1/*
2 * Copyright (C) 2013 Google Inc. All rights reserved.
3 *
4 * Redistribution and use in source and binary forms, with or without
5 * modification, are permitted provided that the following conditions are
6 * met:
7 *
8 *     * Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 *     * Redistributions in binary form must reproduce the above
11 * copyright notice, this list of conditions and the following disclaimer
12 * in the documentation and/or other materials provided with the
13 * distribution.
14 *     * Neither the name of Google Inc. nor the names of its
15 * contributors may be used to endorse or promote products derived from
16 * this software without specific prior written permission.
17 *
18 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
19 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
20 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
21 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
22 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
23 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
24 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
25 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
26 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
27 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
28 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 */
30
31#ifndef BidiTestHarness_h
32#define BidiTestHarness_h
33
34#include <istream>
35#include <map>
36#include <stdio.h>
37#include <string>
38#include <vector>
39
40// FIXME: We don't have any business owning this code. We should try to
41// upstream this to unicode.org if possible (for other implementations to use).
42// Unicode.org provides a reference implmentation, including parser:
43// http://www.unicode.org/Public/PROGRAMS/BidiReferenceC/6.3.0/source/brtest.c
44// But it, like the other implementations I've found, is rather tied to
45// the algorithms it is testing. This file seeks to only implement the parser bits.
46
47// Other C/C++ implementations of this parser:
48// https://github.com/googlei18n/fribidi-vs-unicode/blob/master/test.c
49// http://source.icu-project.org/repos/icu/icu/trunk/source/test/intltest/bidiconf.cpp
50// Both of those are too tied to their respective projects to be use to Blink.
51
52// There are non-C implmentations to parse BidiTest.txt as well, including:
53// https://github.com/twitter/twitter-cldr-rb/blob/master/spec/bidi/bidi_spec.rb
54
55// NOTE: None of this file is currently written to be thread-safe.
56
57namespace bidi_test {
58
59enum ParagraphDirection {
60    DirectionAutoLTR = 1,
61    DirectionLTR = 2,
62    DirectionRTL = 4,
63};
64const int kMaxParagraphDirection = DirectionAutoLTR | DirectionLTR | DirectionRTL;
65
66// For error printing:
67std::string nameFromParagraphDirection(ParagraphDirection paragraphDirection)
68{
69    switch (paragraphDirection) {
70    case bidi_test::DirectionAutoLTR:
71        return "Auto-LTR";
72    case bidi_test::DirectionLTR:
73        return "LTR";
74    case bidi_test::DirectionRTL:
75        return "RTL";
76    }
77    // This should never be reached.
78    return "";
79}
80
81template<class Runner>
82class Harness {
83public:
84    Harness(Runner& runner)
85        : m_runner(runner)
86    {
87    }
88    void parse(std::istream& bidiTestFile);
89
90private:
91    Runner& m_runner;
92};
93
94// We could use boost::trim, but no other part of Blink uses boost yet.
95inline void ltrim(std::string& s)
96{
97    static const std::string separators(" \t");
98    s.erase(0, s.find_first_not_of(separators));
99}
100
101inline void rtrim(std::string& s)
102{
103    static const std::string separators(" \t");
104    size_t lastNonSpace = s.find_last_not_of(separators);
105    if (lastNonSpace == std::string::npos) {
106        s.erase();
107        return;
108    }
109    size_t firstSpaceAtEndOfString = lastNonSpace + 1;
110    if (firstSpaceAtEndOfString >= s.size())
111        return; // lastNonSpace was the last char.
112    s.erase(firstSpaceAtEndOfString, std::string::npos); // erase to the end of the string.
113}
114
115inline void trim(std::string& s)
116{
117    rtrim(s);
118    ltrim(s);
119}
120
121static std::vector<std::string> parseStringList(const std::string& str)
122{
123    std::vector<std::string> strings;
124    static const std::string separators(" \t");
125    size_t lastPos = str.find_first_not_of(separators); // skip leading spaces
126    size_t pos = str.find_first_of(separators, lastPos); // find next space
127
128    while (std::string::npos != pos || std::string::npos != lastPos) {
129        strings.push_back(str.substr(lastPos, pos - lastPos));
130        lastPos = str.find_first_not_of(separators, pos);
131        pos = str.find_first_of(separators, lastPos);
132    }
133    return strings;
134}
135
136static std::vector<int> parseIntList(const std::string& str)
137{
138    std::vector<int> ints;
139    std::vector<std::string> strings = parseStringList(str);
140    for (size_t x = 0; x < strings.size(); x++) {
141        int i = atoi(strings[x].c_str());
142        ints.push_back(i);
143    }
144    return ints;
145}
146
147static std::vector<int> parseLevels(const std::string& line)
148{
149    std::vector<int> levels;
150    std::vector<std::string> strings = parseStringList(line);
151    for (size_t x = 0; x < strings.size(); x++) {
152        const std::string& levelString = strings[x];
153        int i;
154        if (levelString == "x")
155            i = -1;
156        else
157            i = atoi(levelString.c_str());
158        levels.push_back(i);
159    }
160    return levels;
161}
162
163// This is not thread-safe as written.
164static std::basic_string<UChar> parseTestString(const std::string& line)
165{
166    std::basic_string<UChar> testString;
167    static std::map<std::string, UChar> charClassExamples;
168    if (charClassExamples.empty()) {
169        // FIXME: Explicit make_pair is ugly, but required for C++98 compat.
170        charClassExamples.insert(std::make_pair("L", 0x6c)); // 'l' for L
171        charClassExamples.insert(std::make_pair("R", 0x05D0)); // HEBREW ALEF
172        charClassExamples.insert(std::make_pair("EN", 0x33)); // '3' for EN
173        charClassExamples.insert(std::make_pair("ES", 0x2d)); // '-' for ES
174        charClassExamples.insert(std::make_pair("ET", 0x25)); // '%' for ET
175        charClassExamples.insert(std::make_pair("AN", 0x0660)); // arabic 0
176        charClassExamples.insert(std::make_pair("CS", 0x2c)); // ',' for CS
177        charClassExamples.insert(std::make_pair("B", 0x0A)); // <control-000A>
178        charClassExamples.insert(std::make_pair("S", 0x09)); // <control-0009>
179        charClassExamples.insert(std::make_pair("WS", 0x20)); // ' ' for WS
180        charClassExamples.insert(std::make_pair("ON", 0x3d)); // '=' for ON
181        charClassExamples.insert(std::make_pair("NSM", 0x05BF)); // HEBREW POINT RAFE
182        charClassExamples.insert(std::make_pair("AL", 0x0608)); // ARABIC RAY
183        charClassExamples.insert(std::make_pair("BN", 0x00AD)); // SOFT HYPHEN
184        charClassExamples.insert(std::make_pair("LRE", 0x202A));
185        charClassExamples.insert(std::make_pair("RLE", 0x202B));
186        charClassExamples.insert(std::make_pair("PDF", 0x202C));
187        charClassExamples.insert(std::make_pair("LRO", 0x202D));
188        charClassExamples.insert(std::make_pair("RLO", 0x202E));
189        charClassExamples.insert(std::make_pair("LRI", 0x2066));
190        charClassExamples.insert(std::make_pair("RLI", 0x2067));
191        charClassExamples.insert(std::make_pair("FSI", 0x2068));
192        charClassExamples.insert(std::make_pair("PDI", 0x2069));
193    }
194
195    std::vector<std::string> charClasses = parseStringList(line);
196    for (size_t i = 0; i < charClasses.size(); i++) {
197        // FIXME: If the lookup failed we could return false for a parse error.
198        testString.push_back(charClassExamples.find(charClasses[i])->second);
199    }
200    return testString;
201}
202
203static bool parseParagraphDirectionMask(const std::string& line, int& modeMask)
204{
205    modeMask = atoi(line.c_str());
206    return modeMask >= 1 && modeMask <= kMaxParagraphDirection;
207}
208
209static void parseError(const std::string& line, size_t lineNumber)
210{
211    // Use printf to avoid the expense of std::cout.
212    printf("Parse error, line %zu : %s\n", lineNumber, line.c_str());
213}
214
215template<class Runner>
216void Harness<Runner>::parse(std::istream& bidiTestFile)
217{
218    static const std::string levelsPrefix("@Levels");
219    static const std::string reorderPrefix("@Reorder");
220
221    // FIXME: UChar is an ICU type and cheating a bit to use here.
222    // uint16_t might be more portable.
223    std::basic_string<UChar> testString;
224    std::vector<int> levels;
225    std::vector<int> reorder;
226    int paragraphDirectionMask;
227
228    std::string line;
229    size_t lineNumber = 0;
230    while (std::getline(bidiTestFile, line)) {
231        lineNumber++;
232        const std::string originalLine = line;
233        size_t commentStart = line.find_first_of('#');
234        if (commentStart != std::string::npos)
235            line = line.substr(0, commentStart);
236        trim(line);
237        if (line.empty())
238            continue;
239        if (line[0] == '@') {
240            if (!line.find(levelsPrefix)) {
241                levels = parseLevels(line.substr(levelsPrefix.length() + 1));
242                continue;
243            }
244            if (!line.find(reorderPrefix)) {
245                reorder = parseIntList(line.substr(reorderPrefix.length() + 1));
246                continue;
247            }
248        } else {
249            // Assume it's a data line.
250            size_t seperatorIndex = line.find_first_of(';');
251            if (seperatorIndex == std::string::npos) {
252                parseError(originalLine, lineNumber);
253                continue;
254            }
255            testString = parseTestString(line.substr(0, seperatorIndex));
256            if (!parseParagraphDirectionMask(line.substr(seperatorIndex + 1), paragraphDirectionMask)) {
257                parseError(originalLine, lineNumber);
258                continue;
259            }
260
261            if (paragraphDirectionMask & DirectionAutoLTR)
262                m_runner.runTest(testString, reorder, levels, DirectionAutoLTR, originalLine, lineNumber);
263            if (paragraphDirectionMask & DirectionLTR)
264                m_runner.runTest(testString, reorder, levels, DirectionLTR, originalLine, lineNumber);
265            if (paragraphDirectionMask & DirectionRTL)
266                m_runner.runTest(testString, reorder, levels, DirectionRTL, originalLine, lineNumber);
267        }
268    }
269}
270
271} // namespace bidi_test
272
273#endif // BidiTestHarness_h
274