lib/libxml2
1
0
Fork 0
mirror of https://gitlab.gnome.org/GNOME/libxml2.git synced 2025-07-29 11:41:22 +03:00
Code Activity

html: Ignore unexpected DOCTYPE declarations

Browse Source
This commit is contained in:
Nick Wellnhofer
2024-09-11 19:47:07 +02:00
parent 462bf0b7a5
commit be874d7831
2 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
HTMLparser.c
View File

@ -4343,6 +4343,7 @@ htmlParseDocument(htmlParserCtxtPtr ctxt) {
(UPP(4) == 'C') && (UPP(5) == 'T') && (UPP(4) == 'C') && (UPP(5) == 'T') &&
(UPP(6) == 'Y') && (UPP(7) == 'P') && (UPP(6) == 'Y') && (UPP(7) == 'P') &&
(UPP(8) == 'E')) { (UPP(8) == 'E')) {
ctxt->instate = XML_PARSER_MISC;
htmlParseDocTypeDecl(ctxt); htmlParseDocTypeDecl(ctxt);
} }
SKIP_BLANKS; SKIP_BLANKS;
@ -4350,6 +4351,7 @@ htmlParseDocument(htmlParserCtxtPtr ctxt) {
/* /*
* Parse possible comments and PIs before any content * Parse possible comments and PIs before any content
*/ */
ctxt->instate = XML_PARSER_PROLOG;
while (CUR == '<') { while (CUR == '<') {
if ((NXT(1) == '!') && (NXT(2) == '-') && (NXT(3) == '-')) { if ((NXT(1) == '!') && (NXT(2) == '-') && (NXT(3) == '-')) {
SKIP(4); SKIP(4);
@ -4366,6 +4368,7 @@ htmlParseDocument(htmlParserCtxtPtr ctxt) {
/* /*
* Time to start parsing the tree itself * Time to start parsing the tree itself
*/ */
ctxt->instate = XML_PARSER_CONTENT;
htmlParseContent(ctxt); htmlParseContent(ctxt);
/* /*

4
SAX2.c
View File

@ -245,10 +245,10 @@ xmlSAX2InternalSubset(void *ctx, const xmlChar *name,
if (ctxt->myDoc == NULL) if (ctxt->myDoc == NULL)
return; return;
if ((ctxt->html) && (ctxt->instate != XML_PARSER_MISC))
return;
dtd = xmlGetIntSubset(ctxt->myDoc); dtd = xmlGetIntSubset(ctxt->myDoc);
if (dtd != NULL) { if (dtd != NULL) {
if (ctxt->html)
return;
xmlUnlinkNode((xmlNodePtr) dtd); xmlUnlinkNode((xmlNodePtr) dtd);
xmlFreeDtd(dtd); xmlFreeDtd(dtd);
ctxt->myDoc->intSubset = NULL; ctxt->myDoc->intSubset = NULL;