[htdig-dev] Two problems with HTML.cc class?

[Neal R. <ne...@ri...>]

I'm finding that several things are wierd in the htdig/HTML.cc class

1) If a page has an ill-formed comment tag like this:
<!-- hennerik CVSweb $Revision: 1.64  0->

Everything after the start of the comment is eaten.. the entire page.. 
since the comment end is bad.  A browser handles this fine.

2) In the HTML::parse function

223     unsigned char       *text = (unsigned char *)new char[contents->length()+1];


This variable seems intended to store the document contents.  However both 
times it's used as a RHS of an anssignment statment:

     224     unsigned char       *ptext = text;
     [snip]
     380       position = text;
     381       start = position;
     382
     383       while (*position)
     384       {

Note that the while statement (lines 384 to 545) is likely never entered 
since gcc seems to initialize text to zeros on Linux.  The behavior could 
be platform dependent since who knows what's in that memory.

Any feedback?

Thanks

Neal Richter 
Knowledgebase Developer
RightNow Technologies, Inc.
Customer Service for Every Web Site
Office: 406-522-1485