Re: HTML::LinkEx(trac)tor

Fred Douglis (douglis@research.att.com)
Fri, 12 Jul 1996 18:16:32 -0400


My apologies - my last note was apparently due to a red herring. There's still 
a bug, or at least something that might be addressed, but most of my 
description was way off.

Yes, I had just converted, and perhaps the new library is tripping over 
something the old one didn't, but the idea that it wasn't parsing within the 
anchor was my confusion based on the other mail and the results I was seeing.

In fact, after I sent the mail I kept at it (far too long :) because I hate 
reporting bugs without also reporting the fix, if it's at all possible to 
figure it out myself.  And after some debugging I determined that the ... in 
my example was actually the kicker: the text I was parsing had nonconforming 

	 ALT=[ ]

in it!  

Now, it's true that despite this not conforming to the HTML spec, it would be 
better for the parser to handle it than to decide the whole thing isn't a tag 
after all and then treat it as regular text (then converting it to < form 
when outputting it later).

However, I'm not going to try and figure out how to munge the regular 
expression to handle this, and hope someone else can figure that part out 
(somewhere in the vicinity of Parser.pm line 232...)

Fred Douglis 		    MIME accepted	  douglis@research.att.com
AT&T Research					  908 582-3633 (office)
600 Mountain Ave., Rm. 2B-105 			  908 582-3063 (fax)
Murray Hill, NJ 07974     http://www.research.att.com/orgs/ssr/people/douglis/