uri.cc 14.9 KB
Newer Older
1 2 3 4 5 6
// Copyright 2016 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#include "src/uri.h"

7 8
#include <vector>

9 10 11
#include "src/char-predicates-inl.h"
#include "src/handles.h"
#include "src/isolate-inl.h"
12
#include "src/string-search.h"
13
#include "src/unicode-inl.h"
14 15 16 17

namespace v8 {
namespace internal {

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
namespace {  // anonymous namespace for DecodeURI helper functions
bool IsReservedPredicate(uc16 c) {
  switch (c) {
    case '#':
    case '$':
    case '&':
    case '+':
    case ',':
    case '/':
    case ':':
    case ';':
    case '=':
    case '?':
    case '@':
      return true;
    default:
      return false;
  }
}

bool IsReplacementCharacter(const uint8_t* octets, int length) {
  // The replacement character is at codepoint U+FFFD in the Unicode Specials
  // table. Its UTF-8 encoding is 0xEF 0xBF 0xBD.
41 42
  if (length != 3 || octets[0] != 0xEF || octets[1] != 0xBF ||
      octets[2] != 0xBD) {
43 44 45 46 47
    return false;
  }
  return true;
}

48 49
bool DecodeOctets(const uint8_t* octets, int length,
                  std::vector<uc16>* buffer) {
50 51 52 53 54 55 56
  size_t cursor = 0;
  uc32 value = unibrow::Utf8::ValueOf(octets, length, &cursor);
  if (value == unibrow::Utf8::kBadChar &&
      !IsReplacementCharacter(octets, length)) {
    return false;
  }

57
  if (value <= static_cast<uc32>(unibrow::Utf16::kMaxNonSurrogateCharCode)) {
58
    buffer->push_back(value);
59
  } else {
60 61
    buffer->push_back(unibrow::Utf16::LeadSurrogate(value));
    buffer->push_back(unibrow::Utf16::TrailSurrogate(value));
62 63 64 65
  }
  return true;
}

66 67 68 69 70 71 72 73
int TwoDigitHex(uc16 character1, uc16 character2) {
  if (character1 > 'f') return -1;
  int high = HexValue(character1);
  if (high == -1) return -1;
  if (character2 > 'f') return -1;
  int low = HexValue(character2);
  if (low == -1) return -1;
  return (high << 4) + low;
74 75 76 77
}

template <typename T>
void AddToBuffer(uc16 decoded, String::FlatContent* uri_content, int index,
78
                 bool is_uri, std::vector<T>* buffer) {
79
  if (is_uri && IsReservedPredicate(decoded)) {
80
    buffer->push_back('%');
81 82 83 84 85
    uc16 first = uri_content->Get(index + 1);
    uc16 second = uri_content->Get(index + 2);
    DCHECK_GT(std::numeric_limits<T>::max(), first);
    DCHECK_GT(std::numeric_limits<T>::max(), second);

86 87
    buffer->push_back(first);
    buffer->push_back(second);
88
  } else {
89
    buffer->push_back(decoded);
90 91 92 93
  }
}

bool IntoTwoByte(int index, bool is_uri, int uri_length,
94
                 String::FlatContent* uri_content, std::vector<uc16>* buffer) {
95 96 97
  for (int k = index; k < uri_length; k++) {
    uc16 code = uri_content->Get(k);
    if (code == '%') {
98
      int two_digits;
99
      if (k + 2 >= uri_length ||
100 101
          (two_digits = TwoDigitHex(uri_content->Get(k + 1),
                                    uri_content->Get(k + 2))) < 0) {
102 103 104
        return false;
      }
      k += 2;
105
      uc16 decoded = static_cast<uc16>(two_digits);
106 107 108 109 110 111 112 113 114 115
      if (decoded > unibrow::Utf8::kMaxOneByteChar) {
        uint8_t octets[unibrow::Utf8::kMaxEncodedSize];
        octets[0] = decoded;

        int number_of_continuation_bytes = 0;
        while ((decoded << ++number_of_continuation_bytes) & 0x80) {
          if (number_of_continuation_bytes > 3 || k + 3 >= uri_length) {
            return false;
          }
          if (uri_content->Get(++k) != '%' ||
116 117
              (two_digits = TwoDigitHex(uri_content->Get(k + 1),
                                        uri_content->Get(k + 2))) < 0) {
118 119 120
            return false;
          }
          k += 2;
121
          uc16 continuation_byte = static_cast<uc16>(two_digits);
122 123 124 125 126 127 128 129 130 131
          octets[number_of_continuation_bytes] = continuation_byte;
        }

        if (!DecodeOctets(octets, number_of_continuation_bytes, buffer)) {
          return false;
        }
      } else {
        AddToBuffer(decoded, uri_content, k - 2, is_uri, buffer);
      }
    } else {
132
      buffer->push_back(code);
133 134 135 136 137 138
    }
  }
  return true;
}

bool IntoOneAndTwoByte(Handle<String> uri, bool is_uri,
139 140
                       std::vector<uint8_t>* one_byte_buffer,
                       std::vector<uc16>* two_byte_buffer) {
141 142 143 144 145 146 147
  DisallowHeapAllocation no_gc;
  String::FlatContent uri_content = uri->GetFlatContent();

  int uri_length = uri->length();
  for (int k = 0; k < uri_length; k++) {
    uc16 code = uri_content.Get(k);
    if (code == '%') {
148
      int two_digits;
149
      if (k + 2 >= uri_length ||
150 151
          (two_digits = TwoDigitHex(uri_content.Get(k + 1),
                                    uri_content.Get(k + 2))) < 0) {
152 153 154
        return false;
      }

155
      uc16 decoded = static_cast<uc16>(two_digits);
156 157 158 159 160 161 162 163 164 165 166 167
      if (decoded > unibrow::Utf8::kMaxOneByteChar) {
        return IntoTwoByte(k, is_uri, uri_length, &uri_content,
                           two_byte_buffer);
      }

      AddToBuffer(decoded, &uri_content, k, is_uri, one_byte_buffer);
      k += 2;
    } else {
      if (code > unibrow::Utf8::kMaxOneByteChar) {
        return IntoTwoByte(k, is_uri, uri_length, &uri_content,
                           two_byte_buffer);
      }
168
      one_byte_buffer->push_back(code);
169 170 171 172 173 174 175 176 177 178
    }
  }
  return true;
}

}  // anonymous namespace

MaybeHandle<String> Uri::Decode(Isolate* isolate, Handle<String> uri,
                                bool is_uri) {
  uri = String::Flatten(uri);
179 180
  std::vector<uint8_t> one_byte_buffer;
  std::vector<uc16> two_byte_buffer;
181 182 183 184 185

  if (!IntoOneAndTwoByte(uri, is_uri, &one_byte_buffer, &two_byte_buffer)) {
    THROW_NEW_ERROR(isolate, NewURIError(), String);
  }

186 187 188
  if (two_byte_buffer.empty()) {
    return isolate->factory()->NewStringFromOneByte(Vector<const uint8_t>(
        one_byte_buffer.data(), static_cast<int>(one_byte_buffer.size())));
189 190 191
  }

  Handle<SeqTwoByteString> result;
192 193
  int result_length =
      static_cast<int>(one_byte_buffer.size() + two_byte_buffer.size());
194
  ASSIGN_RETURN_ON_EXCEPTION(
195
      isolate, result, isolate->factory()->NewRawTwoByteString(result_length),
196 197
      String);

198 199 200
  CopyChars(result->GetChars(), one_byte_buffer.data(), one_byte_buffer.size());
  CopyChars(result->GetChars() + one_byte_buffer.size(), two_byte_buffer.data(),
            two_byte_buffer.size());
201 202 203 204

  return result;
}

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
namespace {  // anonymous namespace for EncodeURI helper functions
bool IsUnescapePredicateInUriComponent(uc16 c) {
  if (IsAlphaNumeric(c)) {
    return true;
  }

  switch (c) {
    case '!':
    case '\'':
    case '(':
    case ')':
    case '*':
    case '-':
    case '.':
    case '_':
    case '~':
      return true;
    default:
      return false;
  }
}

bool IsUriSeparator(uc16 c) {
  switch (c) {
    case '#':
    case ':':
    case ';':
    case '/':
    case '?':
    case '$':
    case '&':
    case '+':
    case ',':
    case '@':
    case '=':
      return true;
    default:
      return false;
  }
}

246 247 248 249
void AddEncodedOctetToBuffer(uint8_t octet, std::vector<uint8_t>* buffer) {
  buffer->push_back('%');
  buffer->push_back(HexCharOfValue(octet >> 4));
  buffer->push_back(HexCharOfValue(octet & 0x0F));
250 251
}

252
void EncodeSingle(uc16 c, std::vector<uint8_t>* buffer) {
253
  char s[4] = {};
254 255 256 257
  int number_of_bytes;
  number_of_bytes =
      unibrow::Utf8::Encode(s, c, unibrow::Utf16::kNoPreviousCharacter, false);
  for (int k = 0; k < number_of_bytes; k++) {
258
    AddEncodedOctetToBuffer(s[k], buffer);
259 260 261
  }
}

262
void EncodePair(uc16 cc1, uc16 cc2, std::vector<uint8_t>* buffer) {
263
  char s[4] = {};
264 265 266 267
  int number_of_bytes =
      unibrow::Utf8::Encode(s, unibrow::Utf16::CombineSurrogatePair(cc1, cc2),
                            unibrow::Utf16::kNoPreviousCharacter, false);
  for (int k = 0; k < number_of_bytes; k++) {
268
    AddEncodedOctetToBuffer(s[k], buffer);
269
  }
270 271 272 273
}

}  // anonymous namespace

274 275
MaybeHandle<String> Uri::Encode(Isolate* isolate, Handle<String> uri,
                                bool is_uri) {
276 277
  uri = String::Flatten(uri);
  int uri_length = uri->length();
278 279
  std::vector<uint8_t> buffer;
  buffer.reserve(uri_length);
280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298

  {
    DisallowHeapAllocation no_gc;
    String::FlatContent uri_content = uri->GetFlatContent();

    for (int k = 0; k < uri_length; k++) {
      uc16 cc1 = uri_content.Get(k);
      if (unibrow::Utf16::IsLeadSurrogate(cc1)) {
        k++;
        if (k < uri_length) {
          uc16 cc2 = uri->Get(k);
          if (unibrow::Utf16::IsTrailSurrogate(cc2)) {
            EncodePair(cc1, cc2, &buffer);
            continue;
          }
        }
      } else if (!unibrow::Utf16::IsTrailSurrogate(cc1)) {
        if (IsUnescapePredicateInUriComponent(cc1) ||
            (is_uri && IsUriSeparator(cc1))) {
299
          buffer.push_back(cc1);
300 301 302 303 304 305 306
        } else {
          EncodeSingle(cc1, &buffer);
        }
        continue;
      }

      AllowHeapAllocation allocate_error_and_return;
307
      THROW_NEW_ERROR(isolate, NewURIError(), String);
308 309 310
    }
  }

311 312
  return isolate->factory()->NewStringFromOneByte(
      Vector<const uint8_t>(buffer.data(), static_cast<int>(buffer.size())));
313 314
}

315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
namespace {  // Anonymous namespace for Escape and Unescape

template <typename Char>
int UnescapeChar(Vector<const Char> vector, int i, int length, int* step) {
  uint16_t character = vector[i];
  int32_t hi = 0;
  int32_t lo = 0;
  if (character == '%' && i <= length - 6 && vector[i + 1] == 'u' &&
      (hi = TwoDigitHex(vector[i + 2], vector[i + 3])) > -1 &&
      (lo = TwoDigitHex(vector[i + 4], vector[i + 5])) > -1) {
    *step = 6;
    return (hi << 8) + lo;
  } else if (character == '%' && i <= length - 3 &&
             (lo = TwoDigitHex(vector[i + 1], vector[i + 2])) > -1) {
    *step = 3;
    return lo;
  } else {
    *step = 1;
    return character;
  }
}

template <typename Char>
MaybeHandle<String> UnescapeSlow(Isolate* isolate, Handle<String> string,
                                 int start_index) {
  bool one_byte = true;
  int length = string->length();

  int unescaped_length = 0;
  {
    DisallowHeapAllocation no_allocation;
    Vector<const Char> vector = string->GetCharVector<Char>();
    for (int i = start_index; i < length; unescaped_length++) {
      int step;
      if (UnescapeChar(vector, i, length, &step) >
          String::kMaxOneByteCharCode) {
        one_byte = false;
      }
      i += step;
    }
  }

  DCHECK(start_index < length);
  Handle<String> first_part =
      isolate->factory()->NewProperSubString(string, 0, start_index);

  int dest_position = 0;
  Handle<String> second_part;
363
  DCHECK_LE(unescaped_length, String::kMaxLength);
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
  if (one_byte) {
    Handle<SeqOneByteString> dest = isolate->factory()
                                        ->NewRawOneByteString(unescaped_length)
                                        .ToHandleChecked();
    DisallowHeapAllocation no_allocation;
    Vector<const Char> vector = string->GetCharVector<Char>();
    for (int i = start_index; i < length; dest_position++) {
      int step;
      dest->SeqOneByteStringSet(dest_position,
                                UnescapeChar(vector, i, length, &step));
      i += step;
    }
    second_part = dest;
  } else {
    Handle<SeqTwoByteString> dest = isolate->factory()
                                        ->NewRawTwoByteString(unescaped_length)
                                        .ToHandleChecked();
    DisallowHeapAllocation no_allocation;
    Vector<const Char> vector = string->GetCharVector<Char>();
    for (int i = start_index; i < length; dest_position++) {
      int step;
      dest->SeqTwoByteStringSet(dest_position,
                                UnescapeChar(vector, i, length, &step));
      i += step;
    }
    second_part = dest;
  }
  return isolate->factory()->NewConsString(first_part, second_part);
}

bool IsNotEscaped(uint16_t c) {
  if (IsAlphaNumeric(c)) {
    return true;
  }
  //  @*_+-./
  switch (c) {
    case '@':
    case '*':
    case '_':
    case '+':
    case '-':
    case '.':
    case '/':
      return true;
    default:
      return false;
  }
}

template <typename Char>
static MaybeHandle<String> UnescapePrivate(Isolate* isolate,
                                           Handle<String> source) {
  int index;
  {
    DisallowHeapAllocation no_allocation;
    StringSearch<uint8_t, Char> search(isolate, STATIC_CHAR_VECTOR("%"));
    index = search.Search(source->GetCharVector<Char>(), 0);
    if (index < 0) return source;
  }
  return UnescapeSlow<Char>(isolate, source, index);
}

template <typename Char>
static MaybeHandle<String> EscapePrivate(Isolate* isolate,
                                         Handle<String> string) {
  DCHECK(string->IsFlat());
  int escaped_length = 0;
  int length = string->length();

  {
    DisallowHeapAllocation no_allocation;
    Vector<const Char> vector = string->GetCharVector<Char>();
    for (int i = 0; i < length; i++) {
      uint16_t c = vector[i];
      if (c >= 256) {
        escaped_length += 6;
      } else if (IsNotEscaped(c)) {
        escaped_length++;
      } else {
        escaped_length += 3;
      }

      // We don't allow strings that are longer than a maximal length.
447
      DCHECK_LT(String::kMaxLength, 0x7FFFFFFF - 6);   // Cannot overflow.
448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
      if (escaped_length > String::kMaxLength) break;  // Provoke exception.
    }
  }

  // No length change implies no change.  Return original string if no change.
  if (escaped_length == length) return string;

  Handle<SeqOneByteString> dest;
  ASSIGN_RETURN_ON_EXCEPTION(
      isolate, dest, isolate->factory()->NewRawOneByteString(escaped_length),
      String);
  int dest_position = 0;

  {
    DisallowHeapAllocation no_allocation;
    Vector<const Char> vector = string->GetCharVector<Char>();
    for (int i = 0; i < length; i++) {
      uint16_t c = vector[i];
      if (c >= 256) {
        dest->SeqOneByteStringSet(dest_position, '%');
        dest->SeqOneByteStringSet(dest_position + 1, 'u');
        dest->SeqOneByteStringSet(dest_position + 2, HexCharOfValue(c >> 12));
        dest->SeqOneByteStringSet(dest_position + 3,
471
                                  HexCharOfValue((c >> 8) & 0xF));
472
        dest->SeqOneByteStringSet(dest_position + 4,
473 474
                                  HexCharOfValue((c >> 4) & 0xF));
        dest->SeqOneByteStringSet(dest_position + 5, HexCharOfValue(c & 0xF));
475 476 477 478 479 480 481
        dest_position += 6;
      } else if (IsNotEscaped(c)) {
        dest->SeqOneByteStringSet(dest_position, c);
        dest_position++;
      } else {
        dest->SeqOneByteStringSet(dest_position, '%');
        dest->SeqOneByteStringSet(dest_position + 1, HexCharOfValue(c >> 4));
482
        dest->SeqOneByteStringSet(dest_position + 2, HexCharOfValue(c & 0xF));
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
        dest_position += 3;
      }
    }
  }

  return dest;
}

}  // Anonymous namespace

MaybeHandle<String> Uri::Escape(Isolate* isolate, Handle<String> string) {
  Handle<String> result;
  string = String::Flatten(string);
  return string->IsOneByteRepresentationUnderneath()
             ? EscapePrivate<uint8_t>(isolate, string)
             : EscapePrivate<uc16>(isolate, string);
}

MaybeHandle<String> Uri::Unescape(Isolate* isolate, Handle<String> string) {
  Handle<String> result;
  string = String::Flatten(string);
  return string->IsOneByteRepresentationUnderneath()
             ? UnescapePrivate<uint8_t>(isolate, string)
             : UnescapePrivate<uc16>(isolate, string);
}

509 510
}  // namespace internal
}  // namespace v8