FieldDescriptor. So use with + * care. + * + * @return a pointer to the start TokenDescriptor. + */ + TokenDescriptor *getStartTokenPointer() { return &startToken; } + + /** + * Returns a copy of the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * @return a copy of the start TokenDescriptor. + */ + TokenDescriptor getStartToken() const { return startToken; } + + /** + * Sets the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * @param st the new start TokenDescriptor. + */ + void setStartToken(TokenDescriptor st) + { + invalidate(); + startToken = st; + } + + /** + * Returns a pointer to the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @return a pointer to the end TokenDescriptor. + */ + TokenDescriptor *getEndTokenPointer() { return &endToken; } + + /** + * Returns a copy of the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @return a copy of the end TokenDescriptor. + */ + TokenDescriptor getEndToken() const { return endToken; } + + /** + * Sets the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @param e the new end TokenDescriptor. + */ + void setEndToken(TokenDescriptor e) + { + invalidate(); + endToken = e; + } + + /** + * Returns the WhitespaceMode to be used when an instance of this + * FieldDescriptor is parsed. + * + * @return the WhitespaceMode to be used when an instance of this + * FieldDescriptor is parsed. + */ + WhitespaceMode getWhitespaceMode() const { return whitespaceMode; } + + /** + * Sets the WhitespaceMode to be used when an instance of this + * FieldDescriptor is parsed. + * + * @param wm the WhitespaceMode to be used when an instance of this + * FieldDescriptor is parsed. + */ + WhitespaceMode setWhitespaceMode(WhitespaceMode wm) + { + return whitespaceMode = wm; + } + + /** + * Returns the SyntaxDescriptor for this FieldDescriptor. + * + * @return the SyntaxDescriptor for this FieldDescriptor. + */ + SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1) + { + SyntaxDescriptor stx{startToken.id, endToken.id, Tokens::Empty, + const_cast(this), depth}; + return stx; + } + + /** + * Returns a vector of SyntaxDescriptors, one for each Descriptor + * (StructuredClasses, AnnotationClasses or FieldDescriptors) that is + * permitted as child of this FieldDescriptor. This also makes use + * of transparency. + * + * @return a vector of SyntaxDescriptors, one for each Descriptor that is + * permitted as child of this FieldDescriptor + */ + std::vector getPermittedTokens() const; }; /** @@ -460,7 +575,10 @@ public: * * \endcode * - * key="value" inside the A-node would be an attribute, while value + * key="value" inside the A-node would be an attribute, while + * \code{.xml} + * value + * \endcode * would be a primitive field. While equivalent in XML the semantics are * different: An attribute describes indeed attributes, features of one single * node whereas a primitive field describes the _content_ of a node. @@ -472,6 +590,8 @@ class Descriptor : public Node { private: Owned attributesDescriptor; NodeVector fieldDescriptors; + TokenDescriptor startToken; + TokenDescriptor endToken; bool addAndSortFieldDescriptor(Handle fd, Logger &logger); @@ -720,6 +840,85 @@ public: * of an instance of this Descriptor in the structure tree. */ NodeVector getPermittedChildren() const; + + /** + * Returns a pointer to the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * @return a pointer to the start TokenDescriptor. + */ + TokenDescriptor *getStartTokenPointer() { return &startToken; } + + /** + * Returns a copy of the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * @return a copy of the start TokenDescriptor. + */ + TokenDescriptor getStartToken() const { return startToken; } + + /** + * Sets the start TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor starts. + * + * @param st the new start TokenDescriptor. + */ + void setStartToken(TokenDescriptor st) + { + invalidate(); + startToken = st; + } + + /** + * Returns a pointer to the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @return a pointer to the end TokenDescriptor. + */ + TokenDescriptor *getEndTokenPointer() { return &endToken; } + + /** + * Returns a copy of the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @return a copy of the end TokenDescriptor. + */ + TokenDescriptor getEndToken() const { return endToken; } + + /** + * Sets the end TokenDescriptor. This Token is used as a + * signifier during parsing that an instance of this FieldDescriptor ends. + * + * @param e the new end TokenDescriptor. + */ + void setEndToken(TokenDescriptor e) + { + invalidate(); + endToken = e; + } + + /** + * Returns the SyntaxDescriptor for this Descriptor. + * + * @return the SyntaxDescriptor for this Descriptor. + */ + virtual SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1) + { + SyntaxDescriptor stx{startToken.id, endToken.id, Tokens::Empty, + const_cast(this), depth}; + return stx; + } + + /** + * Returns a vector of SyntaxDescriptors, one for each Descriptor + * (StructuredClasses, AnnotationClasses or FieldDescriptors) that is + * permitted as child of this Descriptor. This also makes use + * of transparency. + * + * @return a vector of SyntaxDescriptors, one for each Descriptor that is + * permitted as child of this Descriptor. + */ + std::vector getPermittedTokens() const; }; /* * TODO: We should discuss Cardinalities one more time. Is it smart to define @@ -806,6 +1005,7 @@ private: NodeVector subclasses; bool transparent; bool root; + TokenDescriptor shortToken; /** * Helper method for getFieldDescriptors. @@ -963,6 +1163,50 @@ public: invalidate(); root = std::move(r); } + + /** + * Returns a pointer to the short TokenDescriptor. During parsing an + * occurence of this token will be translated to an empty instance of this + * StructuredClass. + * + * @return a pointer to the short TokenDescriptor. + */ + TokenDescriptor *getShortTokenPointer() { return &shortToken; } + + /** + * Returns a copy of the short TokenDescriptor. During parsing an + * occurence of this token will be translated to an empty instance of this + * StructuredClass. + * + * @return a copy of the short TokenDescriptor. + */ + TokenDescriptor getShortToken() const { return shortToken; } + + /** + * Sets the short TokenDescriptor. During parsing an + * occurence of this token will be translated to an empty instance of this + * StructuredClass. + * + * @param s the new short TokenDescriptor. + */ + void setShortToken(TokenDescriptor s) + { + invalidate(); + shortToken = s; + } + + /** + * Returns the SyntaxDescriptor for this StructuredClass. + * + * @return the SyntaxDescriptor for this StructuredClass. + */ + SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1) override + { + SyntaxDescriptor stx{getStartToken().id, getEndToken().id, + shortToken.id, const_cast(this), + depth}; + return stx; + } }; /** @@ -1188,6 +1432,13 @@ public: { domains.insert(domains.end(), ds.begin(), ds.end()); } + + /** + * Returns all TokenDescriptors of classes and fields in this Ontology. + * + * @return all TokenDescriptors of classes and fields in this Ontology. + */ + std::vector getAllTokenDescriptors() const; }; namespace RttiTypes { @@ -1200,4 +1451,4 @@ extern const Rtti Domain; } } -#endif /* _OUSIA_MODEL_DOMAIN_HPP_ */ +#endif /* _OUSIA_MODEL_DOMAIN_HPP_ */ \ No newline at end of file diff --git a/test/core/model/DomainTest.cpp b/test/core/model/DomainTest.cpp index 6bbf26d..f59e745 100644 --- a/test/core/model/DomainTest.cpp +++ b/test/core/model/DomainTest.cpp @@ -82,9 +82,7 @@ TEST(Domain, testDomainResolving) } // i use this wrapper due to the strange behaviour of GTEST. -static void assertFalse(bool b){ - ASSERT_FALSE(b); -} +static void assertFalse(bool b) { ASSERT_FALSE(b); } static Rooted createUnsortedPrimitiveField( Handle strct, Handle type, Logger &logger, bool tree, @@ -170,7 +168,6 @@ TEST(StructuredClass, getFieldDescriptors) } } - TEST(StructuredClass, getFieldDescriptorsCycles) { Logger logger; @@ -523,6 +520,91 @@ TEST(Descriptor, getPermittedChildrenCycles) ASSERT_EQ(A, children[0]); } +TEST(Descriptor, getSyntaxDescriptor) +{ + // build an ontology with some custom syntax. + Manager mgr{1}; + Logger logger; + Rooted sys{new SystemTypesystem(mgr)}; + // Construct the domain + Rooted domain{new Domain(mgr, sys, "ontology")}; + Rooted A{new StructuredClass( + mgr, "A", domain, Cardinality::any(), {nullptr}, true, true)}; + A->setStartToken(TokenDescriptor(Tokens::Indent)); + A->setEndToken(TokenDescriptor(Tokens::Dedent)); + { + TokenDescriptor sh{"<+>"}; + sh.id = 1; + A->setShortToken(sh); + } + // check the SyntaxDescriptor + SyntaxDescriptor stx = A->getSyntaxDescriptor(); + ASSERT_EQ(Tokens::Indent, stx.start); + ASSERT_EQ(Tokens::Dedent, stx.end); + ASSERT_EQ(1, stx.shortForm); + ASSERT_EQ(A, stx.descriptor); + ASSERT_TRUE(stx.isStruct()); + ASSERT_FALSE(stx.isAnnotation()); + ASSERT_FALSE(stx.isFieldDescriptor()); +} + +TEST(Descriptor, getPermittedTokens) +{ + // build an ontology with some custom syntax. + Manager mgr{1}; + Logger logger; + Rooted sys{new SystemTypesystem(mgr)}; + // Construct the domain + Rooted domain{new Domain(mgr, sys, "ontology")}; + // add one StructuredClass with all tokens set. + Rooted A{new StructuredClass( + mgr, "A", domain, Cardinality::any(), {nullptr}, true, true)}; + A->setStartToken(TokenDescriptor(Tokens::Indent)); + A->setEndToken(TokenDescriptor(Tokens::Dedent)); + { + TokenDescriptor sh{"<+>"}; + sh.id = 1; + A->setShortToken(sh); + } + // add a field with one token set. + Rooted A_field = A->createFieldDescriptor(logger).first; + A_field->setEndToken(TokenDescriptor(Tokens::Newline)); + A_field->addChild(A); + // add an annotation with start and end set. + Rooted A_anno = domain->createAnnotationClass("A"); + { + TokenDescriptor start{"<"}; + start.id = 7; + A_anno->setStartToken(start); + } + { + TokenDescriptor end{">"}; + end.id = 8; + A_anno->setEndToken(end); + } + // add a trivial annotation, which should not be returned. + Rooted B_anno = domain->createAnnotationClass("B"); + ASSERT_TRUE(domain->validate(logger)); + + // check result. + std::vector stxs = A->getPermittedTokens(); + ASSERT_EQ(3, stxs.size()); + // the field should be first, because A itself should not be collected + // directly. + ASSERT_EQ(A_field, stxs[0].descriptor); + ASSERT_EQ(Tokens::Empty, stxs[0].start); + ASSERT_EQ(Tokens::Newline, stxs[0].end); + ASSERT_EQ(Tokens::Empty, stxs[0].shortForm); + ASSERT_EQ(A, stxs[1].descriptor); + ASSERT_EQ(Tokens::Indent, stxs[1].start); + ASSERT_EQ(Tokens::Dedent, stxs[1].end); + ASSERT_EQ(1, stxs[1].shortForm); + ASSERT_EQ(A_anno, stxs[2].descriptor); + ASSERT_EQ(7, stxs[2].start); + ASSERT_EQ(8, stxs[2].end); + ASSERT_EQ(Tokens::Empty, stxs[2].shortForm); +} + TEST(StructuredClass, isSubclassOf) { // create an inheritance hierarchy. @@ -629,6 +711,14 @@ TEST(Domain, validate) base_field->setPrimitiveType(sys->getStringType()); ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); ASSERT_TRUE(domain->validate(logger)); + // add an invalid start token. + base_field->setStartToken(TokenDescriptor("< + >")); + ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); + ASSERT_FALSE(domain->validate(logger)); + // make it valid. + base_field->setStartToken(TokenDescriptor("<")); + ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); + ASSERT_TRUE(domain->validate(logger)); // add a subclass for our base class. Rooted sub{new StructuredClass(mgr, "sub", domain)}; // this should be valid in itself. @@ -686,4 +776,71 @@ TEST(Domain, validate) ASSERT_TRUE(domain->validate(logger)); } } + +TEST(Domain, getAllTokenDescriptors) +{ + // build an ontology with some custom syntax. + Manager mgr{1}; + Logger logger; + Rooted sys{new SystemTypesystem(mgr)}; + // Construct the domain + Rooted domain{new Domain(mgr, sys, "ontology")}; + // add one StructuredClass with all tokens set. + Rooted A{new StructuredClass( + mgr, "A", domain, Cardinality::any(), {nullptr}, true, true)}; + A->setStartToken(TokenDescriptor(Tokens::Indent)); + A->setEndToken(TokenDescriptor(Tokens::Dedent)); + { + TokenDescriptor sh{"<+>"}; + sh.id = 1; + A->setShortToken(sh); + } + // add a field with one token set. + Rooted A_field = A->createFieldDescriptor(logger).first; + A_field->setEndToken(TokenDescriptor(Tokens::Newline)); + A_field->addChild(A); + // add an annotation with start and end set. + Rooted A_anno = domain->createAnnotationClass("A"); + { + TokenDescriptor start{"<"}; + start.id = 7; + A_anno->setStartToken(start); + } + { + TokenDescriptor end{">"}; + end.id = 8; + A_anno->setEndToken(end); + } + // add a trivial annotation, which should not be returned. + Rooted B_anno = domain->createAnnotationClass("B"); + ASSERT_TRUE(domain->validate(logger)); + + // check the result. + std::vector tks = domain->getAllTokenDescriptors(); + + // A short token + ASSERT_EQ("<+>", tks[0]->token); + ASSERT_EQ(1, tks[0]->id); + ASSERT_FALSE(tks[0]->special); + // A start token + ASSERT_EQ("", tks[1]->token); + ASSERT_EQ(Tokens::Indent, tks[1]->id); + ASSERT_TRUE(tks[1]->special); + // A end token + ASSERT_EQ("", tks[2]->token); + ASSERT_EQ(Tokens::Dedent, tks[2]->id); + ASSERT_TRUE(tks[2]->special); + // A field end token + ASSERT_EQ("", tks[3]->token); + ASSERT_EQ(Tokens::Newline, tks[3]->id); + ASSERT_TRUE(tks[3]->special); + // A anno start token + ASSERT_EQ("<", tks[4]->token); + ASSERT_EQ(7, tks[4]->id); + ASSERT_FALSE(tks[4]->special); + // A anno end token + ASSERT_EQ(">", tks[5]->token); + ASSERT_EQ(8, tks[5]->id); + ASSERT_FALSE(tks[5]->special); +} } \ No newline at end of file -- cgit v1.2.3 From 4b5f37d07e4e691848b243ae795bb59893a6379c Mon Sep 17 00:00:00 2001 From: Benjamin Paassen Date: Mon, 2 Mar 2015 15:55:41 +0100 Subject: added another domain test case for invalid syntax tokens. --- test/core/model/DomainTest.cpp | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'test') diff --git a/test/core/model/DomainTest.cpp b/test/core/model/DomainTest.cpp index f59e745..b3c5771 100644 --- a/test/core/model/DomainTest.cpp +++ b/test/core/model/DomainTest.cpp @@ -701,6 +701,14 @@ TEST(Domain, validate) base->setName("myClass"); ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); ASSERT_TRUE(domain->validate(logger)); + // add an invalid short token. + base->setShortToken(TokenDescriptor("bla")); + ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); + ASSERT_FALSE(domain->validate(logger)); + // make it valid. + base->setShortToken(TokenDescriptor("!bla!")); + ASSERT_EQ(ValidationState::UNKNOWN, domain->getValidationState()); + ASSERT_TRUE(domain->validate(logger)); // Let's add a primitive field (without a primitive type at first) Rooted base_field = base->createPrimitiveFieldDescriptor(nullptr, logger).first; -- cgit v1.2.3 From 1c33913ebb5d9202575d3ca99bd17366d30f2261 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Tue, 3 Mar 2015 00:30:38 +0100 Subject: Started restructuring and adapting Stack class, reenabled unit tests (does not compile right now) --- CMakeLists.txt | 2 +- src/core/parser/stack/Stack.cpp | 447 ++++++++------- src/core/parser/stack/Stack.hpp | 13 +- src/core/parser/utils/TokenizedData.cpp | 10 +- src/core/parser/utils/TokenizedData.hpp | 12 + test/core/parser/stack/StackTest.cpp | 959 ++++++++++++++++---------------- 6 files changed, 744 insertions(+), 699 deletions(-) (limited to 'test') diff --git a/CMakeLists.txt b/CMakeLists.txt index f99c212..c7ad7a3 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -326,7 +326,7 @@ IF(TEST) test/core/model/StyleTest test/core/model/TypesystemTest test/core/parser/ParserScopeTest -# test/core/parser/stack/StackTest + test/core/parser/stack/StackTest test/core/parser/stack/StateTest test/core/parser/stack/TokenRegistryTest test/core/parser/utils/SourceOffsetVectorTest diff --git a/src/core/parser/stack/Stack.cpp b/src/core/parser/stack/Stack.cpp index e5bd224..89217ea 100644 --- a/src/core/parser/stack/Stack.cpp +++ b/src/core/parser/stack/Stack.cpp @@ -86,6 +86,12 @@ public: */ bool inImplicitDefaultField : 1; + /** + * Set to true if the handler current is in an implicitly started range + * field. + */ + bool inImplicitRangeField: 1; + /** * Set to false if this field is only opened pro-forma and does not accept * any data. Otherwise set to true. @@ -230,6 +236,18 @@ void HandlerInfo::fieldEnd() * Stub instance of HandlerInfo containing no handler information. */ static HandlerInfo EmptyHandlerInfo{true, true, false, true, true, false, true}; + +/** + * Small helper class makeing sure the reference at some variable is reset once + * the scope is left. + */ +template +struct GuardedTemporaryPointer { + T **ptr; + GuardedTemporaryPointer(T *ref, T **ptr) : ptr(ptr) { *ptr = ref; } + + ~GuardedTemporaryPointer() { *ptr = nullptr; } +}; } /* Helper functions */ @@ -352,11 +370,18 @@ private: HandlerInfo &lastInfo(); /** - * Ends all handlers that currently are not inside a field and already had - * a default field. This method is called whenever the data() and command() - * events are reached. + * Returns a set containing the tokens that should currently be processed + * by the TokenizedData instance. + * + * @return a TokenSet instance containing all tokens that should currently + * be processed. */ - void endOverdueHandlers(); + TokenSet currentTokens() const; + + /** + * Returns the whitespace mode defined by the current command. + */ + WhitespaceMode currentWhitespaceMode() const; /** * Ends the current handler and removes the corresponding element from the @@ -365,13 +390,14 @@ private: void endCurrentHandler(); /** - * Tries to start a default field for the current handler, if currently the - * handler is not inside a field and did not have a default field yet. - * - * @return true if the handler is inside a field, false if no field could - * be started. + * Ends all handlers that currently are not inside a field and already had + * a default field. Tries to start a default field for the current handler, + * if currently the handler is not inside a field and did not have a default + * field yet. This method is called whenever the data(), startAnnotation(), + * startToken(), startCommand(), annotationStart() or annotationEnd() events + * are reached. */ - bool ensureHandlerIsInField(); + void prepareCurrentHandler(); /** * Returns true if all handlers on the stack are currently valid, or false @@ -381,6 +407,30 @@ private: */ bool handlersValid(); + /** + * Called whenever there is an actual data pending on the current + * TokenizedDataReader. Tries to feed this data to the current handler. + */ + void handleData(); + + /** + * Called whenever there is a token waiting to be processed. If possible + * tries to end a current handler with this token or to start a new handler + * with the token. + * + * @param token is the token that should be handled. + */ + void handleToken(const Token &token); + + /** + * Called by the rangeEnd() and fieldEnd() methods to end the current ranged + * command. + * + * @param rangeCommand specifies whether this should end the range of a + * command with range. + */ + void handleFieldEnd(bool rangeCommand); + public: StackImpl(ParserCallbacks &parser, ParserContext &ctx, const std::multimap &states); @@ -403,7 +453,6 @@ public: TokenId registerToken(const std::string &token) override; void unregisterToken(TokenId id) override; Variant readData() override; - bool hasData(); void pushTokens(const std::vector &tokens) override; void popTokens() override; }; @@ -492,16 +541,6 @@ std::set StackImpl::expectedCommands() return res; } -const State &StackImpl::currentState() const -{ - return stack.empty() ? States::None : stack.back().state(); -} - -std::string StackImpl::currentCommandName() const -{ - return stack.empty() ? std::string{} : stack.back().name(); -} - const State *StackImpl::findTargetState(const std::string &name) { const State *currentState = &(this->currentState()); @@ -527,6 +566,28 @@ const State *StackImpl::findTargetStateOrWildcard(const std::string &name) return targetState; } +const State &StackImpl::currentState() const +{ + return stack.empty() ? States::None : stack.back().state(); +} + +std::string StackImpl::currentCommandName() const +{ + return stack.empty() ? std::string{} : stack.back().name(); +} + +TokenSet StackImpl::currentTokens() const +{ + // TODO: Implement + return Tokens{}; +} + +WhitespaceMode currentWhitespaceMode() const +{ + // TODO: Implement + return WhitespaceMode::COLLAPSE; +} + HandlerInfo &StackImpl::currentInfo() { return stack.empty() ? EmptyHandlerInfo : stack.back(); @@ -536,6 +597,8 @@ HandlerInfo &StackImpl::lastInfo() return stack.size() < 2U ? EmptyHandlerInfo : stack[stack.size() - 2]; } +/* Stack helper functions */ + void StackImpl::endCurrentHandler() { if (!stack.empty()) { @@ -563,44 +626,37 @@ void StackImpl::endCurrentHandler() } } -void StackImpl::endOverdueHandlers() +void StackImpl::prepareCurrentHandler() { - if (!stack.empty()) { - // Fetch the handler info for the current top-level element - HandlerInfo &info = stack.back(); + // Repeat until a valid handler is found on the stack + while (true) { + // Fetch the handler for the current top-level element + HandlerInfo &info = currentInfo(); - // Abort if this handler currently is inside a field - if (info.inField || (!info.hadDefaultField && info.valid)) { + // If the current Handler is in a field, there is nothing to be done, + // abort + if (info.inField) { return; } - // Otherwise end the current handler - endCurrentHandler(); - } -} - -bool StackImpl::ensureHandlerIsInField() -{ - // If the current handler is not in a field (and actually has a handler) - // try to start a default field - HandlerInfo &info = currentInfo(); - if (!info.inField && info.handler != nullptr) { - // Abort if the element already had a default field or the handler is - // not valid + // If the current field already had a default field or is not valid, + // end it and repeat if (info.hadDefaultField || !info.valid) { - return false; + endCurrentHandler(); + continue; } // Try to start a new default field, abort if this did not work bool isDefault = true; if (!info.handler->fieldStart(isDefault, info.fieldIdx)) { - return false; + endCurrentHandler(); + continue; } - // Mark the field as started - info.fieldStart(true, true, true); + // Mark the field as started and return -- the field should be marked + // is implicit if this is not a field with range + info.fieldStart(true, !info.range, true, info.range); } - return true; } bool StackImpl::handlersValid() @@ -613,13 +669,105 @@ bool StackImpl::handlersValid() return true; } +void StackImpl::handleData() +{ + // Repeat until we found some handle willingly consuming the data + while (true) { + // Prepare the stack -- make sure all overdue handlers are ended and + // we currently are in an open field + prepareCurrentHandler(); + + // Fetch the current handler information + HandlerInfo &info = currentInfo(); + + // If this field should not get any data, log an error and do not + // call the "data" handler + if (!info.inValidField) { + if (!info.hadDefaultField) { + logger().error("Did not expect any data here", data); + } + return; + } + + // If we're currently in an invalid subtree, just eat the data and abort + if (!handlersValid()) { + return; + } + + // Fork the logger and set it as temporary logger for the "data" + // method. We only want to keep error messages if this was not a + // try to implicitly open a default field. + LoggerFork loggerFork = logger().fork(); + info.handler->setLogger(loggerFork); + + // Pass the data to the current Handler instance + bool valid = false; + try { + valid = info.handler->data(); + } + catch (LoggableException ex) { + loggerFork.log(ex); + } + + // Reset the logger instance of the handler as soon as possible + info.handler->resetLogger(); + + // If placing the data here failed and we're currently in an + // implicitly opened field, just unroll the stack to the next field + // and try again + if (!valid && info.inImplicitDefaultField) { + endCurrentHandler(); + continue; + } + + // Commit the content of the logger fork. Do not change the valid flag. + loggerFork.commit(); + } +} + +void StackImpl::handleToken(const Token &token) { + // TODO: Implement + // Just eat them for now +} + +void StackImpl::handleFieldEnd(bool rangedCommand) +{ + // Throw away all overdue handlers, start the default field at least once + // if this has not been done yet (this is important for range commands) + prepareStack(); + + // Close all implicit default fields + while (!stack.empty()) { + HandlerInfo &info = currentInfo(); + if (!info.inImplicitDefaultField) { + break; + } + endCurrentHandler(); + } + + // Fetch the information attached to the current handler + HandlerInfo &info = currentInfo(); + if (!info.inField || stack.empty()) { + logger().error("Got field end, but there is no field here to end"); + return; + } + + // Only continue if the current handler stack is in a valid state, do not + // call the fieldEnd function if something went wrong before + if (handlersValid()) { + if (info.range && info.inDefaultField) + info.handler->fieldEnd(); + } + + // This command no longer is in a field + info.fieldEnd(); +} + +/* Class StackImpl public functions */ + void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, bool range) { - // End handlers that already had a default field and are currently not - // active. - endOverdueHandlers(); - // Make sure the given identifier is valid (preventing "*" from being // malicously passed to this function) if (!Utils::isNamespacedIdentifier(name.asString())) { @@ -629,6 +777,10 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, } while (true) { + // Prepare the stack -- make sure all overdue handlers are ended and + // we currently are in an open field + prepareCurrentHandler(); + // Try to find a target state for the given command, if none can be // found and the current command does not have an open field, then try // to create an empty default field, otherwise this is an exception @@ -644,12 +796,6 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, } } - // Make sure we're currently inside a field - if (!ensureHandlerIsInField()) { - endCurrentHandler(); - continue; - } - // Fork the logger. We do not want any validation errors to skip LoggerFork loggerFork = logger().fork(); @@ -670,17 +816,14 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, HandlerInfo &parentInfo = lastInfo(); HandlerInfo &info = currentInfo(); - // Call the "start" method of the handler, store the result of the - // start - // method as the validity of the handler -- do not call the start - // method + // Call the "start" method of the handler, store the result of the start + // method as the validity of the handler -- do not call the start method // if the stack is currently invalid (as this may cause further, // unwanted errors) bool validStack = handlersValid(); info.valid = false; if (validStack) { - // Canonicalize the arguments (if this has not already been - // done), + // Canonicalize the arguments (if this has not already been done), // allow additional arguments and numeric indices Variant::mapType canonicalArgs = args; targetState->arguments.validateMap(canonicalArgs, loggerFork, true, @@ -697,10 +840,8 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, } // We started the command within an implicit default field and it is - // not - // valid -- remove both the new handler and the parent field from - // the - // stack + // not valid -- remove both the new handler and the parent field from + // the stack if (!info.valid && parentInfo.inImplicitDefaultField) { endCurrentHandler(); endCurrentHandler(); @@ -708,9 +849,8 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, } // If we ended up here, starting the command may or may not have - // worked, - // but after all, we cannot unroll the stack any further. Update the - // "valid" flag, commit any potential error messages and return. + // worked, but after all, we cannot unroll the stack any further. Update + // the "valid" flag, commit any potential error messages and return. info.valid = parentInfo.valid && info.valid; info.range = range; loggerFork.commit(); @@ -732,106 +872,31 @@ void StackImpl::annotationEnd(const Variant &className, void StackImpl::rangeEnd() { - // TODO + handleFieldEnd(true); } void StackImpl::data(const TokenizedData &data) { - // TODO: Rewrite this function for token handling - // TODO: This loop needs to be refactored out - /*while (!data.atEnd()) { - // End handlers that already had a default field and are currently - not - // active. - endOverdueHandlers(); - - const bool hasNonWhitespaceText = data.hasNonWhitespaceText(); - - // Check whether there is any command the data can be sent to -- if - not, - // make sure the data actually is data - if (stack.empty()) { - if (hasNonWhitespaceText) { - throw LoggableException("No command here to receive data.", - data); - } - return; - } - - // Fetch the current command handler information - HandlerInfo &info = currentInfo(); - - // Make sure the current handler has an open field - if (!ensureHandlerIsInField()) { - endCurrentHandler(); - continue; - } - - // If this field should not get any data, log an error and do not - call - // the "data" handler - if (!info.inValidField) { - // If the "hadDefaultField" flag is set, we already issued an - error - // message - if (!info.hadDefaultField) { - if (hasNonWhitespaceText) { - logger().error("Did not expect any data here", data); - } - return; - } - } - - if (handlersValid() && info.inValidField) { - // Fork the logger and set it as temporary logger for the - "start" - // method. We only want to keep error messages if this was not a - try - // to implicitly open a default field. - LoggerFork loggerFork = logger().fork(); - info.handler->setLogger(loggerFork); - - // Pass the data to the current Handler instance - bool valid = false; - try { - // Create a fork of the TokenizedData and let the handler - work - // on it - TokenizedData dataFork = data; - valid = info.handler->data(dataFork); - - // If the data was validly handled by the handler, commit - the - // change - if (valid) { - data = dataFork; - } - } - catch (LoggableException ex) { - loggerFork.log(ex); - } - - // Reset the logger instance as soon as possible - info.handler->resetLogger(); - - // If placing the data here failed and we're currently in an - // implicitly opened field, just unroll the stack to the next - field - // and try again - if (!valid && info.inImplicitDefaultField) { - endCurrentHandler(); - continue; - } - - // Commit the content of the logger fork. Do not change the - valid - // flag. - loggerFork.commit(); - } - - // There was no reason to unroll the stack any further, so continue - return; - }*/ + // Fetch a reader for the given tokenized data instance. + TokenizedDataReader reader = data.reader(); + + // Use the GuardedTemporaryPointer to make sure that the member variable + // dataReader is resetted to nullptr once this scope is left. + GuardedTemporaryPointer ptr(&reader, &dataReader); + + // Peek a token from the reader, repeat until all tokens have been read + Token token; + while (reader.peek(token, currentTokens(), currentWhitespaceMode())) { + // Handle the token as text data or as actual token + if (token.id == Tokens::Data) { + handleData(); + } else { + handleToken(token); + } + + // Consume the peeked token + reader.consumePeek(); + } } void StackImpl::fieldStart(bool isDefault) @@ -853,8 +918,7 @@ void StackImpl::fieldStart(bool isDefault) } // If the handler already had a default field we cannot start a new - // field - // (the default field always is the last field) -- mark the command as + // field (the default field always is the last field) -- mark the command as // invalid if (info.hadDefaultField) { logger().error(std::string("Got field start, but command \"") + @@ -862,8 +926,7 @@ void StackImpl::fieldStart(bool isDefault) std::string("\" does not have any more fields")); } - // Copy the isDefault flag to a local variable, the fieldStart method - // will + // Copy the isDefault flag to a local variable, the fieldStart method will // write into this variable bool defaultField = isDefault; @@ -891,40 +954,11 @@ void StackImpl::fieldStart(bool isDefault) void StackImpl::fieldEnd() { - // Unroll the stack until the next explicitly open field - while (!stack.empty()) { - HandlerInfo &info = currentInfo(); - if (info.inField && !info.inImplicitDefaultField) { - break; - } - endCurrentHandler(); - } - - // Fetch the information attached to the current handler - HandlerInfo &info = currentInfo(); - if (!info.inField || info.inImplicitDefaultField || stack.empty()) { - logger().error( - "Got field end, but there is no command for which to end the " - "field."); - return; - } - - // Only continue if the current handler stack is in a valid state, do - // not - // call the fieldEnd function if something went wrong before - if (handlersValid() && !info.hadDefaultField && info.inValidField) { - try { - info.handler->fieldEnd(); - } - catch (LoggableException ex) { - logger().log(ex); - } - } - - // This command no longer is in a field - info.fieldEnd(); + handleFieldEnd(false); } +/* Class StackImpl HandlerCallbacks */ + TokenId StackImpl::registerToken(const std::string &token) { return tokenRegistry.registerToken(token); @@ -950,14 +984,7 @@ Variant StackImpl::readData() if (dataReader != nullptr) { TokenizedDataReaderFork dataReaderFork = dataReader->fork(); Token token; - - // TODO: Use correct token set - TokenSet tokens; - - // TODO: Use correct whitespace mode - WhitespaceMode mode = WhitespaceMode::COLLAPSE; - - dataReaderFork.read(token, tokens, mode); + dataReaderFork.read(token, currentTokens(), currentWhitespaceMode()); if (token.id == Tokens::Data) { Variant res = Variant::fromString(token.content); res.setLocation(token.getLocation()); @@ -967,8 +994,6 @@ Variant StackImpl::readData() return Variant{}; } -bool StackImpl::hasData() { return readData() != nullptr; } - /* Class Stack */ Stack::Stack(ParserCallbacks &parser, ParserContext &ctx, @@ -1013,5 +1038,7 @@ void Stack::fieldStart(bool isDefault) { impl->fieldStart(isDefault); } void Stack::fieldEnd() { impl->fieldEnd(); } void Stack::data(const TokenizedData &data) { impl->data(data); } + +void Stack::data(const std::string &str) { data(TokenizedData(str)); } } } diff --git a/src/core/parser/stack/Stack.hpp b/src/core/parser/stack/Stack.hpp index de281d4..1de7cff 100644 --- a/src/core/parser/stack/Stack.hpp +++ b/src/core/parser/stack/Stack.hpp @@ -150,13 +150,24 @@ public: /** * Function that should be called whenever character data is found in the - * input stream. May only be called if the currently is a command on the + * input stream. May only be called if there currently is a command on the * stack. * * @param data is a TokenizedData instance containing the pre-segmented data * that should be read. */ void data(const TokenizedData &data); + + /** + * Function that may be called whenever character data is found in the + * input stream. May only be called if the currently is a command on the + * stack. This method is mainly intended for unit testing. Pass a + * TokenizedData instance to the + * + * @param str is a string containing the data that should be passed to the + * tokenizer. + */ + void data(const std::string &str); }; } } diff --git a/src/core/parser/utils/TokenizedData.cpp b/src/core/parser/utils/TokenizedData.cpp index c3c4f98..d8a8b37 100644 --- a/src/core/parser/utils/TokenizedData.cpp +++ b/src/core/parser/utils/TokenizedData.cpp @@ -29,8 +29,7 @@ namespace ousia { /** * Maximum token length. */ -constexpr TokenLength MaxTokenLength = - std::numeric_limits::max(); +constexpr TokenLength MaxTokenLength = std::numeric_limits::max(); namespace { /** @@ -510,6 +509,13 @@ TokenizedData::TokenizedData(SourceId sourceId) { } +TokenizedData::TokenizedData(const std::string &data, SourceOffset offsStart, + SourceId sourceId) + : TokenizedData(sourceId) +{ + append(data, offsStart); +} + TokenizedData::~TokenizedData() {} size_t TokenizedData::append(const std::string &data, SourceOffset offsStart, diff --git a/src/core/parser/utils/TokenizedData.hpp b/src/core/parser/utils/TokenizedData.hpp index b72ca02..bc937f2 100644 --- a/src/core/parser/utils/TokenizedData.hpp +++ b/src/core/parser/utils/TokenizedData.hpp @@ -95,6 +95,18 @@ public: */ TokenizedData(SourceId sourceId); + /** + * Creates a new instance of TokenizedData, takes a SourceId and an initial + * string buffer. + * + * @param data is the string that should be appended to the buffer. + * @param offsStart is the start offset in bytes in the input file. + * @param sourceId is the source identifier that should be used for + * constructing the location when returning tokens. + */ + TokenizedData(const std::string &data, SourceOffset offsStart = 0, + SourceId sourceId = InvalidSourceId); + /** * Destructor. Needs to be defined explicitly for freeing a shared pointer * of the incomplete TokenizedDataImpl type. diff --git a/test/core/parser/stack/StackTest.cpp b/test/core/parser/stack/StackTest.cpp index 83966d5..8f6c4df 100644 --- a/test/core/parser/stack/StackTest.cpp +++ b/test/core/parser/stack/StackTest.cpp @@ -21,6 +21,7 @@ #include #include +#include #include #include #include @@ -38,70 +39,69 @@ static StandaloneEnvironment env(logger); namespace { +class Parser : public ParserCallbacks { + TokenId registerToken(const std::string &token) override + { + return Tokens::Empty; + } + + void unregisterToken(TokenId id) override + { + // Do nothing here + } +}; + +static Parser parser; + struct Tracker { - int startCount; + int startCommandCount; + int startAnnotationCount; + int startTokenCount; + int endTokenCount; int endCount; int fieldStartCount; int fieldEndCount; - int annotationStartCount; - int annotationEndCount; int dataCount; - Variant::mapType startArgs; - bool fieldStartIsDefault; - size_t fieldStartIdx; - Variant annotationStartClassName; - Variant::mapType annotationStartArgs; - Variant annotationEndClassName; - Variant annotationEndElementName; - TokenizedData dataData; - - bool startResult; - bool fieldStartSetIsDefault; + bool startCommandResult; + bool startAnnotationResult; + bool startTokenResult; + Handler::EndTokenResult endTokenResult; bool fieldStartResult; - bool annotationStartResult; - bool annotationEndResult; bool dataResult; Tracker() { reset(); } void reset() { - startCount = 0; + startCommandCount = 0; + startAnnotationCount = 0; + startTokenCount = 0; + endTokenCount = 0; endCount = 0; fieldStartCount = 0; fieldEndCount = 0; - annotationStartCount = 0; - annotationEndCount = 0; dataCount = 0; - startArgs = Variant::mapType{}; - fieldStartIsDefault = false; - fieldStartIdx = 0; - annotationStartClassName = Variant::fromString(std::string{}); - annotationStartArgs = Variant::mapType{}; - annotationEndClassName = Variant::fromString(std::string{}); - annotationEndElementName = Variant::fromString(std::string{}); - dataData = TokenizedData(); - - startResult = true; - fieldStartSetIsDefault = false; + startCommandResult = true; + startAnnotationResult = true; + startTokenResult = true; + endTokenResult = Handler::EndTokenResult::ENDED_THIS; fieldStartResult = true; - annotationStartResult = true; - annotationEndResult = true; dataResult = true; } - void expect(int startCount, int endCount, int fieldStartCount, - int fieldEndCount, int annotationStartCount, - int annotationEndCount, int dataCount) + void expect(int startCommandCount, int endCount, int fieldStartCount, + int fieldEndCount, int dataCount, int startAnnotationCount = 0, + int startTokenCount = 0, int endTokenCount = 0) { - EXPECT_EQ(startCount, this->startCount); + EXPECT_EQ(startCommandCount, this->startCommandCount); + EXPECT_EQ(startAnnotationCount, this->startAnnotationCount); + EXPECT_EQ(startTokenCount, this->startTokenCount); + EXPECT_EQ(endTokenCount, this->endTokenCount); EXPECT_EQ(endCount, this->endCount); EXPECT_EQ(fieldStartCount, this->fieldStartCount); EXPECT_EQ(fieldEndCount, this->fieldEndCount); - EXPECT_EQ(annotationStartCount, this->annotationStartCount); - EXPECT_EQ(annotationEndCount, this->annotationEndCount); EXPECT_EQ(dataCount, this->dataCount); } }; @@ -113,55 +113,44 @@ private: TestHandler(const HandlerData &handlerData) : Handler(handlerData) {} public: - bool start(Variant::mapType &args) override + bool startCommand(Variant::mapType &args) override { - tracker.startCount++; - tracker.startArgs = args; - if (!tracker.startResult) { - logger().error( - "The TestHandler was told not to allow a field start. So it " - "doesn't. The TestHandler always obeys its master."); - } - return tracker.startResult; + tracker.startCommandCount++; + return tracker.startCommandResult; } - void end() override { tracker.endCount++; } - - bool fieldStart(bool &isDefault, size_t fieldIdx) override + bool startAnnotation(Variant::mapType &args, + AnnotationType annotationType) override { - tracker.fieldStartCount++; - tracker.fieldStartIsDefault = isDefault; - tracker.fieldStartIdx = fieldIdx; - if (tracker.fieldStartSetIsDefault) { - isDefault = true; - } - return tracker.fieldStartResult; + tracker.startAnnotationCount++; + return tracker.startAnnotationResult; } - void fieldEnd() override { tracker.fieldEndCount++; } + bool startToken(Handle node) override + { + tracker.startTokenCount++; + return tracker.startTokenResult; + } - bool annotationStart(const Variant &className, - Variant::mapType &args) override + EndTokenResult endToken(const Token &token, Handle node) override { - tracker.annotationStartCount++; - tracker.annotationStartClassName = className; - tracker.annotationStartArgs = args; - return tracker.annotationStartResult; + tracker.endTokenCount++; + return tracker.endTokenResult; } - bool annotationEnd(const Variant &className, - const Variant &elementName) override + void end() override { tracker.endCount++; } + + bool fieldStart(bool &isDefault, size_t fieldIdx) override { - tracker.annotationEndCount++; - tracker.annotationEndClassName = className; - tracker.annotationEndElementName = elementName; - return tracker.annotationEndResult; + tracker.fieldStartCount++; + return tracker.fieldStartResult; } - bool data(TokenizedData &data) override + void fieldEnd() override { tracker.fieldEndCount++; } + + bool data() override { tracker.dataCount++; - tracker.dataData = data; return tracker.dataResult; } @@ -205,544 +194,544 @@ TEST(Stack, basicTest) tracker.reset(); logger.reset(); { - Stack s{env.context, States::TestHandlers}; + Stack s{parser, env.context, States::TestHandlers}; EXPECT_EQ("", s.currentCommandName()); EXPECT_EQ(&States::None, &s.currentState()); - s.command("document", {}); + s.commandStart("document", {}, true); s.fieldStart(true); s.data("test1"); EXPECT_EQ("document", s.currentCommandName()); EXPECT_EQ(&States::Document, &s.currentState()); - tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fse, dc, sac, stc, etc - s.command("body", {}); + s.commandStart("body", {}, true); s.fieldStart(true); s.data("test2"); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(2, 0, 2, 0, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 2, 0, 2); // scc, ec, fsc, fse, dc, sac, stc, etc - s.command("inner", {}); + s.commandStart("inner", {}, true); s.fieldStart(true); EXPECT_EQ("inner", s.currentCommandName()); EXPECT_EQ(&States::BodyChildren, &s.currentState()); s.fieldEnd(); - tracker.expect(3, 0, 3, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(3, 0, 3, 1, 2); // scc, ec, fsc, fse, dc, sac, stc, etc s.fieldEnd(); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(3, 1, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(3, 1, 3, 2, 2); // scc, ec, fsc, fse, dc, sac, stc, etc - s.command("body", {}); + s.commandStart("body", {}, true); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(4, 2, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(4, 2, 3, 2, 2); // scc, ec, fsc, fse, dc, sac, stc, etc s.fieldStart(true); s.data("test3"); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); s.fieldEnd(); - tracker.expect(4, 2, 4, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(4, 2, 4, 3, 3); // scc, ec, fsc, fse, dc, sac, stc, etc EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); s.fieldEnd(); - tracker.expect(4, 3, 4, 4, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(4, 3, 4, 4, 3); // scc, ec, fsc, fse, dc, sac, stc, etc EXPECT_EQ("document", s.currentCommandName()); EXPECT_EQ(&States::Document, &s.currentState()); } - tracker.expect(4, 4, 4, 4, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(4, 4, 4, 4, 3); // scc, ec, fsc, fse, dc, sac, stc, etc ASSERT_FALSE(logger.hasError()); } - +/* TEST(Stack, errorInvalidCommands) { - Stack s{env.context, States::TestHandlers}; - tracker.reset(); - EXPECT_THROW(s.command("body", {}), LoggableException); - s.command("document", {}); - s.fieldStart(true); - EXPECT_THROW(s.command("document", {}), LoggableException); - s.command("empty", {}); - s.fieldStart(true); - EXPECT_THROW(s.command("body", {}), LoggableException); - s.command("special", {}); - s.fieldStart(true); - s.fieldEnd(); - s.fieldEnd(); - s.fieldEnd(); - - logger.reset(); - s.fieldEnd(); - ASSERT_TRUE(logger.hasError()); - - EXPECT_THROW(s.data("test"), LoggableException); - EXPECT_EQ(&States::None, &s.currentState()); + Stack s{env.context, States::TestHandlers}; + tracker.reset(); + EXPECT_THROW(s.command("body", {}), LoggableException); + s.command("document", {}); + s.fieldStart(true); + EXPECT_THROW(s.command("document", {}), LoggableException); + s.command("empty", {}); + s.fieldStart(true); + EXPECT_THROW(s.command("body", {}), LoggableException); + s.command("special", {}); + s.fieldStart(true); + s.fieldEnd(); + s.fieldEnd(); + s.fieldEnd(); + + logger.reset(); + s.fieldEnd(); + ASSERT_TRUE(logger.hasError()); + + EXPECT_THROW(s.data("test"), LoggableException); + EXPECT_EQ(&States::None, &s.currentState()); } TEST(Stack, validation) { - Stack s{env.context, States::TestHandlers}; - tracker.reset(); - logger.reset(); - - s.command("arguments", {}); - EXPECT_TRUE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); - - logger.reset(); - s.command("arguments", {{"a", 5}}); - EXPECT_TRUE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); - - logger.reset(); - s.command("arguments", {{"a", 5}, {"b", "test"}}); - EXPECT_FALSE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); + Stack s{env.context, States::TestHandlers}; + tracker.reset(); + logger.reset(); + + s.command("arguments", {}); + EXPECT_TRUE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); + + logger.reset(); + s.command("arguments", {{"a", 5}}); + EXPECT_TRUE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); + + logger.reset(); + s.command("arguments", {{"a", 5}, {"b", "test"}}); + EXPECT_FALSE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); } TEST(Stack, invalidCommandName) { - tracker.reset(); - logger.reset(); - - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("a_", {}); - tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(2, 1, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("a_:b", {}); - tracker.expect(3, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(3, 2, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("_a", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("a:", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("a:_b", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.command("a_", {}); + tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(2, 1, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.command("a_:b", {}); + tracker.expect(3, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(3, 2, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + ASSERT_THROW(s.command("_a", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + ASSERT_THROW(s.command("a:", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + ASSERT_THROW(s.command("a:_b", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, multipleFields) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {{"a", false}}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("a", s.currentCommandName()); - EXPECT_EQ(Variant::mapType({{"a", false}}), tracker.startArgs); - - s.fieldStart(false); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_FALSE(tracker.fieldStartIsDefault); - EXPECT_EQ(0U, tracker.fieldStartIdx); - - s.data("test"); - tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test", tracker.dataData.text().asString()); - - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(false); - tracker.expect(1, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_FALSE(tracker.fieldStartIsDefault); - EXPECT_EQ(1U, tracker.fieldStartIdx); - - s.data("test2"); - tracker.expect(1, 0, 2, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test2", tracker.dataData.text().asString()); - - s.fieldEnd(); - tracker.expect(1, 0, 2, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(true); - tracker.expect(1, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_TRUE(tracker.fieldStartIsDefault); - EXPECT_EQ(2U, tracker.fieldStartIdx); - - s.data("test3"); - tracker.expect(1, 0, 3, 2, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test3", tracker.dataData.text().asString()); - - s.fieldEnd(); - tracker.expect(1, 0, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {{"a", false}}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_EQ("a", s.currentCommandName()); + EXPECT_EQ(Variant::mapType({{"a", false}}), tracker.startArgs); + + s.fieldStart(false); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_FALSE(tracker.fieldStartIsDefault); + EXPECT_EQ(0U, tracker.fieldStartIdx); + + s.data("test"); + tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_EQ("test", tracker.dataData.text().asString()); + + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(false); + tracker.expect(1, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_FALSE(tracker.fieldStartIsDefault); + EXPECT_EQ(1U, tracker.fieldStartIdx); + + s.data("test2"); + tracker.expect(1, 0, 2, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_EQ("test2", tracker.dataData.text().asString()); + + s.fieldEnd(); + tracker.expect(1, 0, 2, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(true); + tracker.expect(1, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_TRUE(tracker.fieldStartIsDefault); + EXPECT_EQ(2U, tracker.fieldStartIdx); + + s.data("test3"); + tracker.expect(1, 0, 3, 2, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + EXPECT_EQ("test3", tracker.dataData.text().asString()); + + s.fieldEnd(); + tracker.expect(1, 0, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnNewCommand) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.command("b", {}); + tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnNewCommandWithExplicitDefaultField) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(2, 0, 2, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + + s.command("b", {}); + tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("b", s.currentCommandName()); + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(2, 0, 2, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noImplicitDefaultFieldOnIncompatibleCommand) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - - tracker.fieldStartResult = false; - s.command("b", {}); - tracker.expect(2, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + + tracker.fieldStartResult = false; + s.command("b", {}); + tracker.expect(2, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noImplicitDefaultFieldIfDefaultFieldGiven) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - s.fieldStart(true); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - - s.command("b", {}); - tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + s.fieldStart(true); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + + s.command("b", {}); + tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noEndIfStartFails) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - - tracker.startResult = false; - s.command("b", {}); - tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_TRUE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("a", s.currentCommandName()); + + tracker.startResult = false; + s.command("b", {}); + tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_TRUE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnData) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.data("test"); - tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + { + Stack s{env.context, States::AnyHandlers}; + + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.data("test"); + tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, autoFieldEnd) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, autoImplicitFieldEnd) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - s.command("b", {}); - s.command("c", {}); - s.command("d", {}); - s.command("e", {}); - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(5, 0, 5, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(5, 5, 5, 5, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + s.command("b", {}); + s.command("c", {}); + s.command("d", {}); + s.command("e", {}); + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(5, 0, 5, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(5, 5, 5, 5, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, invalidDefaultField) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.fieldStartResult = false; + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, errorInvalidDefaultFieldData) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - s.fieldStart(true); - ASSERT_FALSE(logger.hasError()); - s.data("test"); - ASSERT_TRUE(logger.hasError()); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.fieldStartResult = false; + s.fieldStart(true); + ASSERT_FALSE(logger.hasError()); + s.data("test"); + ASSERT_TRUE(logger.hasError()); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, errorInvalidFieldData) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - ASSERT_FALSE(logger.hasError()); - s.fieldStart(false); - ASSERT_TRUE(logger.hasError()); - s.data("test"); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.fieldStartResult = false; + ASSERT_FALSE(logger.hasError()); + s.fieldStart(false); + ASSERT_TRUE(logger.hasError()); + s.data("test"); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, errorFieldStartNoCommand) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - Stack s{env.context, States::AnyHandlers}; - ASSERT_THROW(s.fieldStart(false), LoggableException); - ASSERT_THROW(s.fieldStart(true), LoggableException); - tracker.expect(0, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + Stack s{env.context, States::AnyHandlers}; + ASSERT_THROW(s.fieldStart(false), LoggableException); + ASSERT_THROW(s.fieldStart(true), LoggableException); + tracker.expect(0, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, errorMultipleFieldStarts) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(false); - ASSERT_FALSE(logger.hasError()); - s.fieldStart(false); - ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(false); + ASSERT_FALSE(logger.hasError()); + s.fieldStart(false); + ASSERT_TRUE(logger.hasError()); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, errorMultipleFieldEnds) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(false); - s.fieldEnd(); - ASSERT_FALSE(logger.hasError()); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldEnd(); - ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(false); + s.fieldEnd(); + ASSERT_FALSE(logger.hasError()); + tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + ASSERT_TRUE(logger.hasError()); + tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, errorOpenField) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(false); - ASSERT_FALSE(logger.hasError()); - } - ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(false); + ASSERT_FALSE(logger.hasError()); + } + ASSERT_TRUE(logger.hasError()); + tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc } TEST(Stack, fieldEndWhenImplicitDefaultFieldOpen) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - s.fieldStart(true); - s.command("b", {}); - s.data("test"); - s.fieldEnd(); - tracker.expect(2, 1, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(2, 2, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + s.fieldStart(true); + s.command("b", {}); + s.data("test"); + s.fieldEnd(); + tracker.expect(2, 1, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(2, 2, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, fieldAfterDefaultField) { - tracker.reset(); - logger.reset(); - - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldStart(false); - tracker.expect(2, 0, 2, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.data("f1"); - tracker.expect(2, 0, 2, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - s.fieldEnd(); - tracker.expect(2, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - tracker.fieldStartSetIsDefault = true; - - s.fieldStart(false); - tracker.fieldStartSetIsDefault = false; - tracker.expect(2, 0, 3, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - s.data("f2"); - tracker.expect(2, 0, 3, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - s.fieldEnd(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_FALSE(logger.hasError()); - s.fieldStart(false); - ASSERT_TRUE(logger.hasError()); - logger.reset(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - s.data("f3"); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - s.fieldEnd(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - - s.fieldEnd(); - tracker.expect(2, 1, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(2, 2, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); -} + tracker.reset(); + logger.reset(); + + { + Stack s{env.context, States::AnyHandlers}; + s.command("a", {}); + tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(true); + tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.command("b", {}); + tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldStart(false); + tracker.expect(2, 0, 2, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.data("f1"); + tracker.expect(2, 0, 2, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(2, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.fieldStartSetIsDefault = true; + + s.fieldStart(false); + tracker.fieldStartSetIsDefault = false; + tracker.expect(2, 0, 3, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + s.data("f2"); + tracker.expect(2, 0, 3, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + + ASSERT_FALSE(logger.hasError()); + s.fieldStart(false); + ASSERT_TRUE(logger.hasError()); + logger.reset(); + tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + s.data("f3"); + tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + + s.fieldEnd(); + tracker.expect(2, 1, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + } + tracker.expect(2, 2, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + ASSERT_FALSE(logger.hasError()); +}*/ } } -- cgit v1.2.3 From fb8d4cdf01909b61e4e5d0806ec6de178ff0058c Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Tue, 3 Mar 2015 14:34:14 +0100 Subject: Finished stack and adapted all unit tests --- src/core/parser/stack/Stack.cpp | 218 +++++++--- src/core/parser/stack/Stack.hpp | 4 +- test/core/parser/stack/StackTest.cpp | 772 ++++++++++++++++++++--------------- 3 files changed, 595 insertions(+), 399 deletions(-) (limited to 'test') diff --git a/src/core/parser/stack/Stack.cpp b/src/core/parser/stack/Stack.cpp index 89217ea..f341f1d 100644 --- a/src/core/parser/stack/Stack.cpp +++ b/src/core/parser/stack/Stack.cpp @@ -30,9 +30,15 @@ #include "TokenRegistry.hpp" #include "TokenStack.hpp" +#define STACK_DEBUG_OUTPUT 0 +#if STACK_DEBUG_OUTPUT +#include +#endif + namespace ousia { namespace parser_stack { namespace { + /* Class HandlerInfo */ /** @@ -86,12 +92,6 @@ public: */ bool inImplicitDefaultField : 1; - /** - * Set to true if the handler current is in an implicitly started range - * field. - */ - bool inImplicitRangeField: 1; - /** * Set to false if this field is only opened pro-forma and does not accept * any data. Otherwise set to true. @@ -109,11 +109,10 @@ public: HandlerInfo(); /** - * Constructor of the HandlerInfo class, allows to set all flags manually. + * Constructor of the HandlerInfo class, allows to set some flags manually. */ - HandlerInfo(bool valid, bool implicit, bool range, bool inField, - bool inDefaultField, bool inImplicitDefaultField, - bool inValidField); + HandlerInfo(bool implicit, bool inField, bool inDefaultField, + bool inImplicitDefaultField); /** * Constructor of the HandlerInfo class, taking a shared_ptr to the handler @@ -178,18 +177,17 @@ HandlerInfo::HandlerInfo(std::shared_ptr handler) { } -HandlerInfo::HandlerInfo(bool valid, bool implicit, bool range, bool inField, - bool inDefaultField, bool inImplicitDefaultField, - bool inValidField) +HandlerInfo::HandlerInfo(bool implicit, bool inField, bool inDefaultField, + bool inImplicitDefaultField) : handler(nullptr), fieldIdx(0), - valid(valid), + valid(true), implicit(implicit), - range(range), + range(false), inField(inField), inDefaultField(inDefaultField), inImplicitDefaultField(inImplicitDefaultField), - inValidField(inValidField), + inValidField(true), hadDefaultField(false) { } @@ -235,7 +233,7 @@ void HandlerInfo::fieldEnd() /** * Stub instance of HandlerInfo containing no handler information. */ -static HandlerInfo EmptyHandlerInfo{true, true, false, true, true, false, true}; +static HandlerInfo EmptyHandlerInfo{true, true, true, true}; /** * Small helper class makeing sure the reference at some variable is reset once @@ -386,8 +384,10 @@ private: /** * Ends the current handler and removes the corresponding element from the * stack. + * + * @return true if a command was ended, false otherwise. */ - void endCurrentHandler(); + bool endCurrentHandler(); /** * Ends all handlers that currently are not inside a field and already had @@ -396,8 +396,10 @@ private: * field yet. This method is called whenever the data(), startAnnotation(), * startToken(), startCommand(), annotationStart() or annotationEnd() events * are reached. + * + * @return true if the current command is in a valid field. */ - void prepareCurrentHandler(); + bool prepareCurrentHandler(bool startImplicitDefaultField = true); /** * Returns true if all handlers on the stack are currently valid, or false @@ -413,23 +415,23 @@ private: */ void handleData(); - /** - * Called whenever there is a token waiting to be processed. If possible - * tries to end a current handler with this token or to start a new handler - * with the token. - * - * @param token is the token that should be handled. - */ - void handleToken(const Token &token); + /** + * Called whenever there is a token waiting to be processed. If possible + * tries to end a current handler with this token or to start a new handler + * with the token. + * + * @param token is the token that should be handled. + */ + void handleToken(const Token &token); /** * Called by the rangeEnd() and fieldEnd() methods to end the current ranged * command. * - * @param rangeCommand specifies whether this should end the range of a + * @param endRange specifies whether this should end the range of a * command with range. */ - void handleFieldEnd(bool rangeCommand); + void handleFieldEnd(bool endRange); public: StackImpl(ParserCallbacks &parser, ParserContext &ctx, @@ -579,10 +581,10 @@ std::string StackImpl::currentCommandName() const TokenSet StackImpl::currentTokens() const { // TODO: Implement - return Tokens{}; + return TokenSet{}; } -WhitespaceMode currentWhitespaceMode() const +WhitespaceMode StackImpl::currentWhitespaceMode() const { // TODO: Implement return WhitespaceMode::COLLAPSE; @@ -599,7 +601,7 @@ HandlerInfo &StackImpl::lastInfo() /* Stack helper functions */ -void StackImpl::endCurrentHandler() +bool StackImpl::endCurrentHandler() { if (!stack.empty()) { // Fetch the handler info for the current top-level element @@ -623,29 +625,43 @@ void StackImpl::endCurrentHandler() // Remove the element from the stack stack.pop_back(); + return true; } + return false; } -void StackImpl::prepareCurrentHandler() +bool StackImpl::prepareCurrentHandler(bool startImplicitDefaultField) { // Repeat until a valid handler is found on the stack - while (true) { + while (!stack.empty()) { // Fetch the handler for the current top-level element HandlerInfo &info = currentInfo(); // If the current Handler is in a field, there is nothing to be done, // abort if (info.inField) { - return; + return true; } // If the current field already had a default field or is not valid, // end it and repeat - if (info.hadDefaultField || !info.valid) { + if ((info.hadDefaultField || !startImplicitDefaultField) || + !info.valid) { + // We cannot end the command if it is marked as "range" command + if (info.range) { + return false; + } + + // End the current handler endCurrentHandler(); continue; } + // Abort if starting new default fields is not allowed here + if (!startImplicitDefaultField) { + return false; + } + // Try to start a new default field, abort if this did not work bool isDefault = true; if (!info.handler->fieldStart(isDefault, info.fieldIdx)) { @@ -655,8 +671,10 @@ void StackImpl::prepareCurrentHandler() // Mark the field as started and return -- the field should be marked // is implicit if this is not a field with range - info.fieldStart(true, !info.range, true, info.range); + info.fieldStart(true, !info.range, true); + return true; } + return false; } bool StackImpl::handlersValid() @@ -675,7 +693,9 @@ void StackImpl::handleData() while (true) { // Prepare the stack -- make sure all overdue handlers are ended and // we currently are in an open field - prepareCurrentHandler(); + if (stack.empty() || !prepareCurrentHandler()) { + throw LoggableException("Did not expect any data here"); + } // Fetch the current handler information HandlerInfo &info = currentInfo(); @@ -684,7 +704,7 @@ void StackImpl::handleData() // call the "data" handler if (!info.inValidField) { if (!info.hadDefaultField) { - logger().error("Did not expect any data here", data); + logger().error("Did not expect any data here"); } return; } @@ -722,24 +742,25 @@ void StackImpl::handleData() // Commit the content of the logger fork. Do not change the valid flag. loggerFork.commit(); + return; } } -void StackImpl::handleToken(const Token &token) { +void StackImpl::handleToken(const Token &token) +{ // TODO: Implement // Just eat them for now } -void StackImpl::handleFieldEnd(bool rangedCommand) +void StackImpl::handleFieldEnd(bool endRange) { - // Throw away all overdue handlers, start the default field at least once - // if this has not been done yet (this is important for range commands) - prepareStack(); + // Throw away all overdue handlers + prepareCurrentHandler(false); // Close all implicit default fields while (!stack.empty()) { HandlerInfo &info = currentInfo(); - if (!info.inImplicitDefaultField) { + if (!info.inImplicitDefaultField || info.range) { break; } endCurrentHandler(); @@ -747,16 +768,37 @@ void StackImpl::handleFieldEnd(bool rangedCommand) // Fetch the information attached to the current handler HandlerInfo &info = currentInfo(); - if (!info.inField || stack.empty()) { - logger().error("Got field end, but there is no field here to end"); + if (stack.empty() || (!info.inField && !endRange) || + (!info.range && endRange)) { + if (endRange) { + logger().error( + "Got end of range, but there is no command here to end"); + } else { + logger().error("Got field end, but there is no field here to end"); + } return; } // Only continue if the current handler stack is in a valid state, do not // call the fieldEnd function if something went wrong before if (handlersValid()) { - if (info.range && info.inDefaultField) - info.handler->fieldEnd(); + // End the current field if it is valid + if (info.inValidField) { + info.handler->fieldEnd(); + info.fieldEnd(); + } + + // End the complete command if this is a range command, start the + // default field for once if range command did not have a default field + if (info.range && endRange) { + if (!info.hadDefaultField) { + bool isDefault = true; + info.handler->fieldStart(isDefault, true); + info.fieldStart(true, true, true); + } + endCurrentHandler(); + return; + } } // This command no longer is in a field @@ -768,6 +810,9 @@ void StackImpl::handleFieldEnd(bool rangedCommand) void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, bool range) { + // Call prepareCurrentHandler once to end all overdue commands + prepareCurrentHandler(); + // Make sure the given identifier is valid (preventing "*" from being // malicously passed to this function) if (!Utils::isNamespacedIdentifier(name.asString())) { @@ -787,8 +832,8 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, const State *targetState = findTargetStateOrWildcard(name.asString()); if (targetState == nullptr) { HandlerInfo &info = currentInfo(); - if (info.inImplicitDefaultField || !info.inField) { - endCurrentHandler(); + if ((info.inImplicitDefaultField || !info.inField) && + endCurrentHandler()) { continue; } else { throw buildInvalidCommandException(name.asString(), @@ -843,9 +888,10 @@ void StackImpl::commandStart(const Variant &name, const Variant::mapType &args, // not valid -- remove both the new handler and the parent field from // the stack if (!info.valid && parentInfo.inImplicitDefaultField) { - endCurrentHandler(); - endCurrentHandler(); - continue; + // Only continue if the parent handler could actually be removed + if (endCurrentHandler() && endCurrentHandler()) { + continue; + } } // If we ended up here, starting the command may or may not have @@ -870,10 +916,7 @@ void StackImpl::annotationEnd(const Variant &className, // TODO } -void StackImpl::rangeEnd() -{ - handleFieldEnd(true); -} +void StackImpl::rangeEnd() { handleFieldEnd(true); } void StackImpl::data(const TokenizedData &data) { @@ -882,7 +925,7 @@ void StackImpl::data(const TokenizedData &data) // Use the GuardedTemporaryPointer to make sure that the member variable // dataReader is resetted to nullptr once this scope is left. - GuardedTemporaryPointer ptr(&reader, &dataReader); + GuardedTemporaryPointer ptr(&reader, &dataReader); // Peek a token from the reader, repeat until all tokens have been read Token token; @@ -952,10 +995,7 @@ void StackImpl::fieldStart(bool isDefault) info.fieldStart(defaultField, false, valid); } -void StackImpl::fieldEnd() -{ - handleFieldEnd(false); -} +void StackImpl::fieldEnd() { handleFieldEnd(false); } /* Class StackImpl HandlerCallbacks */ @@ -1017,28 +1057,70 @@ std::string Stack::currentCommandName() const void Stack::commandStart(const Variant &name, const Variant::mapType &args, bool range) { +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: commandStart " << name << " " << args << " " << range + << std::endl; +#endif impl->commandStart(name, args, range); } void Stack::annotationStart(const Variant &className, const Variant &args, bool range) { +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: annotationStart " << className << " " << args << " " + << range << std::endl; +#endif impl->annotationStart(className, args, range); } void Stack::annotationEnd(const Variant &className, const Variant &elementName) { +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: annotationEnd " << className << " " << elementName + << std::endl; +#endif impl->annotationEnd(className, elementName); } -void Stack::rangeEnd() { impl->rangeEnd(); } +void Stack::rangeEnd() +{ +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: rangeEnd" << std::endl; +#endif + impl->rangeEnd(); +} -void Stack::fieldStart(bool isDefault) { impl->fieldStart(isDefault); } +void Stack::fieldStart(bool isDefault) +{ +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: fieldStart " << isDefault << std::endl; +#endif + impl->fieldStart(isDefault); +} -void Stack::fieldEnd() { impl->fieldEnd(); } +void Stack::fieldEnd() +{ +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: fieldEnd" << std::endl; +#endif + impl->fieldEnd(); +} -void Stack::data(const TokenizedData &data) { impl->data(data); } +void Stack::data(const TokenizedData &data) +{ +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: data" << std::endl; +#endif + impl->data(data); +} -void Stack::data(const std::string &str) { data(TokenizedData(str)); } +void Stack::data(const std::string &str) +{ +#if STACK_DEBUG_OUTPUT + std::cout << "STACK: data (string) " << str << std::endl; +#endif + data(TokenizedData(str)); +} } } diff --git a/src/core/parser/stack/Stack.hpp b/src/core/parser/stack/Stack.hpp index 1de7cff..6d42f10 100644 --- a/src/core/parser/stack/Stack.hpp +++ b/src/core/parser/stack/Stack.hpp @@ -104,7 +104,7 @@ public: * @param range if true, the started command has an explicit range. */ void commandStart(const Variant &name, const Variant::mapType &args, - bool range); + bool range = false); /** * Function that should be called whenever an annotation starts. @@ -115,7 +115,7 @@ public: * @param range if true, the annotation fields have an explicit range. */ void annotationStart(const Variant &className, const Variant &args, - bool range); + bool range = false); /** * Function that should be called whenever an annotation ends. diff --git a/test/core/parser/stack/StackTest.cpp b/test/core/parser/stack/StackTest.cpp index 8f6c4df..a831c32 100644 --- a/test/core/parser/stack/StackTest.cpp +++ b/test/core/parser/stack/StackTest.cpp @@ -70,6 +70,16 @@ struct Tracker { bool fieldStartResult; bool dataResult; + Variant::mapType startCommandArgs; + Variant::mapType startAnnotationArgs; + + bool fieldStartReturnValue; + size_t fieldStartIdx; + bool fieldStartIsDefault; + bool fieldStartSetIsDefault; + + Variant dataData; + Tracker() { reset(); } void reset() @@ -89,6 +99,15 @@ struct Tracker { endTokenResult = Handler::EndTokenResult::ENDED_THIS; fieldStartResult = true; dataResult = true; + + startCommandArgs = Variant::mapType{}; + startAnnotationArgs = Variant::mapType{}; + + fieldStartIdx = 0; + fieldStartIsDefault = false; + fieldStartSetIsDefault = false; + + dataData = Variant{}; } void expect(int startCommandCount, int endCount, int fieldStartCount, @@ -115,13 +134,20 @@ private: public: bool startCommand(Variant::mapType &args) override { + tracker.startCommandArgs = args; tracker.startCommandCount++; + if (!tracker.startCommandResult) { + logger().error( + "TestHandler was told not to allow a command start. " + "TestHandler always obeys its master."); + } return tracker.startCommandResult; } bool startAnnotation(Variant::mapType &args, AnnotationType annotationType) override { + tracker.startAnnotationArgs = args; tracker.startAnnotationCount++; return tracker.startAnnotationResult; } @@ -142,6 +168,11 @@ public: bool fieldStart(bool &isDefault, size_t fieldIdx) override { + tracker.fieldStartIsDefault = isDefault; + tracker.fieldStartIdx = fieldIdx; + if (tracker.fieldStartSetIsDefault) { + isDefault = true; + } tracker.fieldStartCount++; return tracker.fieldStartResult; } @@ -150,6 +181,7 @@ public: bool data() override { + tracker.dataData = readData(); tracker.dataCount++; return tracker.dataResult; } @@ -199,456 +231,518 @@ TEST(Stack, basicTest) EXPECT_EQ("", s.currentCommandName()); EXPECT_EQ(&States::None, &s.currentState()); - s.commandStart("document", {}, true); + s.commandStart("document", {}); s.fieldStart(true); s.data("test1"); EXPECT_EQ("document", s.currentCommandName()); EXPECT_EQ(&States::Document, &s.currentState()); - tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc - s.commandStart("body", {}, true); + s.commandStart("body", {}); s.fieldStart(true); s.data("test2"); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(2, 0, 2, 0, 2); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(2, 0, 2, 0, 2); // scc, ec, fsc, fec, dc, sac, stc, etc - s.commandStart("inner", {}, true); + s.commandStart("inner", {}); s.fieldStart(true); EXPECT_EQ("inner", s.currentCommandName()); EXPECT_EQ(&States::BodyChildren, &s.currentState()); s.fieldEnd(); - tracker.expect(3, 0, 3, 1, 2); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(3, 0, 3, 1, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldEnd(); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(3, 1, 3, 2, 2); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(3, 1, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc - s.commandStart("body", {}, true); + s.commandStart("body", {}); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); - tracker.expect(4, 2, 3, 2, 2); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(4, 2, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldStart(true); s.data("test3"); EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); s.fieldEnd(); - tracker.expect(4, 2, 4, 3, 3); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(4, 2, 4, 3, 3); // scc, ec, fsc, fec, dc, sac, stc, etc EXPECT_EQ("body", s.currentCommandName()); EXPECT_EQ(&States::Body, &s.currentState()); s.fieldEnd(); - tracker.expect(4, 3, 4, 4, 3); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(4, 3, 4, 4, 3); // scc, ec, fsc, fec, dc, sac, stc, etc EXPECT_EQ("document", s.currentCommandName()); EXPECT_EQ(&States::Document, &s.currentState()); } - tracker.expect(4, 4, 4, 4, 3); // scc, ec, fsc, fse, dc, sac, stc, etc + tracker.expect(4, 4, 4, 4, 3); // scc, ec, fsc, fec, dc, sac, stc, etc ASSERT_FALSE(logger.hasError()); } -/* + +TEST(Stack, basicTestRangeCommands) +{ + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::TestHandlers}; + + EXPECT_EQ("", s.currentCommandName()); + EXPECT_EQ(&States::None, &s.currentState()); + + s.commandStart("document", {}, true); + EXPECT_EQ("document", s.currentCommandName()); + EXPECT_EQ(&States::Document, &s.currentState()); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + s.data("test1"); + tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + + s.commandStart("body", {}, true); + tracker.expect(2, 0, 1, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + s.data("test2"); + tracker.expect(2, 0, 2, 0, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("body", s.currentCommandName()); + EXPECT_EQ(&States::Body, &s.currentState()); + + s.commandStart("inner", {}, true); + tracker.expect(3, 0, 2, 0, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("inner", s.currentCommandName()); + EXPECT_EQ(&States::BodyChildren, &s.currentState()); + s.rangeEnd(); + tracker.expect(3, 1, 3, 1, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("body", s.currentCommandName()); + EXPECT_EQ(&States::Body, &s.currentState()); + s.rangeEnd(); + tracker.expect(3, 2, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + + s.commandStart("body", {}, true); + EXPECT_EQ("body", s.currentCommandName()); + EXPECT_EQ(&States::Body, &s.currentState()); + tracker.expect(4, 2, 3, 2, 2); // scc, ec, fsc, fse, dc, sac, stc, etc + s.fieldStart(true); + tracker.expect(4, 2, 4, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + s.data("test3"); + tracker.expect(4, 2, 4, 2, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("body", s.currentCommandName()); + EXPECT_EQ(&States::Body, &s.currentState()); + s.fieldEnd(); + tracker.expect(4, 2, 4, 3, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("body", s.currentCommandName()); + EXPECT_EQ(&States::Body, &s.currentState()); + s.rangeEnd(); + tracker.expect(4, 3, 4, 3, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + + EXPECT_EQ("document", s.currentCommandName()); + EXPECT_EQ(&States::Document, &s.currentState()); + s.rangeEnd(); + tracker.expect(4, 4, 4, 4, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(4, 4, 4, 4, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); +} + TEST(Stack, errorInvalidCommands) { - Stack s{env.context, States::TestHandlers}; - tracker.reset(); - EXPECT_THROW(s.command("body", {}), LoggableException); - s.command("document", {}); - s.fieldStart(true); - EXPECT_THROW(s.command("document", {}), LoggableException); - s.command("empty", {}); - s.fieldStart(true); - EXPECT_THROW(s.command("body", {}), LoggableException); - s.command("special", {}); - s.fieldStart(true); - s.fieldEnd(); - s.fieldEnd(); - s.fieldEnd(); + Stack s{parser, env.context, States::TestHandlers}; + tracker.reset(); + EXPECT_THROW(s.commandStart("body", {}), LoggableException); + s.commandStart("document", {}); + s.fieldStart(true); + EXPECT_THROW(s.commandStart("document", {}), LoggableException); + s.commandStart("empty", {}); + s.fieldStart(true); + EXPECT_THROW(s.commandStart("body", {}), LoggableException); + s.commandStart("special", {}); + s.fieldStart(true); + s.fieldEnd(); + s.fieldEnd(); + s.fieldEnd(); - logger.reset(); - s.fieldEnd(); - ASSERT_TRUE(logger.hasError()); + logger.reset(); + s.fieldEnd(); + ASSERT_TRUE(logger.hasError()); - EXPECT_THROW(s.data("test"), LoggableException); - EXPECT_EQ(&States::None, &s.currentState()); + EXPECT_THROW(s.data("test"), LoggableException); + EXPECT_EQ(&States::None, &s.currentState()); } TEST(Stack, validation) { - Stack s{env.context, States::TestHandlers}; - tracker.reset(); - logger.reset(); + Stack s{parser, env.context, States::TestHandlers}; + tracker.reset(); + logger.reset(); - s.command("arguments", {}); - EXPECT_TRUE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); + s.commandStart("arguments", {}); + EXPECT_TRUE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); - logger.reset(); - s.command("arguments", {{"a", 5}}); - EXPECT_TRUE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); + logger.reset(); + s.commandStart("arguments", {{"a", 5}}, false); + EXPECT_TRUE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); - logger.reset(); - s.command("arguments", {{"a", 5}, {"b", "test"}}); - EXPECT_FALSE(logger.hasError()); - s.fieldStart(true); - s.fieldEnd(); + logger.reset(); + s.commandStart("arguments", {{"a", 5}, {"b", "test"}}, false); + EXPECT_FALSE(logger.hasError()); + s.fieldStart(true); + s.fieldEnd(); } TEST(Stack, invalidCommandName) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("a_", {}); - tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(2, 1, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - s.command("a_:b", {}); - tracker.expect(3, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(3, 2, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("_a", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("a:", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - - ASSERT_THROW(s.command("a:_b", {}), LoggableException); - tracker.expect(3, 3, 3, 3, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + s.commandStart("a_", {}); + tracker.expect(2, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(2, 1, 2, 2, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + s.commandStart("a_:b", {}); + tracker.expect(3, 2, 2, 2, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(3, 2, 3, 3, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + ASSERT_THROW(s.commandStart("_a", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + ASSERT_THROW(s.commandStart("a:", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + + ASSERT_THROW(s.commandStart("a:_b", {}), LoggableException); + tracker.expect(3, 3, 3, 3, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, multipleFields) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {{"a", false}}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("a", s.currentCommandName()); - EXPECT_EQ(Variant::mapType({{"a", false}}), tracker.startArgs); + s.commandStart("a", {{"a", false}}, false); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("a", s.currentCommandName()); + EXPECT_EQ(Variant::mapType({{"a", false}}), tracker.startCommandArgs); - s.fieldStart(false); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_FALSE(tracker.fieldStartIsDefault); - EXPECT_EQ(0U, tracker.fieldStartIdx); + s.fieldStart(false); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_FALSE(tracker.fieldStartIsDefault); + EXPECT_EQ(0U, tracker.fieldStartIdx); - s.data("test"); - tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test", tracker.dataData.text().asString()); + s.data("test"); + tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("test", tracker.dataData.asString()); - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 1); // scc, ec, fsc, fec, dc, sac, stc, etc - s.fieldStart(false); - tracker.expect(1, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_FALSE(tracker.fieldStartIsDefault); - EXPECT_EQ(1U, tracker.fieldStartIdx); + s.fieldStart(false); + tracker.expect(1, 0, 2, 1, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_FALSE(tracker.fieldStartIsDefault); + EXPECT_EQ(1U, tracker.fieldStartIdx); - s.data("test2"); - tracker.expect(1, 0, 2, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test2", tracker.dataData.text().asString()); + s.data("test2"); + tracker.expect(1, 0, 2, 1, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("test2", tracker.dataData.asString()); - s.fieldEnd(); - tracker.expect(1, 0, 2, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(1, 0, 2, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc - s.fieldStart(true); - tracker.expect(1, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_TRUE(tracker.fieldStartIsDefault); - EXPECT_EQ(2U, tracker.fieldStartIdx); + s.fieldStart(true); + tracker.expect(1, 0, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_TRUE(tracker.fieldStartIsDefault); + EXPECT_EQ(2U, tracker.fieldStartIdx); - s.data("test3"); - tracker.expect(1, 0, 3, 2, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - EXPECT_EQ("test3", tracker.dataData.text().asString()); + s.data("test3"); + tracker.expect(1, 0, 3, 2, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ("test3", tracker.dataData.asString()); - s.fieldEnd(); - tracker.expect(1, 0, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 3, 3, 0, 0, 3); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + s.fieldEnd(); + tracker.expect(1, 0, 3, 3, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 3, 3, 3); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnNewCommand) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + s.commandStart("b", {}); + tracker.expect(2, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(2, 2, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnNewCommandWithExplicitDefaultField) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(2, 0, 2, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 2, 2, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + s.commandStart("b", {}); + tracker.expect(2, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("b", s.currentCommandName()); + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(2, 0, 2, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 2, 2, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noImplicitDefaultFieldOnIncompatibleCommand) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); - tracker.fieldStartResult = false; - s.command("b", {}); - tracker.expect(2, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + tracker.fieldStartResult = false; + s.commandStart("b", {}); + tracker.expect(2, 1, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noImplicitDefaultFieldIfDefaultFieldGiven) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - s.fieldStart(true); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); + s.fieldStart(true); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); - s.command("b", {}); - tracker.expect(2, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(2, 2, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + s.commandStart("b", {}); + tracker.expect(2, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("b", s.currentCommandName()); + } + tracker.expect(2, 2, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, noEndIfStartFails) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("a", s.currentCommandName()); + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_EQ("a", s.currentCommandName()); - tracker.startResult = false; - s.command("b", {}); - tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_EQ("b", s.currentCommandName()); - } - tracker.expect(3, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_TRUE(logger.hasError()); + tracker.startCommandResult = false; + s.commandStart("b", {}); + tracker.expect(3, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + EXPECT_EQ(&States::None, &s.currentState()); + } + tracker.expect(3, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_TRUE(logger.hasError()); } TEST(Stack, implicitDefaultFieldOnData) { - tracker.reset(); - logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; + tracker.reset(); + logger.reset(); + { + Stack s{parser, env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.data("test"); - tracker.expect(1, 0, 1, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + s.data("test"); + tracker.expect(1, 0, 1, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 1, 1); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, autoFieldEnd) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, autoImplicitFieldEnd) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - s.command("b", {}); - s.command("c", {}); - s.command("d", {}); - s.command("e", {}); - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(5, 0, 5, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(5, 5, 5, 5, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + s.commandStart("b", {}); + s.commandStart("c", {}); + s.commandStart("d", {}); + s.commandStart("e", {}); + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(5, 0, 5, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(5, 5, 5, 5, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, invalidDefaultField) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - s.fieldStart(true); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - ASSERT_FALSE(logger.hasError()); + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.fieldStartResult = false; + s.fieldStart(true); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); } TEST(Stack, errorInvalidDefaultFieldData) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - s.fieldStart(true); - ASSERT_FALSE(logger.hasError()); - s.data("test"); - ASSERT_TRUE(logger.hasError()); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.fieldStartResult = false; + s.fieldStart(true); + ASSERT_FALSE(logger.hasError()); + s.data("test"); + ASSERT_TRUE(logger.hasError()); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, errorInvalidFieldData) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.fieldStartResult = false; - ASSERT_FALSE(logger.hasError()); - s.fieldStart(false); - ASSERT_TRUE(logger.hasError()); - s.data("test"); - s.fieldEnd(); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.fieldStartResult = false; + ASSERT_FALSE(logger.hasError()); + s.fieldStart(false); + ASSERT_TRUE(logger.hasError()); + s.data("test"); + s.fieldEnd(); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, errorFieldStartNoCommand) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - Stack s{env.context, States::AnyHandlers}; - ASSERT_THROW(s.fieldStart(false), LoggableException); - ASSERT_THROW(s.fieldStart(true), LoggableException); - tracker.expect(0, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + Stack s{parser, env.context, States::AnyHandlers}; + ASSERT_THROW(s.fieldStart(false), LoggableException); + ASSERT_THROW(s.fieldStart(true), LoggableException); + tracker.expect(0, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, errorMultipleFieldStarts) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.fieldStart(false); - ASSERT_FALSE(logger.hasError()); - s.fieldStart(false); - ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(false); + ASSERT_FALSE(logger.hasError()); + s.fieldStart(false); + ASSERT_TRUE(logger.hasError()); + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.fieldEnd(); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldEnd(); + tracker.expect(1, 0, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, errorMultipleFieldEnds) { - tracker.reset(); - logger.reset(); + tracker.reset(); + logger.reset(); - { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + { + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.fieldStart(false); - s.fieldEnd(); - ASSERT_FALSE(logger.hasError()); - tracker.expect(1, 0, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - s.fieldEnd(); - ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc - } - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.fieldStart(false); + s.fieldEnd(); + ASSERT_FALSE(logger.hasError()); + tracker.expect(1, 0, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.fieldEnd(); + ASSERT_TRUE(logger.hasError()); + tracker.expect(1, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(1, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, errorOpenField) @@ -657,15 +751,15 @@ TEST(Stack, errorOpenField) logger.reset(); { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldStart(false); ASSERT_FALSE(logger.hasError()); } ASSERT_TRUE(logger.hasError()); - tracker.expect(1, 1, 1, 1, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(1, 1, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc } TEST(Stack, fieldEndWhenImplicitDefaultFieldOpen) @@ -674,15 +768,15 @@ TEST(Stack, fieldEndWhenImplicitDefaultFieldOpen) logger.reset(); { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); s.fieldStart(true); - s.command("b", {}); + s.commandStart("b", {}); s.data("test"); s.fieldEnd(); - tracker.expect(2, 1, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 1, 2, 2, 1); // scc, ec, fsc, fec, dc, sac, stc, etc } - tracker.expect(2, 2, 2, 2, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 2, 2, 2, 1); // scc, ec, fsc, fec, dc, sac, stc, etc ASSERT_FALSE(logger.hasError()); } @@ -692,46 +786,66 @@ TEST(Stack, fieldAfterDefaultField) logger.reset(); { - Stack s{env.context, States::AnyHandlers}; - s.command("a", {}); - tracker.expect(1, 0, 0, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + Stack s{parser, env.context, States::AnyHandlers}; + s.commandStart("a", {}); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldStart(true); - tracker.expect(1, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(1, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc - s.command("b", {}); - tracker.expect(2, 0, 1, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + s.commandStart("b", {}); + tracker.expect(2, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldStart(false); - tracker.expect(2, 0, 2, 0, 0, 0, 0); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 2, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc s.data("f1"); - tracker.expect(2, 0, 2, 0, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 2, 0, 1); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldEnd(); - tracker.expect(2, 0, 2, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 2, 1, 1); // scc, ec, fsc, fec, dc, sac, stc, etc tracker.fieldStartSetIsDefault = true; s.fieldStart(false); tracker.fieldStartSetIsDefault = false; - tracker.expect(2, 0, 3, 1, 0, 0, 1); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 1, 1); // scc, ec, fsc, fec, dc, sac, stc, etc s.data("f2"); - tracker.expect(2, 0, 3, 1, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 1, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldEnd(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc ASSERT_FALSE(logger.hasError()); s.fieldStart(false); ASSERT_TRUE(logger.hasError()); logger.reset(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.data("f3"); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldEnd(); - tracker.expect(2, 0, 3, 2, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 0, 3, 2, 2); // scc, ec, fsc, fec, dc, sac, stc, etc s.fieldEnd(); - tracker.expect(2, 1, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 1, 3, 3, 2); // scc, ec, fsc, fec, dc, sac, stc, etc } - tracker.expect(2, 2, 3, 3, 0, 0, 2); // sc, ec, fsc, fse, asc, aec, dc + tracker.expect(2, 2, 3, 3, 2); // scc, ec, fsc, fec, dc, sac, stc, etc ASSERT_FALSE(logger.hasError()); -}*/ +} + +TEST(Stack, rangeCommandUnranged) +{ + tracker.reset(); + logger.reset(); + + { + Stack s{parser, env.context, States::AnyHandlers}; + tracker.expect(0, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.commandStart("a", {}, true); + tracker.expect(1, 0, 0, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.commandStart("b", {}); + tracker.expect(2, 0, 1, 0, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + s.rangeEnd(); + tracker.expect(2, 2, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + } + tracker.expect(2, 2, 1, 1, 0); // scc, ec, fsc, fec, dc, sac, stc, etc + ASSERT_FALSE(logger.hasError()); +} + } } -- cgit v1.2.3