lib/libxml2
1
0
Fork 0
mirror of https://gitlab.gnome.org/GNOME/libxml2.git synced 2025-07-19 05:43:19 +03:00
Code Activity

fixed a nasty bug #119387, bad heuristic from the progressive HTML parser

Browse Source 
* HTMLparser.c: fixed a nasty bug #119387, bad heuristic from
  the progressive HTML parser front-end on large character data
  island leading to an erroneous end of data detection by the
  parser. Some cleanup too to get closer from the XML progressive
  parser.
Daniel
This commit is contained in:
Daniel Veillard
2003-08-09 11:44:50 +00:00
parent 6d13f33d87
commit 14f752c2b7
2 changed files with 33 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
ChangeLog
View File

@ -1,3 +1,11 @@
Sat Aug 9 13:41:21 CEST 2003 Daniel Veillard <daniel@veillard.com>
* HTMLparser.c: fixed a nasty bug #119387, bad heuristic from
the progressive HTML parser front-end on large character data
island leading to an erroneous end of data detection by the
parser. Some cleanup too to get closer from the XML progressive
parser.
Sat Aug 9 00:42:47 HKT 2003 William Brack <wbrack@mmm.com.hk> Sat Aug 9 00:42:47 HKT 2003 William Brack <wbrack@mmm.com.hk>
* win32/configure.js: Added in support for the ISO8859X * win32/configure.js: Added in support for the ISO8859X

33
HTMLparser.c
View File

@ -4950,19 +4950,15 @@ htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
/* TODO: check generation of subtrees if noent !!! */ /* TODO: check generation of subtrees if noent !!! */
htmlParseReference(ctxt); htmlParseReference(ctxt);
} else { } else {
/* TODO Avoid the extra copy, handle directly !!!!!! */
/* /*
* Goal of the following test is: * check that the text sequence is complete
* - minimize calls to the SAX 'character' callback * before handing out the data to the parser
* when they are mergeable * to avoid problems with erroneous end of
* data detection.
*/ */
if ((ctxt->inputNr == 1) &&
(avail < HTML_PARSER_BIG_BUFFER_SIZE)) {
if ((!terminate) && if ((!terminate) &&
(htmlParseLookupSequence( (htmlParseLookupSequence(ctxt, '<', 0, 0, 0) < 0))
ctxt, '<', 0, 0, 0) < 0))
goto done; goto done;
}
ctxt->checkIndex = 0; ctxt->checkIndex = 0;
#ifdef DEBUG_PUSH #ifdef DEBUG_PUSH
xmlGenericError(xmlGenericErrorContext, xmlGenericError(xmlGenericErrorContext,
@ -5160,12 +5156,27 @@ htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size); xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
#endif #endif
#if 0
if ((terminate) || (ctxt->input->buf->buffer->use > 80)) if ((terminate) || (ctxt->input->buf->buffer->use > 80))
htmlParseTryOrFinish(ctxt, terminate); htmlParseTryOrFinish(ctxt, terminate);
#endif
} else if (ctxt->instate != XML_PARSER_EOF) { } else if (ctxt->instate != XML_PARSER_EOF) {
xmlParserInputBufferPush(ctxt->input->buf, 0, ""); if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
htmlParseTryOrFinish(ctxt, terminate); xmlParserInputBufferPtr in = ctxt->input->buf;
if ((in->encoder != NULL) && (in->buffer != NULL) &&
(in->raw != NULL)) {
int nbchars;
nbchars = xmlCharEncInFunc(in->encoder, in->buffer, in->raw);
if (nbchars < 0) {
xmlGenericError(xmlGenericErrorContext,
"htmlParseChunk: encoder error\n");
return(XML_ERR_INVALID_ENCODING);
} }
}
}
}
htmlParseTryOrFinish(ctxt, terminate);
if (terminate) { if (terminate) {
if ((ctxt->instate != XML_PARSER_EOF) && if ((ctxt->instate != XML_PARSER_EOF) &&
(ctxt->instate != XML_PARSER_EPILOG) && (ctxt->instate != XML_PARSER_EPILOG) &&