applied UTF-8 script parsing bug #310229 fix from Jiri Netolicky added the
* HTMLparser.c: applied UTF-8 script parsing bug #310229 fix from
Jiri Netolicky
* result/HTML/script2.html* test/HTML/script2.html: added the test
case from the regression suite
Daniel
diff --git a/HTMLparser.c b/HTMLparser.c
index c6115d0..d724620 100644
--- a/HTMLparser.c
+++ b/HTMLparser.c
@@ -2629,10 +2629,10 @@
htmlParseScript(htmlParserCtxtPtr ctxt) {
xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 1];
int nbchar = 0;
- xmlChar cur;
+ int cur,l;
SHRINK;
- cur = CUR;
+ cur = CUR_CHAR(l);
while (IS_CHAR_CH(cur)) {
if ((cur == '<') && (NXT(1) == '!') && (NXT(2) == '-') &&
(NXT(3) == '-')) {
@@ -2648,7 +2648,7 @@
}
nbchar = 0;
htmlParseComment(ctxt);
- cur = CUR;
+ cur = CUR_CHAR(l);
continue;
} else if ((cur == '<') && (NXT(1) == '/')) {
/*
@@ -2661,7 +2661,7 @@
((NXT(2) >= 'a') && (NXT(2) <= 'z')))
break; /* while */
}
- buf[nbchar++] = cur;
+ COPY_BUF(l,buf,nbchar,cur);
if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
if (ctxt->sax->cdataBlock!= NULL) {
/*
@@ -2673,8 +2673,8 @@
}
nbchar = 0;
}
- NEXT;
- cur = CUR;
+ NEXTL(l);
+ cur = CUR_CHAR(l);
}
if (!(IS_CHAR_CH(cur))) {
htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,