DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Mercurial (7fd12a78625c)

VCS Links

Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180
/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#ifndef __nsCharSeparatedTokenizer_h
#define __nsCharSeparatedTokenizer_h

#include "mozilla/RangedPtr.h"

#include "nsDependentSubstring.h"
#include "nsCRT.h"

/**
 * This parses a SeparatorChar-separated string into tokens.
 * Whitespace surrounding tokens is not treated as part of tokens, however
 * whitespace inside a token is. If the final token is the empty string, it is
 * not returned.
 *
 * Some examples, with SeparatorChar = ',':
 *
 * "foo, bar, baz" ->      "foo" "bar" "baz"
 * "foo,bar,baz" ->        "foo" "bar" "baz"
 * "foo , bar hi , baz" -> "foo" "bar hi" "baz"
 * "foo, ,bar,baz" ->      "foo" "" "bar" "baz"
 * "foo,,bar,baz" ->       "foo" "" "bar" "baz"
 * "foo,bar,baz," ->       "foo" "bar" "baz"
 *
 * The function used for whitespace detection is a template argument.
 * By default, it is NS_IsAsciiWhitespace.
 */
template <typename DependentSubstringType, bool IsWhitespace(char16_t)>
class nsTCharSeparatedTokenizer {
  typedef typename DependentSubstringType::char_type CharType;
  typedef typename DependentSubstringType::substring_type SubstringType;

 public:
  // Flags -- only one for now. If we need more, they should be defined to
  // be 1 << 1, 1 << 2, etc. (They're masks, and aFlags is a bitfield.)
  enum { SEPARATOR_OPTIONAL = 1 };

  nsTCharSeparatedTokenizer(const SubstringType& aSource,
                            CharType aSeparatorChar, uint32_t aFlags = 0)
      : mIter(aSource.Data(), aSource.Length()),
        mEnd(aSource.Data() + aSource.Length(), aSource.Data(),
             aSource.Length()),
        mSeparatorChar(aSeparatorChar),
        mWhitespaceBeforeFirstToken(false),
        mWhitespaceAfterCurrentToken(false),
        mSeparatorAfterCurrentToken(false),
        mSeparatorOptional(aFlags & SEPARATOR_OPTIONAL) {
    // Skip initial whitespace
    while (mIter < mEnd && IsWhitespace(*mIter)) {
      mWhitespaceBeforeFirstToken = true;
      ++mIter;
    }
  }

  /**
   * Checks if any more tokens are available.
   */
  bool hasMoreTokens() const {
    MOZ_ASSERT(mIter == mEnd || !IsWhitespace(*mIter),
               "Should be at beginning of token if there is one");

    return mIter < mEnd;
  }

  /*
   * Returns true if there is whitespace prior to the first token.
   */
  bool whitespaceBeforeFirstToken() const {
    return mWhitespaceBeforeFirstToken;
  }

  /*
   * Returns true if there is a separator after the current token.
   * Useful if you want to check whether the last token has a separator
   * after it which may not be valid.
   */
  bool separatorAfterCurrentToken() const {
    return mSeparatorAfterCurrentToken;
  }

  /*
   * Returns true if there is any whitespace after the current token.
   */
  bool whitespaceAfterCurrentToken() const {
    return mWhitespaceAfterCurrentToken;
  }

  /**
   * Returns the next token.
   */
  const DependentSubstringType nextToken() {
    mozilla::RangedPtr<const CharType> tokenStart = mIter;
    mozilla::RangedPtr<const CharType> tokenEnd = mIter;

    MOZ_ASSERT(mIter == mEnd || !IsWhitespace(*mIter),
               "Should be at beginning of token if there is one");

    // Search until we hit separator or end (or whitespace, if a separator
    // isn't required -- see clause with 'break' below).
    while (mIter < mEnd && *mIter != mSeparatorChar) {
      // Skip to end of the current word.
      while (mIter < mEnd && !IsWhitespace(*mIter) &&
             *mIter != mSeparatorChar) {
        ++mIter;
      }
      tokenEnd = mIter;

      // Skip whitespace after the current word.
      mWhitespaceAfterCurrentToken = false;
      while (mIter < mEnd && IsWhitespace(*mIter)) {
        mWhitespaceAfterCurrentToken = true;
        ++mIter;
      }
      if (mSeparatorOptional) {
        // We've hit (and skipped) whitespace, and that's sufficient to end
        // our token, regardless of whether we've reached a SeparatorChar.
        break;
      }  // (else, we'll keep looping until we hit mEnd or SeparatorChar)
    }

    mSeparatorAfterCurrentToken = (mIter != mEnd && *mIter == mSeparatorChar);
    MOZ_ASSERT(
        mSeparatorOptional || (mSeparatorAfterCurrentToken == (mIter < mEnd)),
        "If we require a separator and haven't hit the end of "
        "our string, then we shouldn't have left the loop "
        "unless we hit a separator");

    // Skip separator (and any whitespace after it), if we're at one.
    if (mSeparatorAfterCurrentToken) {
      ++mIter;

      while (mIter < mEnd && IsWhitespace(*mIter)) {
        mWhitespaceAfterCurrentToken = true;
        ++mIter;
      }
    }

    return Substring(tokenStart.get(), tokenEnd.get());
  }

 private:
  mozilla::RangedPtr<const CharType> mIter;
  const mozilla::RangedPtr<const CharType> mEnd;
  CharType mSeparatorChar;
  bool mWhitespaceBeforeFirstToken;
  bool mWhitespaceAfterCurrentToken;
  bool mSeparatorAfterCurrentToken;
  bool mSeparatorOptional;
};

template <bool IsWhitespace(char16_t) = NS_IsAsciiWhitespace>
class nsCharSeparatedTokenizerTemplate
    : public nsTCharSeparatedTokenizer<nsDependentSubstring, IsWhitespace> {
 public:
  nsCharSeparatedTokenizerTemplate(const nsAString& aSource,
                                   char16_t aSeparatorChar, uint32_t aFlags = 0)
      : nsTCharSeparatedTokenizer<nsDependentSubstring, IsWhitespace>(
            aSource, aSeparatorChar, aFlags) {}
};

typedef nsCharSeparatedTokenizerTemplate<> nsCharSeparatedTokenizer;

template <bool IsWhitespace(char16_t) = NS_IsAsciiWhitespace>
class nsCCharSeparatedTokenizerTemplate
    : public nsTCharSeparatedTokenizer<nsDependentCSubstring, IsWhitespace> {
 public:
  nsCCharSeparatedTokenizerTemplate(const nsACString& aSource,
                                    char aSeparatorChar, uint32_t aFlags = 0)
      : nsTCharSeparatedTokenizer<nsDependentCSubstring, IsWhitespace>(
            aSource, aSeparatorChar, aFlags) {}
};

typedef nsCCharSeparatedTokenizerTemplate<> nsCCharSeparatedTokenizer;

#endif /* __nsCharSeparatedTokenizer_h */