Commit 8915c150 authored by Daniel Veillard's avatar Daniel Veillard

strengthen some of the internal parser limits, add an XML_PARSE_HUGE

* include/libxml/parser.h parser.c xmllint.c: strengthen some
  of the internal parser limits, add an XML_PARSE_HUGE option
  to bypass them all. More internal parser limits will still need
  to be added.
Daniel

svn path=/trunk/; revision=3777
parent bf9c1dad
Tue Aug 26 15:02:58 CEST 2008 Daniel Veillard <daniel@veillard.com>
* include/libxml/parser.h parser.c xmllint.c: strengthen some
of the internal parser limits, add an XML_PARSE_HUGE option
to bypass them all. More internal parser limits will still need
to be added.
Tue Aug 26 09:42:08 CEST 2008 Daniel Veillard <daniel@veillard.com>
* Makefile.am: add the testchar to 'make check'
......
......@@ -1094,7 +1094,8 @@ typedef enum {
the tree allowed afterwards (will possibly
crash if you try to modify the tree) */
XML_PARSE_OLD10 = 1<<17,/* parse using XML-1.0 before update 5 */
XML_PARSE_NOBASEFIX = 1<<18 /* do not fixup XINCLUDE xml:base uris */
XML_PARSE_NOBASEFIX = 1<<18,/* do not fixup XINCLUDE xml:base uris */
XML_PARSE_HUGE = 1<<19 /* relax any hardcoded limit from the parser */
} xmlParserOption;
XMLPUBFUN void XMLCALL
......
......@@ -83,11 +83,12 @@
/**
* xmlParserMaxDepth:
*
* arbitrary depth limit for the XML documents that we allow to
* process. This is not a limitation of the parser but a safety
* boundary feature.
* arbitrary depth limit for the XML documents that we allow to
* process. This is not a limitation of the parser but a safety
* boundary feature. It can be disabled with the XML_PARSE_HUGE
* parser option.
*/
unsigned int xmlParserMaxDepth = 1024;
unsigned int xmlParserMaxDepth = 256;
#define SAX2 1
......@@ -1452,9 +1453,10 @@ nodePush(xmlParserCtxtPtr ctxt, xmlNodePtr value)
ctxt->nodeTab = tmp;
ctxt->nodeMax *= 2;
}
if (((unsigned int) ctxt->nodeNr) > xmlParserMaxDepth) {
if ((((unsigned int) ctxt->nodeNr) > xmlParserMaxDepth) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive depth in document: change xmlParserMaxDepth = %d\n",
"Excessive depth in document: %d use XML_PARSE_HUGE option\n",
xmlParserMaxDepth);
ctxt->instate = XML_PARSER_EOF;
return(0);
......@@ -1463,6 +1465,7 @@ nodePush(xmlParserCtxtPtr ctxt, xmlNodePtr value)
ctxt->node = value;
return (ctxt->nodeNr++);
}
/**
* nodePop:
* @ctxt: an XML parser context
......@@ -2379,7 +2382,8 @@ xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len,
return(NULL);
last = str + len;
if ((ctxt->depth > 40) || (ctxt->nbentities >= 500000)) {
if (((ctxt->depth > 20) || (ctxt->nbentities >= 100000)) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL);
return(NULL);
}
......@@ -2417,7 +2421,8 @@ xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len,
"String decoding Entity Reference: %.30s\n",
str);
ent = xmlParseStringEntityRef(ctxt, &str);
if (ctxt->lastError.code == XML_ERR_ENTITY_LOOP)
if ((ctxt->lastError.code == XML_ERR_ENTITY_LOOP) ||
(ctxt->lastError.code == XML_ERR_INTERNAL_ERROR))
goto int_error;
ctxt->nbentities++;
if (ent != NULL)
......@@ -2505,6 +2510,14 @@ xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len,
c = CUR_SCHAR(str, l);
else
c = 0;
if ((nbchars > 100000) &&
(ctxt->instate == XML_PARSER_ATTRIBUTE_VALUE) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive lenght of attribute: %d use XML_PARSE_HUGE option\n",
nbchars);
goto int_error;
}
}
buffer[nbchars++] = 0;
return(buffer);
......@@ -3647,6 +3660,13 @@ xmlParseAttValueComplex(xmlParserCtxtPtr ctxt, int *attlen, int normalize) {
}
GROW;
c = CUR_CHAR(l);
if ((len > 100000) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive lenght of attribute: %d use XML_PARSE_HUGE option\n",
len);
goto int_error;
}
}
if ((in_space) && (normalize)) {
while (buf[len - 1] == 0x20) len--;
......@@ -3669,6 +3689,7 @@ xmlParseAttValueComplex(xmlParserCtxtPtr ctxt, int *attlen, int normalize) {
mem_error:
xmlErrMemory(ctxt, NULL);
int_error:
if (buf != NULL)
xmlFree(buf);
if (rep != NULL)
......@@ -6494,7 +6515,8 @@ xmlParseReference(xmlParserCtxtPtr ctxt) {
if (!ctxt->wellFormed)
return;
ctxt->nbentities++;
if (ctxt->nbentities >= 500000) {
if ((ctxt->nbentities >= 100000) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL);
return;
}
......@@ -9135,10 +9157,11 @@ xmlParseElement(xmlParserCtxtPtr ctxt) {
xmlNodePtr ret;
int nsNr = ctxt->nsNr;
if ((unsigned int) ctxt->nameNr > xmlParserMaxDepth) {
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive depth in document: change xmlParserMaxDepth = %d\n",
xmlParserMaxDepth);
if (((unsigned int) ctxt->nameNr > xmlParserMaxDepth) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive depth in document: %d use XML_PARSE_HUGE option\n",
xmlParserMaxDepth);
ctxt->instate = XML_PARSER_EOF;
return;
}
......@@ -11826,7 +11849,8 @@ xmlParseCtxtExternalEntity(xmlParserCtxtPtr ctx, const xmlChar *URL,
if (ctx == NULL) return(-1);
if ((ctx->depth > 40) || (ctx->nbentities >= 500000)) {
if (((ctx->depth > 20) || (ctx->nbentities >= 100000)) &&
((ctx->options & XML_PARSE_HUGE) == 0)) {
return(XML_ERR_ENTITY_LOOP);
}
......@@ -12036,13 +12060,12 @@ xmlParseExternalEntityPrivate(xmlDocPtr doc, xmlParserCtxtPtr oldctxt,
xmlChar start[4];
xmlCharEncoding enc;
if ((depth > 40) ||
((oldctxt != NULL) && (oldctxt->nbentities >= 500000))) {
if (((depth > 20) ||
((oldctxt != NULL) && (oldctxt->nbentities >= 100000))) &&
((oldctxt == NULL) || (oldctxt->options & XML_PARSE_HUGE) == 0)) {
return(XML_ERR_ENTITY_LOOP);
}
if (list != NULL)
*list = NULL;
if ((URL == NULL) && (ID == NULL))
......@@ -12282,7 +12305,8 @@ xmlParseBalancedChunkMemoryInternal(xmlParserCtxtPtr oldctxt,
int size;
xmlParserErrors ret = XML_ERR_OK;
if ((oldctxt->depth > 40) || (oldctxt->nbentities >= 500000)) {
if (((oldctxt->depth > 20) || (oldctxt->nbentities >= 100000)) &&
((oldctxt->options & XML_PARSE_HUGE) == 0)) {
return(XML_ERR_ENTITY_LOOP);
}
......@@ -12661,7 +12685,7 @@ xmlParseBalancedChunkMemoryRecover(xmlDocPtr doc, xmlSAXHandlerPtr sax,
int size;
int ret = 0;
if (depth > 40) {
if (depth > 20) {
return(XML_ERR_ENTITY_LOOP);
}
......@@ -13971,6 +13995,14 @@ xmlCtxtUseOptionsInternal(xmlParserCtxtPtr ctxt, int options, const char *encodi
ctxt->options |= XML_PARSE_OLD10;
options -= XML_PARSE_OLD10;
}
if (options & XML_PARSE_NOBASEFIX) {
ctxt->options |= XML_PARSE_NOBASEFIX;
options -= XML_PARSE_NOBASEFIX;
}
if (options & XML_PARSE_HUGE) {
ctxt->options |= XML_PARSE_HUGE;
options -= XML_PARSE_HUGE;
}
ctxt->linenumbers = 1;
return (options);
}
......
......@@ -2833,6 +2833,7 @@ static void usage(const char *name) {
printf("\t--copy : used to test the internal copy implementation\n");
#endif /* LIBXML_TREE_ENABLED */
printf("\t--recover : output what was parsable on broken XML documents\n");
printf("\t--huge : remove any internal arbitrary parser limits\n");
printf("\t--noent : substitute entity references by their value\n");
printf("\t--noout : don't output the result tree\n");
printf("\t--path 'paths': provide a set of paths for resources\n");
......@@ -2972,6 +2973,9 @@ main(int argc, char **argv) {
(!strcmp(argv[i], "--recover"))) {
recovery++;
options |= XML_PARSE_RECOVER;
} else if ((!strcmp(argv[i], "-huge")) ||
(!strcmp(argv[i], "--huge"))) {
options |= XML_PARSE_HUGE;
} else if ((!strcmp(argv[i], "-noent")) ||
(!strcmp(argv[i], "--noent"))) {
noent++;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment