shaba/libxml2
1
0
Fork 0
mirror of https://gitlab.gnome.org/GNOME/libxml2.git synced 2025-03-23 02:50:08 +03:00
Code

Speed up HTML fuzzer

Browse Source 
htmlDocDumpMemory uses the "HTML" encoding if no other encoding was
specified in the source HTML. This encoding can be extremely slow
because of an inefficiency in htmlEntityValueLookup. Stop encoding
the output for now.
This commit is contained in:
Nick Wellnhofer 2021-02-07 13:57:49 +01:00
parent e6495e4789
commit ec808a4415
1 changed files with 11 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
fuzz/html.c
View File

@ -22,7 +22,7 @@ LLVMFuzzerTestOneInput(const char *data, size_t size) {
static const size_t maxChunkSize = 128;
htmlDocPtr doc;
htmlParserCtxtPtr ctxt;
xmlChar *out;
xmlOutputBufferPtr out;
const char *docBuffer;
size_t docSize, consumed, chunkSize;
int opts, outSize;
@ -39,9 +39,16 @@ LLVMFuzzerTestOneInput(const char *data, size_t size) {
/* Pull parser */
doc = htmlReadMemory(docBuffer, docSize, NULL, NULL, opts);
/* Also test the serializer. */
htmlDocDumpMemory(doc, &out, &outSize);
xmlFree(out);
/*
* Also test the serializer. Call htmlDocContentDumpOutput with our
* own buffer to avoid encoding the output. The HTML encoding is
* excruciatingly slow (see htmlEntityValueLookup).
*/
out = xmlAllocOutputBuffer(NULL);
htmlDocContentDumpOutput(out, doc, NULL);
xmlOutputBufferClose(out);
xmlFreeDoc(doc);
/* Push parser */