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 (b66b287a2650)

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set sw=2 ts=2 et tw=79: */
/* ***** BEGIN LICENSE BLOCK *****
 * Version: MPL 1.1/GPL 2.0/LGPL 2.1
 *
 * The contents of this file are subject to the Mozilla Public License Version
 * 1.1 (the "License"); you may not use this file except in compliance with
 * the License. You may obtain a copy of the License at
 * http://www.mozilla.org/MPL/
 *
 * Software distributed under the License is distributed on an "AS IS" basis,
 * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
 * for the specific language governing rights and limitations under the
 * License.
 *
 * The Original Code is mozilla.org code.
 *
 * The Initial Developer of the Original Code is
 * Netscape Communications Corporation.
 * Portions created by the Initial Developer are Copyright (C) 1998
 * the Initial Developer. All Rights Reserved.
 *
 * Contributor(s):
 *   Pierre Phaneuf <pp@ludusdesign.com>
 *   Henri Sivonen <hsivonen@iki.fi>
 *
 * Alternatively, the contents of this file may be used under the terms of
 * either of the GNU General Public License Version 2 or later (the "GPL"),
 * or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
 * in which case the provisions of the GPL or the LGPL are applicable instead
 * of those above. If you wish to allow use of your version of this file only
 * under the terms of either the GPL or the LGPL, and not to allow others to
 * use your version of this file under the terms of the MPL, indicate your
 * decision by deleting the provisions above and replace them with the notice
 * and other provisions required by the GPL or the LGPL. If you do not delete
 * the provisions above, a recipient may use your version of this file under
 * the terms of any one of the MPL, the GPL or the LGPL.
 *
 * ***** END LICENSE BLOCK ***** */

#include "nsCompatibility.h"
#include "nsScriptLoader.h"
#include "nsNetUtil.h"
#include "nsIStyleSheetLinkingElement.h"
#include "nsICharsetConverterManager.h"
#include "nsICharsetAlias.h"
#include "nsIWebShellServices.h"
#include "nsIDocShell.h"
#include "nsEncoderDecoderUtils.h"
#include "nsContentUtils.h"
#include "nsICharsetDetector.h"
#include "nsIScriptElement.h"
#include "nsIMarkupDocumentViewer.h"
#include "nsIDocShellTreeItem.h"
#include "nsIContentViewer.h"
#include "nsIScriptGlobalObjectOwner.h"
#include "nsIScriptSecurityManager.h"
#include "nsHtml5DocumentMode.h"
#include "nsHtml5Tokenizer.h"
#include "nsHtml5UTF16Buffer.h"
#include "nsHtml5TreeBuilder.h"
#include "nsHtml5Parser.h"

static NS_DEFINE_CID(kCharsetAliasCID, NS_CHARSETALIAS_CID);

//-------------- Begin ParseContinue Event Definition ------------------------
/*
The parser can be explicitly interrupted by calling Suspend(). This will cause
the parser to stop processing and allow the application to return to the event
loop. The parser will schedule a nsHtml5ParserContinueEvent which will call
the parser to process the remaining data after returning to the event loop.
*/
class nsHtml5ParserContinueEvent : public nsRunnable
{
public:
  nsRefPtr<nsHtml5Parser> mParser;
  nsHtml5ParserContinueEvent(nsHtml5Parser* aParser)
    : mParser(aParser)
  {}
  NS_IMETHODIMP Run()
  {
    mParser->HandleParserContinueEvent(this);
    return NS_OK;
  }
};
//-------------- End ParseContinue Event Definition ------------------------

NS_IMPL_CYCLE_COLLECTION_CLASS(nsHtml5Parser)

NS_INTERFACE_TABLE_HEAD_CYCLE_COLLECTION_INHERITED(nsHtml5Parser) \
  NS_INTERFACE_TABLE_INHERITED3(nsHtml5Parser, 
                                nsIParser, 
                                nsIStreamListener, 
                                nsIContentSink)
NS_INTERFACE_TABLE_TAIL_INHERITING(nsContentSink)

NS_IMPL_ADDREF_INHERITED(nsHtml5Parser, nsContentSink)

NS_IMPL_RELEASE_INHERITED(nsHtml5Parser, nsContentSink)

NS_IMPL_CYCLE_COLLECTION_TRAVERSE_BEGIN_INHERITED(nsHtml5Parser, nsContentSink)
  NS_IMPL_CYCLE_COLLECTION_TRAVERSE_NSCOMPTR(mScriptElement)
  NS_IMPL_CYCLE_COLLECTION_TRAVERSE_NSCOMPTR(mRequest)
  NS_IMPL_CYCLE_COLLECTION_TRAVERSE_NSCOMPTR(mObserver)
  NS_IMPL_CYCLE_COLLECTION_TRAVERSE_NSCOMPTR(mUnicodeDecoder)
  tmp->mTreeBuilder->DoTraverse(cb);
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_END

NS_IMPL_CYCLE_COLLECTION_UNLINK_BEGIN_INHERITED(nsHtml5Parser, nsContentSink)
  NS_IMPL_CYCLE_COLLECTION_UNLINK_NSCOMPTR(mScriptElement)
  NS_IMPL_CYCLE_COLLECTION_UNLINK_NSCOMPTR(mRequest)
  NS_IMPL_CYCLE_COLLECTION_UNLINK_NSCOMPTR(mObserver)
  NS_IMPL_CYCLE_COLLECTION_UNLINK_NSCOMPTR(mUnicodeDecoder)
  tmp->mTreeBuilder->DoUnlink();
NS_IMPL_CYCLE_COLLECTION_UNLINK_END

nsHtml5Parser::nsHtml5Parser()
  : mFirstBuffer(new nsHtml5UTF16Buffer(NS_HTML5_PARSER_READ_BUFFER_SIZE)),
    // XXX avoid allocating the buffer at all for fragment parser?
    mLastBuffer(mFirstBuffer),
    mTreeBuilder(new nsHtml5TreeBuilder(this)),
    mTokenizer(new nsHtml5Tokenizer(mTreeBuilder))
{
  mTokenizer->setEncodingDeclarationHandler(this);
  // There's a zeroing operator new for everything else
}

nsHtml5Parser::~nsHtml5Parser()
{
  while (mFirstBuffer) {
     nsHtml5UTF16Buffer* old = mFirstBuffer;
     mFirstBuffer = mFirstBuffer->next;
     delete old;
  }
#ifdef GATHER_DOCWRITE_STATISTICS
  delete mSnapshot;
#endif
}

// copied from HTML content sink
static PRBool
IsScriptEnabled(nsIDocument *aDoc, nsIDocShell *aContainer)
{
  NS_ENSURE_TRUE(aDoc && aContainer, PR_TRUE);
  nsCOMPtr<nsIScriptGlobalObject> globalObject = aDoc->GetScriptGlobalObject();
  // Getting context is tricky if the document hasn't had its
  // GlobalObject set yet
  if (!globalObject) {
    nsCOMPtr<nsIScriptGlobalObjectOwner> owner = do_GetInterface(aContainer);
    NS_ENSURE_TRUE(owner, PR_TRUE);
    globalObject = owner->GetScriptGlobalObject();
    NS_ENSURE_TRUE(globalObject, PR_TRUE);
  }
  nsIScriptContext *scriptContext = globalObject->GetContext();
  NS_ENSURE_TRUE(scriptContext, PR_TRUE);
  JSContext* cx = (JSContext *) scriptContext->GetNativeContext();
  NS_ENSURE_TRUE(cx, PR_TRUE);
  PRBool enabled = PR_TRUE;
  nsContentUtils::GetSecurityManager()->
    CanExecuteScripts(cx, aDoc->NodePrincipal(), &enabled);
  return enabled;
}

NS_IMETHODIMP_(void)
nsHtml5Parser::SetContentSink(nsIContentSink* aSink)
{
  NS_ASSERTION(aSink == static_cast<nsIContentSink*> (this), 
               "Attempt to set a foreign sink.");
}

NS_IMETHODIMP_(nsIContentSink*)
nsHtml5Parser::GetContentSink(void)
{
  return static_cast<nsIContentSink*> (this);
}

NS_IMETHODIMP_(void)
nsHtml5Parser::GetCommand(nsCString& aCommand)
{
  aCommand.Assign("view");
}

NS_IMETHODIMP_(void)
nsHtml5Parser::SetCommand(const char* aCommand)
{
  NS_ASSERTION(!strcmp(aCommand, "view"), "Parser command was not view");
}

NS_IMETHODIMP_(void)
nsHtml5Parser::SetCommand(eParserCommands aParserCommand)
{
  NS_ASSERTION(aParserCommand == eViewNormal, 
               "Parser command was not eViewNormal.");
}

NS_IMETHODIMP_(void)
nsHtml5Parser::SetDocumentCharset(const nsACString& aCharset, PRInt32 aCharsetSource)
{
  mCharset = aCharset;
  mCharsetSource = aCharsetSource;
}

NS_IMETHODIMP_(void)
nsHtml5Parser::SetParserFilter(nsIParserFilter* aFilter)
{
  NS_ERROR("Attempt to set a parser filter on HTML5 parser.");
}

NS_IMETHODIMP
nsHtml5Parser::GetChannel(nsIChannel** aChannel)
{
  return mRequest ? CallQueryInterface(mRequest, aChannel) :
                    NS_ERROR_NOT_AVAILABLE;
}

NS_IMETHODIMP
nsHtml5Parser::GetDTD(nsIDTD** aDTD)
{
  *aDTD = nsnull;
  return NS_OK;
}

NS_IMETHODIMP
nsHtml5Parser::ContinueParsing()
{
  mBlocked = PR_FALSE;
  return ContinueInterruptedParsing();
}

NS_IMETHODIMP
nsHtml5Parser::ContinueInterruptedParsing()
{
  // If there are scripts executing, then the content sink is jumping the gun
  // (probably due to a synchronous XMLHttpRequest) and will re-enable us
  // later, see bug 460706.
  if (IsScriptExecutingImpl()) {
    return NS_OK;
  }
  // If the stream has already finished, there's a good chance
  // that we might start closing things down when the parser
  // is reenabled. To make sure that we're not deleted across
  // the reenabling process, hold a reference to ourselves.
  // XXX is this really necessary? -- hsivonen?
  nsCOMPtr<nsIParser> kungFuDeathGrip(this);
  // XXX Stop speculative script thread but why?
  mTreeBuilder->MaybeFlush();
  ParseUntilSuspend();
  return NS_OK;
}

NS_IMETHODIMP_(void)
nsHtml5Parser::BlockParser()
{
  NS_NOTREACHED("No one should call this");
}

NS_IMETHODIMP_(void)
nsHtml5Parser::UnblockParser()
{
  mBlocked = PR_FALSE;
}

NS_IMETHODIMP_(PRBool)
nsHtml5Parser::IsParserEnabled()
{
  return !mBlocked;
}

NS_IMETHODIMP_(PRBool)
nsHtml5Parser::IsComplete()
{
  // XXX old parser says
  // return !(mFlags & NS_PARSER_FLAG_PENDING_CONTINUE_EVENT);
  return (mLifeCycle == TERMINATED);
}

NS_IMETHODIMP
nsHtml5Parser::Parse(nsIURI* aURL, // legacy parameter; ignored
                     nsIRequestObserver* aObserver,
                     void* aKey,
                     nsDTDMode aMode) // legacy; ignored
{
  mObserver = aObserver;
  mRootContextKey = aKey;
  mCanInterruptParser = PR_TRUE;
  NS_PRECONDITION(mLifeCycle == NOT_STARTED, 
                  "Tried to start parse without initializing the parser properly.");
  return NS_OK;
}

NS_IMETHODIMP
nsHtml5Parser::Parse(const nsAString& aSourceBuffer,
                     void* aKey,
                     const nsACString& aContentType, // ignored
                     PRBool aLastCall,
                     nsDTDMode aMode) // ignored
{
  NS_PRECONDITION(!mFragmentMode, "Document.write called in fragment mode!");
  // Return early if the parser has processed EOF
  switch (mLifeCycle) {
    case TERMINATED:
      return NS_OK;
    case NOT_STARTED:
      mTreeBuilder->setScriptingEnabled(IsScriptEnabled(mDocument, mDocShell));
      mTokenizer->start();
      mLifeCycle = PARSING;
      mParser = this;
      mCharsetSource = kCharsetFromOtherComponent;
      break;
    default:
      break;
  }

  if (aLastCall && aSourceBuffer.IsEmpty() && aKey == GetRootContextKey()) {
    // document.close()
    mLifeCycle = STREAM_ENDING;
    MaybePostContinueEvent();
    return NS_OK;
  }

  // XXX stop speculative script thread here

  PRInt32 lineNumberSave = mTokenizer->getLineNumber();

  // Maintain a reference to ourselves so we don't go away
  // till we're completely done.
  // XXX is this still necessary? -- hsivonen
  nsCOMPtr<nsIParser> kungFuDeathGrip(this);

  if (!aSourceBuffer.IsEmpty()) {
    nsHtml5UTF16Buffer* buffer = new nsHtml5UTF16Buffer(aSourceBuffer.Length());
    memcpy(buffer->getBuffer(), aSourceBuffer.BeginReading(), aSourceBuffer.Length() * sizeof(PRUnichar));
    buffer->setEnd(aSourceBuffer.Length());
    if (!mBlocked) {
      WillResumeImpl();
      WillParseImpl();
      while (buffer->hasMore()) {
        buffer->adjust(mLastWasCR);
        mLastWasCR = PR_FALSE;
        if (buffer->hasMore()) {
          mUninterruptibleDocWrite = PR_TRUE;
          mLastWasCR = mTokenizer->tokenizeBuffer(buffer);
          if (mScriptElement) {
            mTreeBuilder->Flush();
            mUninterruptibleDocWrite = PR_FALSE;
            ExecuteScript();
          }
          if (mBlocked) {
            // XXX is the tail insertion and script exec in the wrong order?
            WillInterruptImpl();
            break;
          }
          // Ignore suspension requests
        }
      }
    }

    mUninterruptibleDocWrite = PR_FALSE;

    if (buffer->hasMore()) {
      // If we got here, the buffer wasn't parsed synchronously to completion
      // and its tail needs to go into the chain of pending buffers.
      // The script is identified by aKey. If there's nothing in the buffer
      // chain for that key, we'll insert at the head of the queue.
      // When the script leaves something in the queue, a zero-length
      // key-holder "buffer" is inserted in the queue. If the same script
      // leaves something in the chain again, it will be inserted immediately
      // before the old key holder belonging to the same script.
      nsHtml5UTF16Buffer* prevSearchBuf = nsnull;
      nsHtml5UTF16Buffer* searchBuf = mFirstBuffer;
      if (aKey) { // after document.open, the first level of document.write has null key
        while (searchBuf != mLastBuffer) {
          if (searchBuf->key == aKey) {
            // found a key holder
            // now insert the new buffer between the previous buffer
            // and the key holder.
            buffer->next = searchBuf;
            if (prevSearchBuf) {
              prevSearchBuf->next = buffer;
            } else {
              mFirstBuffer = buffer;
            }
            break;
          }
          prevSearchBuf = searchBuf;
          searchBuf = searchBuf->next;
        }
      }
      if (searchBuf == mLastBuffer || !aKey) {
        // key was not found or we have a first-level write after document.open
        // we'll insert to the head of the queue
        nsHtml5UTF16Buffer* keyHolder = new nsHtml5UTF16Buffer(aKey);
        keyHolder->next = mFirstBuffer;
        buffer->next = keyHolder;
        mFirstBuffer = buffer;
      }
      MaybePostContinueEvent();
    } else {
      delete buffer;
    }
  }

  // Scripting semantics require a forced tree builder flush here
  mTreeBuilder->Flush();
  mTokenizer->setLineNumber(lineNumberSave);
  return NS_OK;
}

/**
 * This magic value is passed to the previous method on document.close()
 */
NS_IMETHODIMP_(void *)
nsHtml5Parser::GetRootContextKey()
{
  return mRootContextKey;
}

NS_IMETHODIMP
nsHtml5Parser::Terminate(void)
{
  // We should only call DidBuildModel once, so don't do anything if this is
  // the second time that Terminate has been called.
  if (mLifeCycle == TERMINATED) {
    return NS_OK;
  }
  // XXX - [ until we figure out a way to break parser-sink circularity ]
  // Hack - Hold a reference until we are completely done...
  nsCOMPtr<nsIParser> kungFuDeathGrip(this);
  // CancelParsingEvents must be called to avoid leaking the nsParser object
  // @see bug 108049
  CancelParsingEvents();
  return DidBuildModel(PR_TRUE);
}

NS_IMETHODIMP
nsHtml5Parser::ParseFragment(const nsAString& aSourceBuffer,
                             void* aKey,
                             nsTArray<nsString>& aTagStack,
                             PRBool aXMLMode,
                             const nsACString& aContentType,
                             nsDTDMode aMode)
{
  return NS_ERROR_NOT_IMPLEMENTED;
}

NS_IMETHODIMP
nsHtml5Parser::ParseFragment(const nsAString& aSourceBuffer,
                             nsISupports* aTargetNode,
                             nsIAtom* aContextLocalName,
                             PRInt32 aContextNamespace,
                             PRBool aQuirks)
{
  nsCOMPtr<nsIContent> target = do_QueryInterface(aTargetNode);
  NS_ASSERTION(target, "Target did not QI to nsIContent");

  nsIDocument* doc = target->GetOwnerDoc();
  NS_ENSURE_TRUE(doc, NS_ERROR_NOT_AVAILABLE);
  
  nsIURI* uri = doc->GetDocumentURI();
  NS_ENSURE_TRUE(uri, NS_ERROR_NOT_AVAILABLE);

  Initialize(doc, uri, nsnull, nsnull);

  // Initialize() doesn't deal with base URI
  mDocumentBaseURI = doc->GetBaseURI();

  mTreeBuilder->setFragmentContext(aContextLocalName, aContextNamespace, target, aQuirks);
  mFragmentMode = PR_TRUE;
  mCanInterruptParser = PR_FALSE;
  NS_PRECONDITION(mLifeCycle == NOT_STARTED, "Tried to start parse without initializing the parser properly.");
  mTreeBuilder->setScriptingEnabled(IsScriptEnabled(mDocument, mDocShell));
  mTokenizer->start();
  mLifeCycle = PARSING;
  mParser = this;
  mNodeInfoManager = target->GetOwnerDoc()->NodeInfoManager();
  if (!aSourceBuffer.IsEmpty()) {
    PRBool lastWasCR = PR_FALSE;
    nsHtml5UTF16Buffer buffer(aSourceBuffer.Length());
    memcpy(buffer.getBuffer(), aSourceBuffer.BeginReading(), aSourceBuffer.Length() * sizeof(PRUnichar));
    buffer.setEnd(aSourceBuffer.Length());
    while (buffer.hasMore()) {
      buffer.adjust(lastWasCR);
      lastWasCR = PR_FALSE;
      if (buffer.hasMore()) {
        lastWasCR = mTokenizer->tokenizeBuffer(&buffer);
        if (mScriptElement) {
          nsCOMPtr<nsIScriptElement> script = do_QueryInterface(mScriptElement);
          NS_ASSERTION(script, "mScriptElement didn't QI to nsIScriptElement!");
          script->PreventExecution();
          mScriptElement = nsnull;
        }
      }
    }
  }
  mTokenizer->eof();
  mTokenizer->end();
  mLifeCycle = TERMINATED;
  DropParserAndPerfHint();
  return NS_OK;
}

NS_IMETHODIMP
nsHtml5Parser::BuildModel(void)
{
  // XXX who calls this? Should this be a no-op?
  ParseUntilSuspend();
  return NS_OK;
}

NS_IMETHODIMP
nsHtml5Parser::CancelParsingEvents()
{
  mContinueEvent = nsnull;
  return NS_OK;
}

void
nsHtml5Parser::Reset()
{
  mNeedsCharsetSwitch = PR_FALSE;
  mLastWasCR = PR_FALSE;
  mFragmentMode = PR_FALSE;
  mBlocked = PR_FALSE;
  mSuspending = PR_FALSE;
  mLifeCycle = NOT_STARTED;
  mScriptElement = nsnull;
  mUninterruptibleDocWrite = PR_FALSE;
  mRootContextKey = nsnull;
  mRequest = nsnull;
  mObserver = nsnull;
  mContinueEvent = nsnull;  // weak ref
  // encoding-related stuff
  mCharsetSource = kCharsetUninitialized;
  mCharset.Truncate();
  mPendingCharset.Truncate();
  mUnicodeDecoder = nsnull;
  mSniffingBuffer = nsnull;
  mSniffingLength = 0;
  mBomState = BOM_SNIFFING_NOT_STARTED;
  mMetaScanner = nsnull;
  // Portable parser objects
  while (mFirstBuffer->next) {
    nsHtml5UTF16Buffer* oldBuf = mFirstBuffer;
    mFirstBuffer = mFirstBuffer->next;
    delete oldBuf;
  }
  mFirstBuffer->setStart(0);
  mFirstBuffer->setEnd(0);
#ifdef DEBUG
  mStreamListenerState = eNone;
#endif
}

PRBool
nsHtml5Parser::CanInterrupt()
{
  return !(mFragmentMode || mUninterruptibleDocWrite);
}

/* End nsIParser  */

// nsIRequestObserver methods:
nsresult
nsHtml5Parser::OnStartRequest(nsIRequest* aRequest, nsISupports* aContext)
{
  NS_PRECONDITION(eNone == mStreamListenerState,
                  "Parser's nsIStreamListener API was not setup "
                  "correctly in constructor.");
  if (mObserver) {
    mObserver->OnStartRequest(aRequest, aContext);
  }
#ifdef DEBUG
  mStreamListenerState = eOnStart;
#endif
  mRequest = aRequest;
  
  if (mCharsetSource < kCharsetFromChannel) {
    // we aren't ready to commit to an encoding yet
    // leave converter uninstantiated for now
    return NS_OK;
  }
  
  nsresult rv = NS_OK;
  nsCOMPtr<nsICharsetConverterManager> convManager = do_GetService(NS_CHARSETCONVERTERMANAGER_CONTRACTID, &rv);
  NS_ENSURE_SUCCESS(rv, rv);
  rv = convManager->GetUnicodeDecoder(mCharset.get(), getter_AddRefs(mUnicodeDecoder));
  NS_ENSURE_SUCCESS(rv, rv);
  mUnicodeDecoder->SetInputErrorBehavior(nsIUnicodeDecoder::kOnError_Recover);
  return NS_OK;
}

nsresult
nsHtml5Parser::OnStopRequest(nsIRequest* aRequest,
                             nsISupports* aContext,
                             nsresult status)
{
  mTreeBuilder->MaybeFlush();
  NS_ASSERTION(mRequest == aRequest, "Got Stop on wrong stream.");
  nsresult rv = NS_OK;
  if (!mUnicodeDecoder) {
    PRUint32 writeCount;
    rv = FinalizeSniffing(nsnull, 0, &writeCount, 0);
    NS_ENSURE_SUCCESS(rv, rv);
  }
  switch (mLifeCycle) {
    case TERMINATED:
      break;
    case NOT_STARTED:
      mTreeBuilder->setScriptingEnabled(IsScriptEnabled(mDocument, mDocShell));
      mTokenizer->start();
      mLifeCycle = STREAM_ENDING;
      mParser = this;
      break;
    case STREAM_ENDING:
      NS_ERROR("OnStopRequest when the stream lifecycle was already ending.");
      break;
    default:
      mLifeCycle = STREAM_ENDING;
      break;
  }
#ifdef DEBUG
  mStreamListenerState = eOnStop;
#endif
  if (!IsScriptExecutingImpl()) {
    ParseUntilSuspend();
  }
  if (mObserver) {
    mObserver->OnStopRequest(aRequest, aContext, status);
  }
  return NS_OK;
}

// nsIStreamListener method:
/*
 * This function is invoked as a result of a call to a stream's
 * ReadSegments() method. It is called for each contiguous buffer
 * of data in the underlying stream or pipe. Using ReadSegments
 * allows us to avoid copying data to read out of the stream.
 */
static NS_METHOD
ParserWriteFunc(nsIInputStream* aInStream,
                void* aHtml5Parser,
                const char* aFromSegment,
                PRUint32 aToOffset,
                PRUint32 aCount,
                PRUint32* aWriteCount)
{
  nsHtml5Parser* parser = static_cast<nsHtml5Parser*> (aHtml5Parser);
  if (parser->HasDecoder()) {
    return parser->WriteStreamBytes((const PRUint8*)aFromSegment, aCount, aWriteCount);
  } else {
    return parser->SniffStreamBytes((const PRUint8*)aFromSegment, aCount, aWriteCount);
  }
}

nsresult
nsHtml5Parser::OnDataAvailable(nsIRequest* aRequest,
                               nsISupports* aContext,
                               nsIInputStream* aInStream,
                               PRUint32 aSourceOffset,
                               PRUint32 aLength)
{
  mTreeBuilder->MaybeFlush();
  NS_PRECONDITION(eOnStart == mStreamListenerState ||
                  eOnDataAvail == mStreamListenerState,
            "Error: OnStartRequest() must be called before OnDataAvailable()");
  NS_ASSERTION(mRequest == aRequest, "Got data on wrong stream.");
  PRUint32 totalRead;
  nsresult rv = aInStream->ReadSegments(ParserWriteFunc, static_cast<void*> (this), aLength, &totalRead);
  NS_ASSERTION(totalRead == aLength, "ReadSegments read the wrong number of bytes.");
  if (!IsScriptExecutingImpl()) {
    ParseUntilSuspend();
  }
  return rv;
}

// EncodingDeclarationHandler
void
nsHtml5Parser::internalEncodingDeclaration(nsString* aEncoding)
{
  if (mCharsetSource >= kCharsetFromMetaTag) { // this threshold corresponds to "confident" in the HTML5 spec
    return;
  }
  nsresult rv = NS_OK;
  nsCOMPtr<nsICharsetAlias> calias(do_GetService(kCharsetAliasCID, &rv));
  if (NS_FAILED(rv)) {
    return;
  }
  nsCAutoString newEncoding;
  CopyUTF16toUTF8(*aEncoding, newEncoding);
  PRBool eq;
  rv = calias->Equals(newEncoding, mCharset, &eq);
  if (NS_FAILED(rv)) {
    return;
  }
  if (eq) {
    mCharsetSource = kCharsetFromMetaTag; // become confident
    return;
  }
  
  // XXX check HTML5 non-IANA aliases here

  // The encodings are different. We want to reparse.
  nsCOMPtr<nsIHttpChannel> httpChannel(do_QueryInterface(mRequest, &rv));
  if (NS_SUCCEEDED(rv)) {
    nsCAutoString method;
    httpChannel->GetRequestMethod(method);
    // XXX does Necko have a way to renavigate POST, etc. without hitting
    // the network?
    if (!method.EqualsLiteral("GET")) {
      // This is the old Gecko behavior but the spec disagrees.
      // Don't reparse on POST.
      return;
    }
  }
  
  // we still want to reparse
  mNeedsCharsetSwitch = PR_TRUE;
  mPendingCharset.Assign(newEncoding);
}

// DocumentModeHandler
void
nsHtml5Parser::documentMode(nsHtml5DocumentMode m)
{
  nsCompatibility mode = eCompatibility_NavQuirks;
  switch (m) {
    case STANDARDS_MODE:
      mode = eCompatibility_FullStandards;
      break;
    case ALMOST_STANDARDS_MODE:
      mode = eCompatibility_AlmostStandards;
      break;
    case QUIRKS_MODE:
      mode = eCompatibility_NavQuirks;
      break;
  }
  nsCOMPtr<nsIHTMLDocument> htmlDocument = do_QueryInterface(mDocument);
  NS_ASSERTION(htmlDocument, "Document didn't QI into HTML document.");
  htmlDocument->SetCompatibilityMode(mode);
}

// nsIContentSink
NS_IMETHODIMP
nsHtml5Parser::WillParse()
{
  NS_NOTREACHED("No one should call this");
  return NS_ERROR_NOT_IMPLEMENTED;
}

NS_IMETHODIMP
nsHtml5Parser::WillBuildModel(nsDTDMode aDTDMode)
{
  NS_NOTREACHED("No one should call this");
  return NS_ERROR_NOT_IMPLEMENTED;
}

// This is called when the tree construction has ended
NS_IMETHODIMP
nsHtml5Parser::DidBuildModel(PRBool aTerminated)
{
  NS_ASSERTION(mLifeCycle == STREAM_ENDING, "Bad life cycle.");
  mTokenizer->eof();
  mTokenizer->end();
  mLifeCycle = TERMINATED;
  // This is comes from nsXMLContentSink
  DidBuildModelImpl(aTerminated);
  mDocument->ScriptLoader()->RemoveObserver(this);
  StartLayout(PR_FALSE);
  ScrollToRef();
  mDocument->RemoveObserver(this);
  mDocument->EndLoad();
  DropParserAndPerfHint();
#ifdef GATHER_DOCWRITE_STATISTICS
  printf("UNSAFE SCRIPTS: %d\n", sUnsafeDocWrites);
  printf("TOKENIZER-SAFE SCRIPTS: %d\n", sTokenSafeDocWrites);
  printf("TREEBUILDER-SAFE SCRIPTS: %d\n", sTreeSafeDocWrites);
#endif
  return NS_OK;
}

NS_IMETHODIMP
nsHtml5Parser::WillInterrupt()
{
  return WillInterruptImpl();
}

NS_IMETHODIMP
nsHtml5Parser::WillResume()
{
  NS_NOTREACHED("No one should call this.");
  return NS_ERROR_NOT_IMPLEMENTED;
}

NS_IMETHODIMP
nsHtml5Parser::SetParser(nsIParser* aParser)
{
  NS_NOTREACHED("No one should be setting a parser on the HTML5 pseudosink.");
  return NS_ERROR_NOT_IMPLEMENTED;
}

void
nsHtml5Parser::FlushPendingNotifications(mozFlushType aType)
{
}

NS_IMETHODIMP
nsHtml5Parser::SetDocumentCharset(nsACString& aCharset)
{
  if (mDocShell) {
    // the following logic to get muCV is copied from
    // nsHTMLDocument::StartDocumentLoad
    // We need to call muCV->SetPrevDocCharacterSet here in case
    // the charset is detected by parser DetectMetaTag
    nsCOMPtr<nsIMarkupDocumentViewer> muCV;
    nsCOMPtr<nsIContentViewer> cv;
    mDocShell->GetContentViewer(getter_AddRefs(cv));
    if (cv) {
      muCV = do_QueryInterface(cv);
    } else {
      // in this block of code, if we get an error result, we return
      // it but if we get a null pointer, that's perfectly legal for
      // parent and parentContentViewer
      nsCOMPtr<nsIDocShellTreeItem> docShellAsItem =
        do_QueryInterface(mDocShell);
      NS_ENSURE_TRUE(docShellAsItem, NS_ERROR_FAILURE);
      nsCOMPtr<nsIDocShellTreeItem> parentAsItem;
      docShellAsItem->GetSameTypeParent(getter_AddRefs(parentAsItem));
      nsCOMPtr<nsIDocShell> parent(do_QueryInterface(parentAsItem));
      if (parent) {
        nsCOMPtr<nsIContentViewer> parentContentViewer;
        nsresult rv =
          parent->GetContentViewer(getter_AddRefs(parentContentViewer));
        if (NS_SUCCEEDED(rv) && parentContentViewer) {
          muCV = do_QueryInterface(parentContentViewer);
        }
      }
    }
    if (muCV) {
      muCV->SetPrevDocCharacterSet(aCharset);
    }
  }
  if (mDocument) {
    mDocument->SetDocumentCharacterSet(aCharset);
  }
  return NS_OK;
}

nsISupports*
nsHtml5Parser::GetTarget()
{
  return mDocument;
}

// not from interface
void
nsHtml5Parser::HandleParserContinueEvent(nsHtml5ParserContinueEvent* ev)
{
  // Ignore any revoked continue events...
  if (mContinueEvent != ev)
    return;
  mContinueEvent = nsnull;
  NS_ASSERTION(!IsScriptExecutingImpl(), "Interrupted in the middle of a script?");
  ContinueInterruptedParsing();
}

NS_IMETHODIMP
nsHtml5Parser::Notify(const char* aCharset, nsDetectionConfident aConf)
{
  if (aConf == eBestAnswer || aConf == eSureAnswer) {
    mCharset.Assign(aCharset);
    mCharsetSource = kCharsetFromAutoDetection;
    SetDocumentCharset(mCharset);
  }
  return NS_OK;
}

nsresult
nsHtml5Parser::SetupDecodingAndWriteSniffingBufferAndCurrentSegment(const PRUint8* aFromSegment, // can be null
                                                                    PRUint32 aCount,
                                                                    PRUint32* aWriteCount)
{
  nsresult rv = NS_OK;
  nsCOMPtr<nsICharsetConverterManager> convManager = do_GetService(NS_CHARSETCONVERTERMANAGER_CONTRACTID, &rv);
  NS_ENSURE_SUCCESS(rv, rv);
  rv = convManager->GetUnicodeDecoder(mCharset.get(), getter_AddRefs(mUnicodeDecoder));
  if (rv == NS_ERROR_UCONV_NOCONV) {
    mCharset.Assign("windows-1252"); // lower case is the raw form
    mCharsetSource = kCharsetFromWeakDocTypeDefault;
    rv = convManager->GetUnicodeDecoderRaw(mCharset.get(), getter_AddRefs(mUnicodeDecoder));
    SetDocumentCharset(mCharset);
  }
  NS_ENSURE_SUCCESS(rv, rv);
  mUnicodeDecoder->SetInputErrorBehavior(nsIUnicodeDecoder::kOnError_Recover);
  return WriteSniffingBufferAndCurrentSegment(aFromSegment, aCount, aWriteCount);
}

nsresult
nsHtml5Parser::WriteSniffingBufferAndCurrentSegment(const PRUint8* aFromSegment, // can be null
                                                    PRUint32 aCount,
                                                    PRUint32* aWriteCount)
{
  nsresult rv = NS_OK;
  if (mSniffingBuffer) {
    PRUint32 writeCount;
    rv = WriteStreamBytes(mSniffingBuffer, mSniffingLength, &writeCount);
    NS_ENSURE_SUCCESS(rv, rv);
    mSniffingBuffer = nsnull;
  }
  mMetaScanner = nsnull;
  if (aFromSegment) {
    rv = WriteStreamBytes(aFromSegment, aCount, aWriteCount);
  }
  return rv;
}

nsresult
nsHtml5Parser::SetupDecodingFromBom(const char* aCharsetName, const char* aDecoderCharsetName)
{
  nsresult rv = NS_OK;
  nsCOMPtr<nsICharsetConverterManager> convManager = do_GetService(NS_CHARSETCONVERTERMANAGER_CONTRACTID, &rv);
  NS_ENSURE_SUCCESS(rv, rv);
  rv = convManager->GetUnicodeDecoderRaw(aDecoderCharsetName, getter_AddRefs(mUnicodeDecoder));
  NS_ENSURE_SUCCESS(rv, rv);
  mCharset.Assign(aCharsetName);
  mCharsetSource = kCharsetFromByteOrderMark;
  SetDocumentCharset(mCharset);
  mSniffingBuffer = nsnull;
  mMetaScanner = nsnull;
  mBomState = BOM_SNIFFING_OVER;
  return rv;
}

nsresult
nsHtml5Parser::FinalizeSniffing(const PRUint8* aFromSegment, // can be null
                                PRUint32 aCount,
                                PRUint32* aWriteCount,
                                PRUint32 aCountToSniffingLimit)
{
  // meta scan failed.
  if (mCharsetSource >= kCharsetFromHintPrevDoc) {
    return SetupDecodingAndWriteSniffingBufferAndCurrentSegment(aFromSegment, aCount, aWriteCount);
  }
  // maybe try chardet now; instantiation copied from nsDOMFile
  const nsAdoptingString& detectorName = nsContentUtils::GetLocalizedStringPref("intl.charset.detector");
  if (!detectorName.IsEmpty()) {
    nsCAutoString detectorContractID;
    detectorContractID.AssignLiteral(NS_CHARSET_DETECTOR_CONTRACTID_BASE);
    AppendUTF16toUTF8(detectorName, detectorContractID);
    nsCOMPtr<nsICharsetDetector> detector = do_CreateInstance(detectorContractID.get());
    if (detector) {
      nsresult rv = detector->Init(this);
      NS_ENSURE_SUCCESS(rv, rv);
      PRBool dontFeed = PR_FALSE;
      if (mSniffingBuffer) {
        rv = detector->DoIt((const char*)mSniffingBuffer.get(), mSniffingLength, &dontFeed);
        NS_ENSURE_SUCCESS(rv, rv);
      }
      if (!dontFeed && aFromSegment) {
        rv = detector->DoIt((const char*)aFromSegment, aCountToSniffingLimit, &dontFeed);
        NS_ENSURE_SUCCESS(rv, rv);
      }
      rv = detector->Done();
      NS_ENSURE_SUCCESS(rv, rv);
      // fall thru; callback may have changed charset
    } else {
      NS_ERROR("Could not instantiate charset detector.");
    }
  }
  if (mCharsetSource == kCharsetUninitialized) {
    // Hopefully this case is never needed, but dealing with it anyway
    mCharset.Assign("windows-1252");
    mCharsetSource = kCharsetFromWeakDocTypeDefault;
    SetDocumentCharset(mCharset);
  }
  return SetupDecodingAndWriteSniffingBufferAndCurrentSegment(aFromSegment, aCount, aWriteCount);
}

nsresult
nsHtml5Parser::SniffStreamBytes(const PRUint8* aFromSegment,
                                PRUint32 aCount,
                                PRUint32* aWriteCount)
{
  nsresult rv = NS_OK;
  PRUint32 writeCount;
  for (PRUint32 i = 0; i < aCount; i++) {
    switch (mBomState) {
      case BOM_SNIFFING_NOT_STARTED:
        NS_ASSERTION(i == 0, "Bad BOM sniffing state.");
        switch (*aFromSegment) {
          case 0xEF:
            mBomState = SEEN_UTF_8_FIRST_BYTE;
            break;
          case 0xFF:
            mBomState = SEEN_UTF_16_LE_FIRST_BYTE;
            break;
          case 0xFE:
            mBomState = SEEN_UTF_16_BE_FIRST_BYTE;
            break;
          default:
            mBomState = BOM_SNIFFING_OVER;
            break;
        }
        break;
      case SEEN_UTF_16_LE_FIRST_BYTE:
        if (aFromSegment[i] == 0xFE) {
          rv = SetupDecodingFromBom("UTF-16", "UTF-16LE"); // upper case is the raw form
          NS_ENSURE_SUCCESS(rv, rv);
          PRUint32 count = aCount - (i + 1);
          rv = WriteStreamBytes(aFromSegment + (i + 1), count, &writeCount);
          NS_ENSURE_SUCCESS(rv, rv);
          *aWriteCount = writeCount + (i + 1);
          return rv;
        }
        mBomState = BOM_SNIFFING_OVER;
        break;
      case SEEN_UTF_16_BE_FIRST_BYTE:
        if (aFromSegment[i] == 0xFF) {
          rv = SetupDecodingFromBom("UTF-16", "UTF-16BE"); // upper case is the raw form
          NS_ENSURE_SUCCESS(rv, rv);
          PRUint32 count = aCount - (i + 1);
          rv = WriteStreamBytes(aFromSegment + (i + 1), count, &writeCount);
          NS_ENSURE_SUCCESS(rv, rv);
          *aWriteCount = writeCount + (i + 1);
          return rv;
        }
        mBomState = BOM_SNIFFING_OVER;
        break;
      case SEEN_UTF_8_FIRST_BYTE:
        if (aFromSegment[i] == 0xBB) {
          mBomState = SEEN_UTF_8_SECOND_BYTE;
        } else {
          mBomState = BOM_SNIFFING_OVER;
        }
        break;
      case SEEN_UTF_8_SECOND_BYTE:
        if (aFromSegment[i] == 0xBF) {
          rv = SetupDecodingFromBom("UTF-8", "UTF-8"); // upper case is the raw form
          NS_ENSURE_SUCCESS(rv, rv);
          PRUint32 count = aCount - (i + 1);
          rv = WriteStreamBytes(aFromSegment + (i + 1), count, &writeCount);
          NS_ENSURE_SUCCESS(rv, rv);
          *aWriteCount = writeCount + (i + 1);
          return rv;
        }
        mBomState = BOM_SNIFFING_OVER;
        break;
      default:
        goto bom_loop_end;
    }
  }
  // if we get here, there either was no BOM or the BOM sniffing isn't complete yet
  bom_loop_end:
  
  if (!mMetaScanner) {
    mMetaScanner = new nsHtml5MetaScanner();
  }
  
  if (mSniffingLength + aCount >= NS_HTML5_PARSER_SNIFFING_BUFFER_SIZE) {
    // this is the last buffer
    PRUint32 countToSniffingLimit = NS_HTML5_PARSER_SNIFFING_BUFFER_SIZE - mSniffingLength;
    nsHtml5ByteReadable readable(aFromSegment, aFromSegment + countToSniffingLimit);
    mMetaScanner->sniff(&readable, getter_AddRefs(mUnicodeDecoder), mCharset);
    if (mUnicodeDecoder) {
      mUnicodeDecoder->SetInputErrorBehavior(nsIUnicodeDecoder::kOnError_Recover);
      // meta scan successful
      mCharsetSource = kCharsetFromMetaPrescan;
      SetDocumentCharset(mCharset);
      mMetaScanner = nsnull;
      return WriteSniffingBufferAndCurrentSegment(aFromSegment, aCount, aWriteCount);
    }
    return FinalizeSniffing(aFromSegment, aCount, aWriteCount, countToSniffingLimit);
  }

  // not the last buffer
  nsHtml5ByteReadable readable(aFromSegment, aFromSegment + aCount);
  mMetaScanner->sniff(&readable, getter_AddRefs(mUnicodeDecoder), mCharset);
  if (mUnicodeDecoder) {
    // meta scan successful
    mCharsetSource = kCharsetFromMetaPrescan;
    SetDocumentCharset(mCharset);
    mMetaScanner = nsnull;
    return WriteSniffingBufferAndCurrentSegment(aFromSegment, aCount, aWriteCount);
  }
  if (!mSniffingBuffer) {
    mSniffingBuffer = new PRUint8[NS_HTML5_PARSER_SNIFFING_BUFFER_SIZE];
  }
  memcpy(mSniffingBuffer + mSniffingLength, aFromSegment, aCount);
  mSniffingLength += aCount;
  *aWriteCount = aCount;
  return NS_OK;
}

nsresult
nsHtml5Parser::WriteStreamBytes(const PRUint8* aFromSegment,
                                PRUint32 aCount,
                                PRUint32* aWriteCount)
{
  // mLastBuffer always points to a buffer of the size NS_HTML5_PARSER_READ_BUFFER_SIZE.
  if (mLastBuffer->getEnd() == NS_HTML5_PARSER_READ_BUFFER_SIZE) {
    mLastBuffer = (mLastBuffer->next = new nsHtml5UTF16Buffer(NS_HTML5_PARSER_READ_BUFFER_SIZE));
  }
  PRUint32 totalByteCount = 0;
  for (;;) {
    PRInt32 end = mLastBuffer->getEnd();
    PRInt32 byteCount = aCount - totalByteCount;
    PRInt32 utf16Count = NS_HTML5_PARSER_READ_BUFFER_SIZE - end;

    NS_ASSERTION(utf16Count, "Trying to convert into a buffer with no free space!");

    nsresult convResult = mUnicodeDecoder->Convert((const char*)aFromSegment, &byteCount, mLastBuffer->getBuffer() + end, &utf16Count);

    end += utf16Count;
    mLastBuffer->setEnd(end);
    totalByteCount += byteCount;
    aFromSegment += byteCount;

    NS_ASSERTION(mLastBuffer->getEnd() <= NS_HTML5_PARSER_READ_BUFFER_SIZE, "The Unicode decoder wrote too much data.");

    if (NS_FAILED(convResult)) {
      if (totalByteCount < aCount) { // mimicking nsScanner even though this seems wrong
        ++totalByteCount;
        ++aFromSegment;
      }
      mLastBuffer->getBuffer()[end] = 0xFFFD;
      ++end;
      mLastBuffer->setEnd(end);
      if (end == NS_HTML5_PARSER_READ_BUFFER_SIZE) {
          mLastBuffer = (mLastBuffer->next = new nsHtml5UTF16Buffer(NS_HTML5_PARSER_READ_BUFFER_SIZE));
      }
      mUnicodeDecoder->Reset();
      if (totalByteCount == aCount) {
        *aWriteCount = totalByteCount;
        return NS_OK;
      }
    } else if (convResult == NS_PARTIAL_MORE_OUTPUT) {
      mLastBuffer = (mLastBuffer->next = new nsHtml5UTF16Buffer(NS_HTML5_PARSER_READ_BUFFER_SIZE));
      NS_ASSERTION(totalByteCount < aCount, "The Unicode decoder has consumed too many bytes.");
    } else {
      NS_ASSERTION(totalByteCount == aCount, "The Unicode decoder consumed the wrong number of bytes.");
      *aWriteCount = totalByteCount;
      return NS_OK;
    }
  }
}

void
nsHtml5Parser::ParseUntilSuspend()
{
  NS_PRECONDITION(!mFragmentMode, "ParseUntilSuspend called in fragment mode.");
  NS_PRECONDITION(!mNeedsCharsetSwitch, "ParseUntilSuspend called when charset switch needed.");

  if (mBlocked) {
    return;
  }

  switch (mLifeCycle) {
    case TERMINATED:
      return;
    case NOT_STARTED:
      mTreeBuilder->setScriptingEnabled(IsScriptEnabled(mDocument, mDocShell));
      mTokenizer->start();
      mLifeCycle = PARSING;
      mParser = this;
      break;
    default:
      break;
  }

  WillResumeImpl();
  WillParseImpl();
  mSuspending = PR_FALSE;
  for (;;) {
    if (!mFirstBuffer->hasMore()) {
      if (mFirstBuffer == mLastBuffer) {
        switch (mLifeCycle) {
          case TERMINATED:
            // something like cache manisfests stopped the parse in mid-flight
            return;
          case PARSING:
            // never release the last buffer. instead just zero its indeces for refill
            mFirstBuffer->setStart(0);
            mFirstBuffer->setEnd(0);
            return; // no more data for now but expecting more
          case STREAM_ENDING:
            DidBuildModel(PR_FALSE);
            return; // no more data and not expecting more
          default:
            NS_NOTREACHED("It should be impossible to reach this.");
            return;
        }
      } else {
        nsHtml5UTF16Buffer* oldBuf = mFirstBuffer;
        mFirstBuffer = mFirstBuffer->next;
        delete oldBuf;
        continue;
      }
    }

    if (mBlocked || (mLifeCycle == TERMINATED)) {
      return;
    }

#ifdef GATHER_DOCWRITE_STATISTICS
    if (mSnapshot && mFirstBuffer->key == GetRootContextKey()) {
      if (mTokenizer->isInDataState()) {
        if (mTreeBuilder->snapshotMatches(mSnapshot)) {
          sTreeSafeDocWrites++;
        } else {
          sTokenSafeDocWrites++;
        }
      } else {
        sUnsafeDocWrites++;
      }
      delete mSnapshot;
      mSnapshot = nsnull;
    }
#endif

    // now we have a non-empty buffer
    mFirstBuffer->adjust(mLastWasCR);
    mLastWasCR = PR_FALSE;
    if (mFirstBuffer->hasMore()) {
      mLastWasCR = mTokenizer->tokenizeBuffer(mFirstBuffer);
      NS_ASSERTION(!(mScriptElement && mNeedsCharsetSwitch), "Can't have both script and charset switch.");
      if (mScriptElement) {
        mTreeBuilder->Flush();
        ExecuteScript();
      } else if (mNeedsCharsetSwitch) {
        if (PerformCharsetSwitch() == NS_ERROR_HTMLPARSER_STOPPARSING) {
          return;
        } else {
          // let's continue if we failed to restart
          mNeedsCharsetSwitch = PR_FALSE;
        }
      }
      if (mBlocked) {
        WillInterruptImpl();
        return;
      }
      if (mSuspending) {
        MaybePostContinueEvent();
        WillInterruptImpl();
        return;
      }
    }
    continue;
  }
}

nsresult
nsHtml5Parser::PerformCharsetSwitch()
{
  // this code comes from nsObserverBase.cpp
  nsresult rv = NS_OK;
  nsCOMPtr<nsIWebShellServices> wss = do_QueryInterface(mDocShell);
  if (!wss) {
    return NS_ERROR_HTMLPARSER_CONTINUE;
  }
#ifndef DONT_INFORM_WEBSHELL
  // ask the webshellservice to load the URL
  if (NS_FAILED(rv = wss->SetRendering(PR_FALSE))) {
    // do nothing and fall thru
  } else if (NS_FAILED(rv = wss->StopDocumentLoad())) {
    rv = wss->SetRendering(PR_TRUE); // turn on the rendering so at least we will see something.
  } else if (NS_FAILED(rv = wss->ReloadDocument(mPendingCharset.get(), kCharsetFromMetaTag))) {
    rv = wss->SetRendering(PR_TRUE); // turn on the rendering so at least we will see something.
  } else {
    rv = NS_ERROR_HTMLPARSER_STOPPARSING; // We're reloading a new document...stop loading the current.
  }
#endif
  // if our reload request is not accepted, we should tell parser to go on
  if (rv != NS_ERROR_HTMLPARSER_STOPPARSING)
    rv = NS_ERROR_HTMLPARSER_CONTINUE;
  return rv;
}

/**
 * This method executes a script element set by nsHtml5TreeBuilder. The reason
 * why this code is here and not in the tree builder is to allow the control
 * to return from the tokenizer before scripts run. This way, the tokenizer
 * is not invoked re-entrantly although the parser is.
 */
void
nsHtml5Parser::ExecuteScript()
{
  NS_PRECONDITION(mScriptElement, "Trying to run a script without having one!");
#ifdef GATHER_DOCWRITE_STATISTICS
  if (!mSnapshot) {
    mSnapshot = mTreeBuilder->newSnapshot();
  }
#endif
  nsCOMPtr<nsIScriptElement> sele = do_QueryInterface(mScriptElement);
   // Notify our document that we're loading this script.
  nsCOMPtr<nsIHTMLDocument> htmlDocument = do_QueryInterface(mDocument);
  NS_ASSERTION(htmlDocument, "Document didn't QI into HTML document.");
  htmlDocument->ScriptLoading(sele);
   // Copied from nsXMLContentSink
  // Now tell the script that it's ready to go. This may execute the script
  // or return NS_ERROR_HTMLPARSER_BLOCK. Or neither if the script doesn't
  // need executing.
  nsresult rv = mScriptElement->DoneAddingChildren(PR_TRUE);
  // If the act of insertion evaluated the script, we're fine.
  // Else, block the parser till the script has loaded.
  if (rv == NS_ERROR_HTMLPARSER_BLOCK) {
    mScriptElements.AppendObject(sele);
    mBlocked = PR_TRUE;
  } else {
    // This may have already happened if the script executed, but in case
    // it didn't then remove the element so that it doesn't get stuck forever.
    htmlDocument->ScriptExecuted(sele);
  }
  mScriptElement = nsnull;
}

void
nsHtml5Parser::MaybePostContinueEvent()
{
  NS_PRECONDITION(mLifeCycle != TERMINATED, 
                  "Tried to post continue event when the parser is done.");
  if (mContinueEvent) {
    return; // we already have a pending event
  }
  // This creates a reference cycle between this and the event that is
  // broken when the event fires.
  nsCOMPtr<nsIRunnable> event = new nsHtml5ParserContinueEvent(this);
  if (NS_FAILED(NS_DispatchToCurrentThread(event))) {
    NS_WARNING("failed to dispatch parser continuation event");
  } else {
    mContinueEvent = event;
  }
}

void
nsHtml5Parser::Suspend()
{
  mSuspending = PR_TRUE;
}

nsresult
nsHtml5Parser::Initialize(nsIDocument* aDoc,
                          nsIURI* aURI,
                          nsISupports* aContainer,
                          nsIChannel* aChannel)
{
  nsresult rv = nsContentSink::Init(aDoc, aURI, aContainer, aChannel);
  NS_ENSURE_SUCCESS(rv, rv);
  aDoc->AddObserver(this);
  return NS_OK;
}

nsresult
nsHtml5Parser::ProcessBASETag(nsIContent* aContent)
{
  NS_ASSERTION(aContent, "missing base-element");
  nsresult rv = NS_OK;
  if (mDocument) {
    nsAutoString value;
    if (aContent->GetAttr(kNameSpaceID_None, nsHtml5Atoms::target, value)) {
      mDocument->SetBaseTarget(value);
    }
    if (aContent->GetAttr(kNameSpaceID_None, nsHtml5Atoms::href, value)) {
      nsCOMPtr<nsIURI> baseURI;
      rv = NS_NewURI(getter_AddRefs(baseURI), value);
      if (NS_SUCCEEDED(rv)) {
        rv = mDocument->SetBaseURI(baseURI); // The document checks if it is legal to set this base
        if (NS_SUCCEEDED(rv)) {
          mDocumentBaseURI = mDocument->GetBaseURI();
        }
      }
    }
  }
  return rv;
}

void
nsHtml5Parser::UpdateStyleSheet(nsIContent* aElement)
{
  nsCOMPtr<nsIStyleSheetLinkingElement> ssle(do_QueryInterface(aElement));
  if (ssle) {
    ssle->SetEnableUpdates(PR_TRUE);
    PRBool willNotify;
    PRBool isAlternate;
    nsresult rv = ssle->UpdateStyleSheet(this, &willNotify, &isAlternate);
    if (NS_SUCCEEDED(rv) && willNotify && !isAlternate) {
      ++mPendingSheetCount;
      mScriptLoader->AddExecuteBlocker();
    }
  }
}

void
nsHtml5Parser::SetScriptElement(nsIContent* aScript)
{
  mScriptElement = aScript;
}

// nsContentSink overrides

void
nsHtml5Parser::UpdateChildCounts()
{
  // No-op
}

nsresult
nsHtml5Parser::FlushTags()
{
    return NS_OK;
}

void
nsHtml5Parser::PostEvaluateScript(nsIScriptElement *aElement)
{
  nsCOMPtr<nsIHTMLDocument> htmlDocument = do_QueryInterface(mDocument);
  NS_ASSERTION(htmlDocument, "Document didn't QI into HTML document.");
  htmlDocument->ScriptExecuted(aElement);
}


#ifdef GATHER_DOCWRITE_STATISTICS
PRUint32 nsHtml5Parser::sUnsafeDocWrites = 0;
PRUint32 nsHtml5Parser::sTokenSafeDocWrites = 0;
PRUint32 nsHtml5Parser::sTreeSafeDocWrites = 0;
#endif