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.

Header

Mercurial (5350524bb654)

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
/* -*- 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/. */

#include "mozilla/IncrementalTokenizer.h"

#include "mozilla/AutoRestore.h"

#include "nsIInputStream.h"
#include "IncrementalTokenizer.h"
#include <algorithm>

namespace mozilla {

IncrementalTokenizer::IncrementalTokenizer(Consumer aConsumer,
                                           const char * aWhitespaces,
                                           const char * aAdditionalWordChars,
                                           uint32_t aRawMinBuffered)
  : TokenizerBase(aWhitespaces, aAdditionalWordChars)
#ifdef DEBUG
  , mConsuming(false)
#endif
  , mNeedMoreInput(false)
  , mRollback(false)
  , mInputCursor(0)
  , mConsumer(aConsumer)
{
  mInputFinished = false;
  mMinRawDelivery = aRawMinBuffered;
}

nsresult IncrementalTokenizer::FeedInput(const nsACString & aInput)
{
  NS_ENSURE_TRUE(mConsumer, NS_ERROR_NOT_INITIALIZED);
  MOZ_ASSERT(!mInputFinished);

  mInput.Cut(0, mInputCursor);
  mInputCursor = 0;

  mInput.Append(aInput);

  return Process();
}

nsresult IncrementalTokenizer::FeedInput(nsIInputStream * aInput, uint32_t aCount)
{
  NS_ENSURE_TRUE(mConsumer, NS_ERROR_NOT_INITIALIZED);
  MOZ_ASSERT(!mInputFinished);
  MOZ_ASSERT(!mConsuming);

  mInput.Cut(0, mInputCursor);
  mInputCursor = 0;

  nsresult rv = NS_OK;
  while (NS_SUCCEEDED(rv) && aCount) {
    nsCString::index_type remainder = mInput.Length();
    nsCString::index_type load =
      std::min<nsCString::index_type>(aCount, PR_UINT32_MAX - remainder);

    if (!load) {
      // To keep the API simple, we fail if the input data buffer if filled.
      // It's highly unlikely there will ever be such amout of data cumulated
      // unless a logic fault in the consumer code.
      NS_ERROR("IncrementalTokenizer consumer not reading data?");
      return NS_ERROR_OUT_OF_MEMORY;
    }

    if (!mInput.SetLength(remainder + load, fallible)) {
      return NS_ERROR_OUT_OF_MEMORY;
    }

    nsCString::char_iterator buffer = mInput.BeginWriting() + remainder;

    uint32_t read;
    rv = aInput->Read(buffer, load, &read);
    if (NS_SUCCEEDED(rv)) {
      // remainder + load fits the uint32_t size, so must remainder + read.
      mInput.SetLength(remainder + read);
      aCount -= read;

      rv = Process();
    }
  }

  return rv;
}

nsresult IncrementalTokenizer::FinishInput()
{
  NS_ENSURE_TRUE(mConsumer, NS_ERROR_NOT_INITIALIZED);
  MOZ_ASSERT(!mInputFinished);
  MOZ_ASSERT(!mConsuming);

  mInput.Cut(0, mInputCursor);
  mInputCursor = 0;

  mInputFinished = true;
  nsresult rv = Process();
  mConsumer = nullptr;
  return rv;
}

bool IncrementalTokenizer::Next(Token & aToken)
{
  // Assert we are called only from the consumer callback
  MOZ_ASSERT(mConsuming);

  if (mPastEof) {
    return false;
  }

  nsACString::const_char_iterator next = Parse(aToken);
  mPastEof = aToken.Type() == TOKEN_EOF;
  if (next == mCursor && !mPastEof) {
    // Not enough input to make a deterministic decision.
    return false;
  }

  AssignFragment(aToken, mCursor, next);
  mCursor = next;
  return true;
}

void IncrementalTokenizer::NeedMoreInput()
{
  // Assert we are called only from the consumer callback
  MOZ_ASSERT(mConsuming);

  // When the input has been finished, we can't set the flag to prevent
  // indefinite wait for more input (that will never come)
  mNeedMoreInput = !mInputFinished;
}

void IncrementalTokenizer::Rollback()
{
  // Assert we are called only from the consumer callback
  MOZ_ASSERT(mConsuming);

  mRollback = true;
}

nsresult IncrementalTokenizer::Process()
{
#ifdef DEBUG
  // Assert we are not re-entered
  MOZ_ASSERT(!mConsuming);

  AutoRestore<bool> consuming(mConsuming);
  mConsuming = true;
#endif

  MOZ_ASSERT(!mPastEof);

  nsresult rv = NS_OK;

  mInput.BeginReading(mCursor);
  mCursor += mInputCursor;
  mInput.EndReading(mEnd);

  while (NS_SUCCEEDED(rv) && !mPastEof) {
    Token token;
    nsACString::const_char_iterator next = Parse(token);
    mPastEof = token.Type() == TOKEN_EOF;
    if (next == mCursor && !mPastEof) {
      // Not enough input to make a deterministic decision.
      break;
    }

    AssignFragment(token, mCursor, next);

    nsACString::const_char_iterator rollback = mCursor;
    mCursor = next;

    mNeedMoreInput = mRollback = false;

    rv = mConsumer(token, *this);
    if (NS_FAILED(rv)) {
      break;
    }
    if (mNeedMoreInput || mRollback) {
      mCursor = rollback;
      mPastEof = false;
      if (mNeedMoreInput) {
        break;
      }
    }
  }

  mInputCursor = mCursor - mInput.BeginReading();
  return rv;
}

} // mozilla