Fix more quadratic runtime issues in HTML push parser
Make sure that checkIndex is set when returning without match from
inside a comment. Also track parser state in htmlParseLookupChars.
Found by OSS-Fuzz.
diff --git a/HTMLparser.c b/HTMLparser.c
index 366c19b..9b12dd1 100644
--- a/HTMLparser.c
+++ b/HTMLparser.c
@@ -5205,7 +5205,7 @@
}
if (incomment) {
if (base + 3 > len)
- return (-1);
+ break;
if ((buf[base] == '-') && (buf[base + 1] == '-') &&
(buf[base + 2] == '>')) {
incomment = 0;
@@ -5294,8 +5294,11 @@
if (base < 0)
return (-1);
- if (ctxt->checkIndex > base)
+ if (ctxt->checkIndex > base) {
base = ctxt->checkIndex;
+ /* Abuse hasPErefs member to restore current state. */
+ incomment = ctxt->hasPErefs & 1 ? 1 : 0;
+ }
if (in->buf == NULL) {
buf = in->base;
@@ -5316,7 +5319,7 @@
}
if (incomment) {
if (base + 3 > len)
- return (-1);
+ break;
if ((buf[base] == '-') && (buf[base + 1] == '-') &&
(buf[base + 2] == '>')) {
incomment = 0;
@@ -5332,6 +5335,8 @@
}
}
ctxt->checkIndex = base;
+ /* Abuse hasPErefs member to track current state. */
+ ctxt->hasPErefs = incomment;
return (-1);
}
