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.

Implementation

Mercurial (b6d82b1a6b02)

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319
/* -*- 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/. */

/*
 * A class which represents a fragment of text (eg inside a text
 * node); if only codepoints below 256 are used, the text is stored as
 * a char*; otherwise the text is stored as a char16_t*
 */

#ifndef nsTextFragment_h___
#define nsTextFragment_h___

#include "mozilla/Attributes.h"
#include "mozilla/MemoryReporting.h"

#include "nsCharTraits.h"
#include "nsString.h"
#include "nsStringBuffer.h"
#include "nsReadableUtils.h"
#include "nsISupportsImpl.h"

// XXX should this normalize the code to keep a \u0000 at the end?

// XXX nsTextFragmentPool?

/**
 * A fragment of text. If mIs2b is 1 then the m2b pointer is valid
 * otherwise the m1b pointer is valid. If m1b is used then each byte
 * of data represents a single ucs2 character with the high byte being
 * zero.
 *
 * This class does not have a virtual destructor therefore it is not
 * meant to be subclassed.
 */
class nsTextFragment final {
 public:
  static nsresult Init();
  static void Shutdown();

  /**
   * Default constructor. Initialize the fragment to be empty.
   */
  nsTextFragment() : m1b(nullptr), mAllBits(0) {
    MOZ_COUNT_CTOR(nsTextFragment);
    NS_ASSERTION(sizeof(FragmentBits) == 4, "Bad field packing!");
  }

  ~nsTextFragment();

  /**
   * Change the contents of this fragment to be a copy of the
   * the argument fragment, or to "" if unable to allocate enough memory.
   */
  nsTextFragment& operator=(const nsTextFragment& aOther);

  /**
   * Return true if this fragment is represented by char16_t data
   */
  bool Is2b() const { return mState.mIs2b; }

  /**
   * Return true if this fragment contains Bidi text
   * For performance reasons this flag is only set if explicitely requested (by
   * setting the aUpdateBidi argument on SetTo or Append to true).
   */
  bool IsBidi() const { return mState.mIsBidi; }

  /**
   * Get a pointer to constant char16_t data.
   */
  const char16_t* Get2b() const {
    MOZ_ASSERT(Is2b(), "not 2b text");
    return static_cast<char16_t*>(m2b->Data());
  }

  /**
   * Get a pointer to constant char data.
   */
  const char* Get1b() const {
    NS_ASSERTION(!Is2b(), "not 1b text");
    return (const char*)m1b;
  }

  /**
   * Get the length of the fragment. The length is the number of logical
   * characters, not the number of bytes to store the characters.
   */
  uint32_t GetLength() const { return mState.mLength; }

  bool CanGrowBy(size_t n) const {
    return n < (1 << 29) && mState.mLength + n < (1 << 29);
  }

  /**
   * Change the contents of this fragment to be a copy of the given
   * buffer. If aUpdateBidi is true, contents of the fragment will be scanned,
   * and mState.mIsBidi will be turned on if it includes any Bidi characters.
   * If aForce2b is true, aBuffer will be stored as char16_t as is.  Then,
   * you can access the value faster but may waste memory if all characters
   * are less than U+0100.
   */
  bool SetTo(const char16_t* aBuffer, int32_t aLength, bool aUpdateBidi,
             bool aForce2b);

  bool SetTo(const nsString& aString, bool aUpdateBidi, bool aForce2b) {
    ReleaseText();
    if (aForce2b && !aUpdateBidi) {
      nsStringBuffer* buffer = nsStringBuffer::FromString(aString);
      if (buffer) {
        NS_ADDREF(m2b = buffer);
        mState.mInHeap = true;
        mState.mIs2b = true;
        mState.mLength = aString.Length();
        return true;
      }
    }

    return SetTo(aString.get(), aString.Length(), aUpdateBidi, aForce2b);
  }

  /**
   * Append aData to the end of this fragment. If aUpdateBidi is true, contents
   * of the fragment will be scanned, and mState.mIsBidi will be turned on if
   * it includes any Bidi characters.
   * If aForce2b is true, the string will be stored as char16_t as is.  Then,
   * you can access the value faster but may waste memory if all characters
   * are less than U+0100.
   */
  bool Append(const char16_t* aBuffer, uint32_t aLength, bool aUpdateBidi,
              bool aForce2b);

  /**
   * Append the contents of this string fragment to aString
   */
  void AppendTo(nsAString& aString) const {
    if (!AppendTo(aString, mozilla::fallible)) {
      aString.AllocFailed(aString.Length() + GetLength());
    }
  }

  /**
   * Append the contents of this string fragment to aString
   * @return false if an out of memory condition is detected, true otherwise
   */
  MOZ_MUST_USE
  bool AppendTo(nsAString& aString,
                const mozilla::fallible_t& aFallible) const {
    if (mState.mIs2b) {
      if (aString.IsEmpty()) {
        m2b->ToString(mState.mLength, aString);
        return true;
      }
      bool ok = aString.Append(Get2b(), mState.mLength, aFallible);
      if (!ok) {
        return false;
      }

      return true;
    } else {
      return AppendASCIItoUTF16(Substring(m1b, mState.mLength), aString,
                                aFallible);
    }
  }

  /**
   * Append a substring of the contents of this string fragment to aString.
   * @param aOffset where to start the substring in this text fragment
   * @param aLength the length of the substring
   */
  void AppendTo(nsAString& aString, int32_t aOffset, int32_t aLength) const {
    if (!AppendTo(aString, aOffset, aLength, mozilla::fallible)) {
      aString.AllocFailed(aString.Length() + aLength);
    }
  }

  /**
   * Append a substring of the contents of this string fragment to aString.
   * @param aString the string in which to append
   * @param aOffset where to start the substring in this text fragment
   * @param aLength the length of the substring
   * @return false if an out of memory condition is detected, true otherwise
   */
  MOZ_MUST_USE
  bool AppendTo(nsAString& aString, int32_t aOffset, int32_t aLength,
                const mozilla::fallible_t& aFallible) const {
    if (mState.mIs2b) {
      bool ok = aString.Append(Get2b() + aOffset, aLength, aFallible);
      if (!ok) {
        return false;
      }

      return true;
    } else {
      return AppendASCIItoUTF16(Substring(m1b + aOffset, aLength), aString,
                                aFallible);
    }
  }

  /**
   * Make a copy of the fragments contents starting at offset for
   * count characters. The offset and count will be adjusted to
   * lie within the fragments data. The fragments data is converted if
   * necessary.
   */
  void CopyTo(char16_t* aDest, int32_t aOffset, int32_t aCount);

  /**
   * Return the character in the text-fragment at the given
   * index. This always returns a char16_t.
   */
  char16_t CharAt(int32_t aIndex) const {
    MOZ_ASSERT(uint32_t(aIndex) < mState.mLength, "bad index");
    return mState.mIs2b ? Get2b()[aIndex]
                        : static_cast<unsigned char>(m1b[aIndex]);
  }

  /**
   * IsHighSurrogateFollowedByLowSurrogateAt() returns true if character at
   * aIndex is high surrogate and it's followed by low surrogate.
   */
  inline bool IsHighSurrogateFollowedByLowSurrogateAt(int32_t aIndex) const {
    MOZ_ASSERT(aIndex >= 0);
    MOZ_ASSERT(aIndex < mState.mLength);
    if (!mState.mIs2b || aIndex + 1 >= mState.mLength) {
      return false;
    }
    return NS_IS_HIGH_SURROGATE(Get2b()[aIndex]) &&
           NS_IS_LOW_SURROGATE(Get2b()[aIndex + 1]);
  }

  /**
   * IsLowSurrogateFollowingHighSurrogateAt() returns true if character at
   * aIndex is low surrogate and it follows high surrogate.
   */
  inline bool IsLowSurrogateFollowingHighSurrogateAt(int32_t aIndex) const {
    MOZ_ASSERT(aIndex >= 0);
    MOZ_ASSERT(aIndex < mState.mLength);
    if (!mState.mIs2b || aIndex <= 0) {
      return false;
    }
    return NS_IS_LOW_SURROGATE(Get2b()[aIndex]) &&
           NS_IS_HIGH_SURROGATE(Get2b()[aIndex - 1]);
  }

  /**
   * ScalarValueAt() returns a Unicode scalar value at aIndex.  If the character
   * at aIndex is a high surrogate followed by low surrogate, returns character
   * code for the pair.  If the index is low surrogate, or a high surrogate but
   * not in a pair, returns 0.
   */
  inline char32_t ScalarValueAt(int32_t aIndex) const {
    MOZ_ASSERT(aIndex >= 0);
    MOZ_ASSERT(aIndex < mState.mLength);
    if (!mState.mIs2b) {
      return static_cast<unsigned char>(m1b[aIndex]);
    }
    char16_t ch = Get2b()[aIndex];
    if (!IS_SURROGATE(ch)) {
      return ch;
    }
    if (aIndex + 1 < mState.mLength && NS_IS_HIGH_SURROGATE(ch)) {
      char16_t nextCh = Get2b()[aIndex + 1];
      if (NS_IS_LOW_SURROGATE(nextCh)) {
        return SURROGATE_TO_UCS4(ch, nextCh);
      }
    }
    return 0;
  }

  void SetBidi(bool aBidi) { mState.mIsBidi = aBidi; }

  struct FragmentBits {
    // uint32_t to ensure that the values are unsigned, because we
    // want 0/1, not 0/-1!
    // Making these bool causes Windows to not actually pack them,
    // which causes crashes because we assume this structure is no more than
    // 32 bits!
    uint32_t mInHeap : 1;
    uint32_t mIs2b : 1;
    uint32_t mIsBidi : 1;
    // Note that when you change the bits of mLength, you also need to change
    // NS_MAX_TEXT_FRAGMENT_LENGTH.
    uint32_t mLength : 29;
  };

#define NS_MAX_TEXT_FRAGMENT_LENGTH (static_cast<uint32_t>(0x1FFFFFFF))

  size_t SizeOfExcludingThis(mozilla::MallocSizeOf aMallocSizeOf) const;

  /**
   * Check whether the text in this fragment is the same as the text in the
   * other fragment.
   */
  MOZ_MUST_USE bool TextEquals(const nsTextFragment& aOther) const;

 private:
  void ReleaseText();

  /**
   * Scan the contents of the fragment and turn on mState.mIsBidi if it
   * includes any Bidi characters.
   */
  void UpdateBidiFlag(const char16_t* aBuffer, uint32_t aLength);

  union {
    nsStringBuffer* m2b;
    const char* m1b;  // This is const since it can point to shared data
  };

  union {
    uint32_t mAllBits;
    FragmentBits mState;
  };
};

#endif /* nsTextFragment_h___ */