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 --- src/core/common/Utils.hpp | 53 ++++++++++++++++++++++++++++------------------- 1 file changed, 32 insertions(+), 21 deletions(-) (limited to 'src/core/common/Utils.hpp') diff --git a/src/core/common/Utils.hpp b/src/core/common/Utils.hpp index 7d96562..82a8f8c 100644 --- a/src/core/common/Utils.hpp +++ b/src/core/common/Utils.hpp @@ -123,14 +123,6 @@ public: */ static bool hasNonWhitepaceChar(const std::string &s); - /** - * Removes whitespace at the beginning and the end of the given string. - * - * @param s is the string that should be trimmed. - * @return a trimmed copy of s. - */ - static std::string trim(const std::string &s); - /** * Trims the given string or vector of chars by returning the start and end * index. @@ -153,8 +145,8 @@ public: * * @param s is the container that should be trimmed. * @param len is the number of elements in the container. - * @param f is a function that returns true for values that should be - * removed. + * @param f is a function that returns true for values at a certain index + * that should be removed. * @return start and end index. Note that "end" points at the character * beyond the end, thus "end" minus "start" */ @@ -163,7 +155,7 @@ public: { size_t start = 0; for (size_t i = 0; i < len; i++) { - if (!f(s[i])) { + if (!f(i)) { start = i; break; } @@ -171,7 +163,7 @@ public: size_t end = 0; for (ssize_t i = len - 1; i >= static_cast(start); i--) { - if (!f(s[i])) { + if (!f(i)) { end = i + 1; break; } @@ -198,16 +190,32 @@ public: * the collapsed version of the string ends. * @return start and end index. Note that "end" points at the character * beyond the end, thus "end" minus "start" + * @param f is a function that returns true for values at a certain index + * that should be removed. */ - template - static std::string trim(const T &s, size_t len, size_t &start, size_t &end) + template + static std::string trim(const T &s, size_t len, size_t &start, size_t &end, + Filter f) { - auto res = trim(s, len, isWhitespace); + auto res = trim(s, len, f); start = res.first; end = res.second; return std::string(&s[start], end - start); } + /** + * Removes whitespace at the beginning and the end of the given string. + * + * @param s is the string that should be trimmed. + * @return a trimmed copy of s. + */ + static std::string trim(const std::string &s) + { + std::pair bounds = + trim(s, [&s](size_t i) { return isWhitespace(s[i]); }); + return s.substr(bounds.first, bounds.second - bounds.first); + } + /** * Collapses the whitespaces in the given string (trims the string and * replaces all whitespace characters by a single one). @@ -219,7 +227,8 @@ public: { size_t start; size_t end; - return collapse(s, s.size(), start, end); + return collapse(s, s.size(), start, end, + [&s](size_t i) { return isWhitespace(s[i]); }); } /** @@ -236,7 +245,8 @@ public: static std::string collapse(const std::string &s, size_t &start, size_t &end) { - return collapse(s, s.size(), start, end); + return collapse(s, s.size(), start, end, + [&s](size_t i) { return isWhitespace(s[i]); }); } /** @@ -244,6 +254,8 @@ public: * replaces all whitespace characters by a single one). * * @tparam T is the string type that should be used. + * @tparam Filter is a filter function used for detecting the character + * indices that might be removed. * @param s is the string in which the whitespace should be collapsed. * @param len is the length of the input string * @param start is an output parameter which is set to the offset at which @@ -252,9 +264,9 @@ public: * the collapsed version of the string ends. * @return a copy of s with collapsed whitespace. */ - template + template static std::string collapse(const T &s, size_t len, size_t &start, - size_t &end) + size_t &end, Filter f) { // Result vector std::vector res; @@ -268,8 +280,7 @@ public: bool hadWhitespace = false; for (size_t i = 0; i < len; i++) { const char c = s[i]; - const bool whitespace = isWhitespace(c); - if (whitespace) { + if (f(i)) { hadWhitespace = !res.empty(); } else { // Adapt the start and end position -- 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 'src/core/common/Utils.hpp') 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 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 'src/core/common/Utils.hpp') 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