inspector/ContentSearchUtilities.cpp

   1 /*
   2  * Copyright (C) 2011 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  * 1. Redistributions of source code must retain the above copyright
   9  * notice, this list of conditions and the following disclaimer.
  10  *
  11  * 2. Redistributions in binary form must reproduce the above
  12  * copyright notice, this list of conditions and the following disclaimer
  13  * in the documentation and/or other materials provided with the
  14  * distribution.
  15  *
  16  * THIS SOFTWARE IS PROVIDED BY GOOGLE INC. AND ITS CONTRIBUTORS
  17  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  18  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  19  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GOOGLE INC.
  20  * OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  21  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  22  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  23  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  24  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  25  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  26  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  27  */
  28
  29 #include "config.h"
  30 #include "ContentSearchUtilities.h"
  31
  32 #include "InspectorValues.h"
  33 #include "RegularExpression.h"
  34 #include "Yarr.h"
  35 #include <wtf/BumpPointerAllocator.h>
  36 #include <wtf/StdLibExtras.h>
  37 #include <wtf/text/StringBuilder.h>
  38
  39 using namespace JSC::Yarr;
  40
  41 namespace Inspector {
  42 namespace ContentSearchUtilities {
  43
  44 static const char regexSpecialCharacters[] = "[](){}+-*.,?\\^$|";
  45
  46 static String createSearchRegexSource(const String& text)
  47 {
  48     StringBuilder result;
  49
  50     for (unsigned i = 0; i < text.length(); i++) {
  51         UChar character = text[i];
  52         if (isASCII(character) && strchr(regexSpecialCharacters, character))
  53             result.append('\\');
  54         result.append(character);
  55     }
  56
  57     return result.toString();
  58 }
  59
  60 static inline size_t sizetExtractor(const size_t* value)
  61 {
  62     return *value;
  63 }
  64
  65 TextPosition textPositionFromOffset(size_t offset, const Vector<size_t>& lineEndings)
  66 {
  67     const size_t* foundNextStart = approximateBinarySearch<size_t, size_t>(lineEndings, lineEndings.size(), offset, sizetExtractor);
  68     size_t lineIndex = foundNextStart - &lineEndings.at(0);
  69     if (offset >= *foundNextStart)
  70         ++lineIndex;
  71     size_t lineStartOffset = lineIndex > 0 ? lineEndings.at(lineIndex - 1) : 0;
  72     size_t column = offset - lineStartOffset;
  73     return TextPosition(OrdinalNumber::fromZeroBasedInt(lineIndex), OrdinalNumber::fromZeroBasedInt(column));
  74 }
  75
  76 static Vector<std::pair<size_t, String>> getRegularExpressionMatchesByLines(const JSC::Yarr::RegularExpression& regex, const String& text)
  77 {
  78     Vector<std::pair<size_t, String>> result;
  79     if (text.isEmpty())
  80         return result;
  81
  82     std::unique_ptr<Vector<size_t>> endings(lineEndings(text));
  83     size_t size = endings->size();
  84     size_t start = 0;
  85
  86     for (size_t lineNumber = 0; lineNumber < size; ++lineNumber) {
  87         size_t nextStart = endings->at(lineNumber);
  88         String line = text.substring(start, nextStart - start);
  89
  90         int matchLength;
  91         if (regex.match(line, 0, &matchLength) != -1)
  92             result.append(std::pair<size_t, String>(lineNumber, line));
  93
  94         start = nextStart;
  95     }
  96
  97     return result;
  98 }
  99
 100 std::unique_ptr<Vector<size_t>> lineEndings(const String& text)
 101 {
 102     auto result = std::make_unique<Vector<size_t>>();
 103
 104     size_t start = 0;
 105     while (start < text.length()) {
 106         size_t nextStart = text.findNextLineStart(start);
 107         if (nextStart == notFound) {
 108             result->append(text.length());
 109             break;
 110         }
 111
 112         result->append(nextStart);
 113         start = nextStart;
 114     }
 115
 116     result->append(text.length());
 117
 118     return result;
 119 }
 120
 121 static Ref<Inspector::Protocol::GenericTypes::SearchMatch> buildObjectForSearchMatch(size_t lineNumber, const String& lineContent)
 122 {
 123     return Inspector::Protocol::GenericTypes::SearchMatch::create()
 124         .setLineNumber(lineNumber)
 125         .setLineContent(lineContent)
 126         .release();
 127 }
 128
 129 JSC::Yarr::RegularExpression createSearchRegex(const String& query, bool caseSensitive, bool isRegex)
 130 {
 131     String regexSource = isRegex ? query : createSearchRegexSource(query);
 132     return JSC::Yarr::RegularExpression(regexSource, caseSensitive ? TextCaseSensitive : TextCaseInsensitive);
 133 }
 134
 135 int countRegularExpressionMatches(const JSC::Yarr::RegularExpression& regex, const String& content)
 136 {
 137     if (content.isEmpty())
 138         return 0;
 139
 140     int result = 0;
 141     int position;
 142     unsigned start = 0;
 143     int matchLength;
 144     while ((position = regex.match(content, start, &matchLength)) != -1) {
 145         if (start >= content.length())
 146             break;
 147         if (matchLength > 0)
 148             ++result;
 149         start = position + 1;
 150     }
 151     return result;
 152 }
 153
 154 Ref<Inspector::Protocol::Array<Inspector::Protocol::GenericTypes::SearchMatch>> searchInTextByLines(const String& text, const String& query, const bool caseSensitive, const bool isRegex)
 155 {
 156     Ref<Inspector::Protocol::Array<Inspector::Protocol::GenericTypes::SearchMatch>> result = Inspector::Protocol::Array<Inspector::Protocol::GenericTypes::SearchMatch>::create();
 157
 158     JSC::Yarr::RegularExpression regex = ContentSearchUtilities::createSearchRegex(query, caseSensitive, isRegex);
 159     Vector<std::pair<size_t, String>> matches = getRegularExpressionMatchesByLines(regex, text);
 160
 161     for (const auto& match : matches) {
 162         Ref<Inspector::Protocol::GenericTypes::SearchMatch> matchObject = buildObjectForSearchMatch(match.first, match.second);
 163         result->addItem(WTF::move(matchObject));
 164     }
 165
 166     return result;
 167 }
 168
 169 static String scriptCommentPattern(const String& name)
 170 {
 171     // "//# <name>=<value>" and deprecated "//@"
 172     return "//[#@][\040\t]" + name + "=[\040\t]*([^\\s\'\"]*)[\040\t]*$";
 173 }
 174
 175 static String stylesheetCommentPattern(const String& name)
 176 {
 177     // "/*# <name>=<value> */" and deprecated "/*@"
 178     return "/\\*[#@][\040\t]" + name + "=[\040\t]*([^\\s\'\"]*)[\040\t]*\\*/";
 179 }
 180
 181 static String findMagicComment(const String& content, const String& patternString)
 182 {
 183     ASSERT(!content.isNull());
 184     const char* error = nullptr;
 185     JSC::Yarr::YarrPattern pattern(patternString, false, true, &error);
 186     ASSERT(!error);
 187     BumpPointerAllocator regexAllocator;
 188     auto bytecodePattern = JSC::Yarr::byteCompile(pattern, &regexAllocator);
 189     ASSERT(bytecodePattern);
 190
 191     ASSERT(pattern.m_numSubpatterns == 1);
 192     Vector<int, 4> matches;
 193     matches.resize(4);
 194     unsigned result = JSC::Yarr::interpret(bytecodePattern.get(), content, 0, reinterpret_cast<unsigned*>(matches.data()));
 195     if (result == JSC::Yarr::offsetNoMatch)
 196         return String();
 197
 198     ASSERT(matches[2] > 0 && matches[3] > 0);
 199     return content.substring(matches[2], matches[3] - matches[2]);
 200 }
 201
 202 String findScriptSourceURL(const String& content)
 203 {
 204     return findMagicComment(content, scriptCommentPattern(ASCIILiteral("sourceURL")));
 205 }
 206
 207 String findScriptSourceMapURL(const String& content)
 208 {
 209     return findMagicComment(content, scriptCommentPattern(ASCIILiteral("sourceMappingURL")));
 210 }
 211
 212 String findStylesheetSourceMapURL(const String& content)
 213 {
 214     return findMagicComment(content, stylesheetCommentPattern(ASCIILiteral("sourceMappingURL")));
 215 }
 216
 217 } // namespace ContentSearchUtilities
 218 } // namespace Inspector