From 4854509f8add1e2ff167623fb0e8d4216d9d6023 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 8 Feb 2015 17:54:27 +0100 Subject: Implemented DynamicTokenizer and unit tests --- src/plugins/plain/DynamicTokenizer.hpp | 154 ++++++++++++++++++++++----------- 1 file changed, 105 insertions(+), 49 deletions(-) (limited to 'src/plugins/plain/DynamicTokenizer.hpp') diff --git a/src/plugins/plain/DynamicTokenizer.hpp b/src/plugins/plain/DynamicTokenizer.hpp index f7fef13..760bebf 100644 --- a/src/plugins/plain/DynamicTokenizer.hpp +++ b/src/plugins/plain/DynamicTokenizer.hpp @@ -28,34 +28,63 @@ #ifndef _OUSIA_DYNAMIC_TOKENIZER_HPP_ #define _OUSIA_DYNAMIC_TOKENIZER_HPP_ +#include +#include +#include + #include +#include "TokenTrie.hpp" + namespace ousia { // Forward declarations class CharReader; -class TokenDescriptor; /** * The DynamicToken structure describes a token discovered by the Tokenizer. */ struct DynamicToken { /** - * Pointer pointing at the TokenDescriptor instance this token corresponds - * to. May be one of the special TokenDescriptors defined as static members - * of the DynamicTokenizer class. + * Id of the type of this token. */ - TokenDescriptor const *descriptor; + TokenTypeId type; /** * String that was matched. */ - std::string str; + std::string content; /** * Location from which the string was extracted. */ SourceLocation location; + + /** + * Default constructor. + */ + DynamicToken() : type(EmptyToken) {} + + /** + * Constructor of the DynamicToken struct. + * + * @param id represents the token type. + * @param content is the string content that has been extracted. + * @param location is the location of the extracted string content in the + * source file. + */ + DynamicToken(TokenTypeId type, const std::string &content, + SourceLocation location) + : type(type), content(content), location(location) + { + } + + /** + * Constructor of the DynamicToken struct, only initializes the token type + * + * @param type is the id corresponding to the type of the token. + */ + DynamicToken(TokenTypeId type) : type(type) {} }; /** @@ -64,43 +93,70 @@ struct DynamicToken { */ enum class WhitespaceMode { /** - * Preserves all whitespaces as they are found in the source file. - */ + * Preserves all whitespaces as they are found in the source file. + */ PRESERVE, /** - * Trims whitespace at the beginning and the end of the found text. - */ + * Trims whitespace at the beginning and the end of the found text. + */ TRIM, /** - * Whitespaces are trimmed and collapsed, multiple whitespace characters - * are replaced by a single space character. - */ + * Whitespaces are trimmed and collapsed, multiple whitespace characters + * are replaced by a single space character. + */ COLLAPSE }; /** * The DynamicTokenizer is used to extract tokens and chunks of text from a * CharReader. It allows to register and unregister tokens while parsing and - * to modify the handling of whitespace characters. + * to modify the handling of whitespace characters. Note that the + * DynamicTokenizer always tries to extract the longest possible token from the + * tokenizer. */ class DynamicTokenizer { private: /** - * Reference at the char reader. + * CharReader instance from which the tokens should be read. */ CharReader &reader; + /** + * Internally used token trie. This object holds all registered tokens. + */ + TokenTrie trie; + /** * Flag defining whether whitespaces should be preserved or not. */ WhitespaceMode whitespaceMode; /** - * Vector containing all registered token descriptors. + * Vector containing all registered token types. + */ + std::vector tokens; + + /** + * Next index in the tokens list where to search for a new token id. */ - std::vector> descriptors; + size_t nextTokenTypeId; + + /** + * Templated function used internally to read the current token. The + * function is templated in order to force code generation for all six + * combiations of whitespace modes and reading/peeking. + * + * @tparam TextHandler is the type to be used for the textHandler instance. + * @tparam read specifies whether the function should start from and advance + * the read pointer of the char reader. + * @param token is the token structure into which the token information + * should be written. + * @return false if the end of the stream has been reached, true otherwise. + */ + template + bool next(DynamicToken &token); public: /** @@ -108,43 +164,44 @@ public: * * @param reader is the CharReader that should be used for reading the * tokens. - * @param preserveWhitespaces should be set to true if all whitespaces - * should be preserved (for preformated environments). - */ - DynamicTokenizer(CharReader &reader) - : reader(reader), - preserveWhitespaces(preserveWhitespaces), - location(reader.getSourceId()), - empty(true), - hasWhitespace(false) - { - } - - /** - * Destructor of the DynamicTokenizer class. + * @param whitespaceMode specifies how whitespace should be handled. */ - ~DynamicTokenizer(); + DynamicTokenizer(CharReader &reader, + WhitespaceMode whitespaceMode = WhitespaceMode::COLLAPSE); /** * Registers the given string as a token. Returns a const pointer at a * TokenDescriptor that will be used to reference the newly created token. * * @param token is the token string that should be registered. - * @return a pointer at a TokenDescriptor which is representative for the - * newly registered token. Returns nullptr if a token with this string - * was already registered. + * @return a unique identifier for the registered token or EmptyToken if + * an error occured. */ - const TokenDescriptor* registerToken(const std::string &token); + TokenTypeId registerToken(const std::string &token); /** - * Unregisters the token belonging to the given TokenDescriptor. + * Unregisters the token belonging to the given TokenTypeId. * - * @param descr is a TokenDescriptor that was previously returned by - * registerToken. + * @param type is the token type that should be unregistered. The + *TokenTypeId + * must have been returned by registerToken. * @return true if the operation was successful, false otherwise (e.g. * because the given TokenDescriptor was already unregistered). */ - bool unregisterToken(const TokenDescriptor *descr); + bool unregisterToken(TokenTypeId type); + + /** + * Returns the token that was registered under the given TokenTypeId id or + *an + * empty string if an invalid TokenTypeId id is given. + * + * @param type is the TokenTypeId id for which the corresponding token + *string + * should be returned. + * @return the registered token string or an empty string if the given type + * was invalid. + */ + std::string getTokenString(TokenTypeId type); /** * Sets the whitespace mode. @@ -173,17 +230,16 @@ public: bool read(DynamicToken &token); /** - * TokenDescriptor representing an empty token. - */ - static const *TokenDescriptor Empty; - - /** - * TokenDescriptor representing generic text. + * The peek method does not advance the read position of the char reader, + * but reads the next token from the current char reader peek position. + * + * @param token is a reference at the token instance into which the Token + * information should be written. + * @return true if a token could be read, false if the end of the stream + * has been reached. */ - static const *TokenDescriptor Text; - + bool peek(DynamicToken &token); }; - } #endif /* _OUSIA_DYNAMIC_TOKENIZER_HPP_ */ -- cgit v1.2.3 From f713b1d393230e7083727d457623fdac878eb248 Mon Sep 17 00:00:00 2001 From: Andreas Stöckel Date: Sun, 8 Feb 2015 18:48:07 +0100 Subject: DynamicTokenizer now gets the reader as a parameter to read and peek -- the beauty of this tokenizer is that it has no internal state depending on the reader, so it doesn't need to hold a reference to it --- src/plugins/plain/DynamicTokenizer.cpp | 35 ++++++------- src/plugins/plain/DynamicTokenizer.hpp | 22 ++++---- test/plugins/plain/DynamicTokenizerTest.cpp | 81 ++++++++++++++--------------- 3 files changed, 67 insertions(+), 71 deletions(-) (limited to 'src/plugins/plain/DynamicTokenizer.hpp') diff --git a/src/plugins/plain/DynamicTokenizer.cpp b/src/plugins/plain/DynamicTokenizer.cpp index a8f2317..f2cfcd1 100644 --- a/src/plugins/plain/DynamicTokenizer.cpp +++ b/src/plugins/plain/DynamicTokenizer.cpp @@ -345,14 +345,13 @@ public: /* Class DynamicTokenizer */ -DynamicTokenizer::DynamicTokenizer(CharReader &reader, - WhitespaceMode whitespaceMode) - : reader(reader), whitespaceMode(whitespaceMode), nextTokenTypeId(0) +DynamicTokenizer::DynamicTokenizer(WhitespaceMode whitespaceMode) + : whitespaceMode(whitespaceMode), nextTokenTypeId(0) { } template -bool DynamicTokenizer::next(DynamicToken &token) +bool DynamicTokenizer::next(CharReader &reader, DynamicToken &token) { // If we're in the read mode, reset the char reader peek position to the // current read position @@ -437,28 +436,28 @@ bool DynamicTokenizer::next(DynamicToken &token) return match.hasMatch(); } -bool DynamicTokenizer::read(DynamicToken &token) +bool DynamicTokenizer::read(CharReader &reader,DynamicToken &token) { switch (whitespaceMode) { case WhitespaceMode::PRESERVE: - return next(token); + return next(reader, token); case WhitespaceMode::TRIM: - return next(token); + return next(reader, token); case WhitespaceMode::COLLAPSE: - return next(token); + return next(reader, token); } return false; } -bool DynamicTokenizer::peek(DynamicToken &token) +bool DynamicTokenizer::peek(CharReader &reader,DynamicToken &token) { switch (whitespaceMode) { case WhitespaceMode::PRESERVE: - return next(token); + return next(reader, token); case WhitespaceMode::TRIM: - return next(token); + return next(reader, token); case WhitespaceMode::COLLAPSE: - return next(token); + return next(reader, token); } return false; } @@ -530,16 +529,16 @@ WhitespaceMode DynamicTokenizer::getWhitespaceMode() { return whitespaceMode; } /* Explicitly instantiate all possible instantiations of the "next" member function */ template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader, DynamicToken &token); template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader, DynamicToken &token); template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader,DynamicToken &token); template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader,DynamicToken &token); template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader,DynamicToken &token); template bool DynamicTokenizer::next( - DynamicToken &token); + CharReader &reader,DynamicToken &token); } diff --git a/src/plugins/plain/DynamicTokenizer.hpp b/src/plugins/plain/DynamicTokenizer.hpp index 760bebf..0b4dd39 100644 --- a/src/plugins/plain/DynamicTokenizer.hpp +++ b/src/plugins/plain/DynamicTokenizer.hpp @@ -118,11 +118,6 @@ enum class WhitespaceMode { */ class DynamicTokenizer { private: - /** - * CharReader instance from which the tokens should be read. - */ - CharReader &reader; - /** * Internally used token trie. This object holds all registered tokens. */ @@ -151,23 +146,22 @@ private: * @tparam TextHandler is the type to be used for the textHandler instance. * @tparam read specifies whether the function should start from and advance * the read pointer of the char reader. + * @param reader is the CharReader instance from which the data should be + * read. * @param token is the token structure into which the token information * should be written. * @return false if the end of the stream has been reached, true otherwise. */ template - bool next(DynamicToken &token); + bool next(CharReader &reader, DynamicToken &token); public: /** * Constructor of the DynamicTokenizer class. * - * @param reader is the CharReader that should be used for reading the - * tokens. * @param whitespaceMode specifies how whitespace should be handled. */ - DynamicTokenizer(CharReader &reader, - WhitespaceMode whitespaceMode = WhitespaceMode::COLLAPSE); + DynamicTokenizer(WhitespaceMode whitespaceMode = WhitespaceMode::COLLAPSE); /** * Registers the given string as a token. Returns a const pointer at a @@ -222,23 +216,27 @@ public: * Reads a new token from the CharReader and stores it in the given * DynamicToken instance. * + * @param reader is the CharReader instance from which the data should be + * read. * @param token is a reference at the token instance into which the Token * information should be written. * @return true if a token could be read, false if the end of the stream * has been reached. */ - bool read(DynamicToken &token); + bool read(CharReader &reader, DynamicToken &token); /** * The peek method does not advance the read position of the char reader, * but reads the next token from the current char reader peek position. * + * @param reader is the CharReader instance from which the data should be + * read. * @param token is a reference at the token instance into which the Token * information should be written. * @return true if a token could be read, false if the end of the stream * has been reached. */ - bool peek(DynamicToken &token); + bool peek(CharReader &reader, DynamicToken &token); }; } diff --git a/test/plugins/plain/DynamicTokenizerTest.cpp b/test/plugins/plain/DynamicTokenizerTest.cpp index 63fa466..5183fdd 100644 --- a/test/plugins/plain/DynamicTokenizerTest.cpp +++ b/test/plugins/plain/DynamicTokenizerTest.cpp @@ -25,8 +25,7 @@ namespace ousia { TEST(DynamicTokenizer, tokenRegistration) { - CharReader reader{"test"}; - DynamicTokenizer tokenizer{reader}; + DynamicTokenizer tokenizer; ASSERT_EQ(EmptyToken, tokenizer.registerToken("")); @@ -57,10 +56,10 @@ TEST(DynamicTokenizer, textTokenPreserveWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::PRESERVE}; + DynamicTokenizer tokenizer{WhitespaceMode::PRESERVE}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ(" this \t is only a \n\n test text ", token.content); @@ -68,17 +67,17 @@ TEST(DynamicTokenizer, textTokenPreserveWhitespace) ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(36U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::PRESERVE}; + DynamicTokenizer tokenizer{WhitespaceMode::PRESERVE}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("this \t is only a \n\n test text", token.content); @@ -86,7 +85,7 @@ TEST(DynamicTokenizer, textTokenPreserveWhitespace) ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(32U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } } @@ -96,10 +95,10 @@ TEST(DynamicTokenizer, textTokenTrimWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::TRIM}; + DynamicTokenizer tokenizer{WhitespaceMode::TRIM}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("this \t is only a \n\n test text", token.content); @@ -107,17 +106,17 @@ TEST(DynamicTokenizer, textTokenTrimWhitespace) ASSERT_EQ(1U, loc.getStart()); ASSERT_EQ(33U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::TRIM}; + DynamicTokenizer tokenizer{WhitespaceMode::TRIM}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("this \t is only a \n\n test text", token.content); @@ -125,7 +124,7 @@ TEST(DynamicTokenizer, textTokenTrimWhitespace) ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(32U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } } @@ -135,10 +134,10 @@ TEST(DynamicTokenizer, textTokenCollapseWhitespace) CharReader reader{" this \t is only a \n\n test text "}; // 012345 6789012345678 9 0123456789012345 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::COLLAPSE}; + DynamicTokenizer tokenizer{WhitespaceMode::COLLAPSE}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("this is only a test text", token.content); @@ -146,17 +145,17 @@ TEST(DynamicTokenizer, textTokenCollapseWhitespace) ASSERT_EQ(1U, loc.getStart()); ASSERT_EQ(33U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } { CharReader reader{"this \t is only a \n\n test text"}; // 01234 5678901234567 8 9012345678901 // 0 1 2 3 - DynamicTokenizer tokenizer{reader, WhitespaceMode::COLLAPSE}; + DynamicTokenizer tokenizer{WhitespaceMode::COLLAPSE}; DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("this is only a test text", token.content); @@ -164,21 +163,21 @@ TEST(DynamicTokenizer, textTokenCollapseWhitespace) ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(32U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } } TEST(DynamicTokenizer, simpleReadToken) { CharReader reader{"test1:test2"}; - DynamicTokenizer tokenizer{reader}; + DynamicTokenizer tokenizer; const TokenTypeId tid = tokenizer.registerToken(":"); ASSERT_EQ(0U, tid); { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test1", token.content); @@ -194,7 +193,7 @@ TEST(DynamicTokenizer, simpleReadToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(tid, token.type); ASSERT_EQ(":", token.content); @@ -210,7 +209,7 @@ TEST(DynamicTokenizer, simpleReadToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test2", token.content); @@ -227,14 +226,14 @@ TEST(DynamicTokenizer, simpleReadToken) TEST(DynamicTokenizer, simplePeekToken) { CharReader reader{"test1:test2"}; - DynamicTokenizer tokenizer{reader}; + DynamicTokenizer tokenizer; const TokenTypeId tid = tokenizer.registerToken(":"); ASSERT_EQ(0U, tid); { DynamicToken token; - ASSERT_TRUE(tokenizer.peek(token)); + ASSERT_TRUE(tokenizer.peek(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test1", token.content); @@ -248,7 +247,7 @@ TEST(DynamicTokenizer, simplePeekToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.peek(token)); + ASSERT_TRUE(tokenizer.peek(reader, token)); ASSERT_EQ(tid, token.type); ASSERT_EQ(":", token.content); @@ -262,7 +261,7 @@ TEST(DynamicTokenizer, simplePeekToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.peek(token)); + ASSERT_TRUE(tokenizer.peek(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test2", token.content); @@ -276,7 +275,7 @@ TEST(DynamicTokenizer, simplePeekToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test1", token.content); @@ -290,7 +289,7 @@ TEST(DynamicTokenizer, simplePeekToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(tid, token.type); ASSERT_EQ(":", token.content); @@ -304,7 +303,7 @@ TEST(DynamicTokenizer, simplePeekToken) { DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("test2", token.content); @@ -320,7 +319,7 @@ TEST(DynamicTokenizer, simplePeekToken) TEST(DynamicTokenizer, ambiguousTokens) { CharReader reader{"abc"}; - DynamicTokenizer tokenizer(reader); + DynamicTokenizer tokenizer; TokenTypeId t1 = tokenizer.registerToken("abd"); TokenTypeId t2 = tokenizer.registerToken("bc"); @@ -329,7 +328,7 @@ TEST(DynamicTokenizer, ambiguousTokens) ASSERT_EQ(1U, t2); DynamicToken token; - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(TextToken, token.type); ASSERT_EQ("a", token.content); @@ -338,7 +337,7 @@ TEST(DynamicTokenizer, ambiguousTokens) ASSERT_EQ(0U, loc.getStart()); ASSERT_EQ(1U, loc.getEnd()); - ASSERT_TRUE(tokenizer.read(token)); + ASSERT_TRUE(tokenizer.read(reader, token)); ASSERT_EQ(t2, token.type); ASSERT_EQ("bc", token.content); @@ -347,7 +346,7 @@ TEST(DynamicTokenizer, ambiguousTokens) ASSERT_EQ(1U, loc.getStart()); ASSERT_EQ(3U, loc.getEnd()); - ASSERT_FALSE(tokenizer.read(token)); + ASSERT_FALSE(tokenizer.read(reader, token)); } TEST(DynamicTokenizer, commentTestWhitespacePreserve) @@ -355,7 +354,7 @@ TEST(DynamicTokenizer, commentTestWhitespacePreserve) CharReader reader{"Test/Test /* Block Comment */", 0}; // 012345678901234567890123456789 // 0 1 2 - DynamicTokenizer tokenizer(reader, WhitespaceMode::PRESERVE); + DynamicTokenizer tokenizer(WhitespaceMode::PRESERVE); const TokenTypeId t1 = tokenizer.registerToken("/"); const TokenTypeId t2 = tokenizer.registerToken("/*"); @@ -371,14 +370,14 @@ TEST(DynamicTokenizer, commentTestWhitespacePreserve) DynamicToken t; for (auto &te : expected) { - EXPECT_TRUE(tokenizer.read(t)); + EXPECT_TRUE(tokenizer.read(reader, t)); EXPECT_EQ(te.type, t.type); 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(t)); + ASSERT_FALSE(tokenizer.read(reader, t)); } TEST(DynamicTokenizer, commentTestWhitespaceCollapse) @@ -386,7 +385,7 @@ TEST(DynamicTokenizer, commentTestWhitespaceCollapse) CharReader reader{"Test/Test /* Block Comment */", 0}; // 012345678901234567890123456789 // 0 1 2 - DynamicTokenizer tokenizer(reader, WhitespaceMode::COLLAPSE); + DynamicTokenizer tokenizer(WhitespaceMode::COLLAPSE); const TokenTypeId t1 = tokenizer.registerToken("/"); const TokenTypeId t2 = tokenizer.registerToken("/*"); @@ -402,14 +401,14 @@ TEST(DynamicTokenizer, commentTestWhitespaceCollapse) DynamicToken t; for (auto &te : expected) { - EXPECT_TRUE(tokenizer.read(t)); + EXPECT_TRUE(tokenizer.read(reader, t)); EXPECT_EQ(te.type, t.type); 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(t)); + ASSERT_FALSE(tokenizer.read(reader, t)); } } -- cgit v1.2.3