Add harness for running Unicode.org Bidi tests.

Source/platform/text/BidiTestHarness.h

+/*
+ * Copyright (C) 2013 Google, Inc. All Rights Reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY

[leviw_travelin_and_unemployed, 2013/10/28 21:33:11]

Nit: We don't need to use the "Provided by Apple" license.

+ * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+ * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+ * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+ * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+ * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
+ * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef BidiTestHarness_h
+#define BidiTestHarness_h
+
+#include <fstream>
+#include <iostream>
+#include <map>
+#include <sstream>
+#include <string>
+
+// Unicode.org provides a reference implmentation, including parser:
+// http://www.unicode.org/Public/PROGRAMS/BidiReferenceC/6.3.0/source/brtest.c
+// I failed to notice that until after I had written this one.
+// This implementation was written in std:: c++ with the intention
+// of being upstreamed to unicode.org.
+// FIXME: We should probably replace this implementation with the one from unicode.org.
+
+// Other C/C++ implementations of this parser:
+// https://github.com/googlei18n/fribidi-vs-unicode/blob/master/test.c
+// http://source.icu-project.org/repos/icu/icu/trunk/source/test/intltest/bidiconf.cpp
+// Both of those are too tied to their respective projects to be use to Blink.
+
+// There are non-C implmentations to parse BidiTest.txt as well, including:
+// https://github.com/twitter/twitter-cldr-rb/blob/master/spec/bidi/bidi_spec.rb
+
+namespace bidi_test {
+
+enum ParagraphDirection {
+    DirectionAutoLTR = 1,
+    DirectionLTR = 2,
+    DirectionRTL = 4,
+};
+const int kMaxParagraphDirection = DirectionAutoLTR | DirectionLTR | DirectionRTL;
+
+// For error printing:
+std::string nameFromParagraphDirection(ParagraphDirection paragraphDirection)
+{
+    switch (paragraphDirection) {
+    case bidi_test::DirectionAutoLTR:
+        return "Auto-LTR";
+    case bidi_test::DirectionLTR:
+        return "LTR";
+    case bidi_test::DirectionRTL:
+        return "RTL";
+    }
+}
+
+template<class Runner>
+class Harness {
+public:
+    Harness(Runner& runner)
+        : m_runner(runner)
+    {
+    }
+    void parse(std::istream& bidiTestFile);
+
+private:
+    Runner& m_runner;
+};
+
+// This trim() example comes from:
+// http://stackoverflow.com/questions/216823/whats-the-best-way-to-trim-stdstring
+// We could use boost::trim, but no other part of Blink uses boost yet.
+
+inline std::string& ltrim(std::string& s)
+{
+    s.erase(s.begin(), std::find_if(s.begin(), s.end(), std::not1(std::ptr_fun<int, int>(std::isspace))));

[leviw_travelin_and_unemployed, 2013/10/28 21:33:11]

jyasskin for this and the other std C++ stuff.

+    return s;
+}
+
+inline std::string& rtrim(std::string& s)
+{
+    s.erase(std::find_if(s.rbegin(), s.rend(), std::not1(std::ptr_fun<int, int>(std::isspace))).base(), s.end());
+    return s;
+}
+
+inline std::string& trim(std::string& s)
+{
+    return ltrim(rtrim(s));
+}
+
+static bool parseStringList(std::string str, std::vector<std::string>& strings)
+{
+    std::string separators(" \t");
+    strings.clear();
+    size_t lastPos = str.find_first_not_of(separators); // skip leading spaces
+    size_t pos = str.find_first_of(separators, lastPos); // find next space
+
+    while (std::string::npos != pos || std::string::npos != lastPos) {
+        strings.push_back(str.substr(lastPos, pos - lastPos));
+        lastPos = str.find_first_not_of(separators, pos);
+        pos = str.find_first_of(separators, lastPos);
+    }
+    return true;
+}
+
+static bool parseIntList(std::string str, std::vector<int>& ints)
+{
+    ints.clear();
+    std::vector<std::string> strings;
+    if (!parseStringList(str, strings))
+        return false;
+    for (int x = 0; x < strings.size(); x++) {
+        int i;
+        std::istringstream(strings[x]) >> i;
+        ints.push_back(i);
+    }
+    return true;
+}
+
+static bool parseLevelRule(std::string line, std::vector<int>& levels)
+{
+    levels.clear();
+    std::vector<std::string> strings;
+    if (!parseStringList(line, strings))
+        return false;
+    for (int x = 0; x < strings.size(); x++) {
+        const std::string& levelString = strings[x];
+        int i;
+        if (levelString == "x")
+            i = -1;
+        else
+            std::istringstream(levelString) >> i;
+        levels.push_back(i);
+    }
+    return true;}
+
+static bool parseReorderRule(std::string line, std::vector<int>& reorder)
+{
+    return parseIntList(line, reorder);
+}
+
+static bool parseTestString(std::string line, std::basic_string<UChar>& testString)
+{
+    testString.clear();
+    static std::map<std::string, UChar> charClassExamples;
+    if (charClassExamples.empty()) {
+        charClassExamples.insert({"L", 0x6c}); // 'l' for L
+        charClassExamples.insert({"R", 0x05D0}); // HEBREW ALEF
+        charClassExamples.insert({"EN", 0x33}); // '3' for EN
+        charClassExamples.insert({"ES", 0x2d}); // '-' for ES
+        charClassExamples.insert({"ET", 0x25}); // '%' for ET
+        charClassExamples.insert({"AN", 0x0660}); // arabic 0
+        charClassExamples.insert({"CS", 0x2c}); // ',' for CS
+        charClassExamples.insert({"B", 0x0A}); // <control-000A>
+        charClassExamples.insert({"S", 0x09}); // <control-0009>
+        charClassExamples.insert({"WS", 0x20}); // ' ' for WS
+        charClassExamples.insert({"ON", 0x3d}); // '=' for ON
+        charClassExamples.insert({"NSM", 0x05BF}); // HEBREW POINT RAFE
+        charClassExamples.insert({"AL", 0x0608}); // ARABIC RAY
+        charClassExamples.insert({"BN", 0x00AD}); // SOFT HYPHEN
+        charClassExamples.insert({"LRE", 0x202A});
+        charClassExamples.insert({"RLE", 0x202B});
+        charClassExamples.insert({"PDF", 0x202C});
+        charClassExamples.insert({"LRO", 0x202D});
+        charClassExamples.insert({"RLO", 0x202E});
+        charClassExamples.insert({"LRI", 0x2066});
+        charClassExamples.insert({"RLI", 0x2067});
+        charClassExamples.insert({"FSI", 0x2068});
+        charClassExamples.insert({"PDI", 0x2069});
+    }
+
+    std::vector<std::string> charClasses;
+    parseStringList(line, charClasses);
+    for (int i = 0; i < charClasses.size(); i++) {
+        testString.push_back(charClassExamples.find(charClasses[i])->second);
+    }
+    return true;
+}
+
+static bool parseParagraphDirectionMask(std::string line, int& modeMask)
+{
+    std::istringstream(line) >> modeMask;
+    return modeMask >= 1 && modeMask <= kMaxParagraphDirection;
+}
+
+static void parseError(const std::string& line, size_t lineNumber)
+{
+    std::cout << "Parse error, line " << lineNumber << ":" << line;
+}
+
+template<class Runner>
+void Harness<Runner>::parse(std::istream& bidiTestFile)
+{
+    std::string levelsPrefix("@Levels");
+    std::string reorderPrefix("@Reorder");
+
+    // FIXME: UChar is an ICU type and cheating a bit to use here.
+    // uint16_t might be more portable.
+    std::basic_string<UChar> testString;
+    std::vector<int> levels;
+    std::vector<int> reorder;
+    int paragraphDirectionMask;
+
+    std::string originalLine;
+    std::string line;
+    size_t lineNumber = 0;
+    while (std::getline(bidiTestFile, originalLine)) {
+        lineNumber++;
+        line = originalLine;
+        size_t commentStart = originalLine.find_first_of('#');
+        if (commentStart != std::string::npos)
+            line = line.substr(0, commentStart);
+        line = trim(line);
+        if (line.empty())
+            continue;
+        if (line[0] == '@') {
+            if (!line.find(levelsPrefix)) {
+                if (!parseLevelRule(line.substr(levelsPrefix.length() + 1), levels))
+                    parseError(originalLine, lineNumber);
+                continue;
+            }
+            if (!line.find(reorderPrefix)) {
+                if (!parseReorderRule(line.substr(reorderPrefix.length() + 1), reorder))
+                    parseError(originalLine, lineNumber);
+                continue;
+            }
+        } else {
+            // Assume it's a data line.
+            size_t seperatorIndex = line.find_first_of(';');
+            if (seperatorIndex == std::string::npos) {
+                parseError(originalLine, lineNumber);
+                continue;
+            }
+            if (!parseTestString(line.substr(0, seperatorIndex), testString)) {
+                parseError(originalLine, lineNumber);
+                continue;
+            }
+            if (!parseParagraphDirectionMask(line.substr(seperatorIndex + 1), paragraphDirectionMask)) {
+                parseError(originalLine, lineNumber);
+                continue;
+            }
+
+            if (paragraphDirectionMask & DirectionAutoLTR)
+                m_runner.runTest(testString, reorder, levels, DirectionAutoLTR, originalLine, lineNumber);
+            if (paragraphDirectionMask & DirectionLTR)
+                m_runner.runTest(testString, reorder, levels, DirectionLTR, originalLine, lineNumber);
+            if (paragraphDirectionMask & DirectionRTL)
+                m_runner.runTest(testString, reorder, levels, DirectionRTL, originalLine, lineNumber);
+        }
+    }
+}
+
+} // namespace bidi_test
+
+#endif // BidiTestHarness_h