From: Daniel Veillard (Daniel.Veillard@w3.org)
Date: Wed Oct 13 1999 - 02:38:50 EDT
> Huh? I though just moving the test down below the four cases would
> solve the problem; you couldn't have autoclosed anything before
> parsing some content anyway, so moving the test wont disturb the first
> iteration of the loop, and after the first iteration the loop
> basically looks the same as before... What am I missing?
The fact that the parser tries to accomodate with invalid
input. Your change was breaking the test3.html test and the whole
problem started to get really messy. I actually rewrote most
of the HTML parsing logic yesterday, and commited it to the W3C
CVS base, there is still a bug left (at least) but now at least
the parsing engine logic seems quite more robust.
> > I certainly need to add more test cases, as the code progresses
> > catching those should be done as much as possible in a systematic
> > way. Any idea ?
>
> Erh... not really. I usually try to write small examples to test the
> individual functions as I go along, and I keep those around for
> regression testing. But even so I dont really hope to cover all
> combinations, as different parts of my program starts to interact,
> cf. the bug that started this thread. And anyway this is open
> source - leave the debugging to the users :-)
Well, I need regression tests anyway to check before commits.
The fact that it's open source doesn't prevent people from complaining
when something gets broken ;-)
Daniel
-- Daniel.Veillard@w3.org | W3C, INRIA Rhone-Alpes | Today's Bookmarks : Tel : +33 476 615 257 | 655, avenue de l'Europe | Linux, WWW, rpmfind, Fax : +33 476 615 207 | 38330 Montbonnot FRANCE | rpm2html, XML, http://www.w3.org/People/W3Cpeople.html#Veillard | badminton, and Kaffe. ---- Message from the list xml@rufus.w3.org Archived at : http://rufus.w3.org/veillard/XML/messages to unsubscribe: echo "unsubscribe xml" | mail majordomo@rufus.w3.org
This archive was generated by hypermail 2b29 : Wed Aug 02 2000 - 12:29:49 EDT