[Chicago-talk] HTML::Tree progress

Leland Johnson easyasy2k at gmail.com
Sun Sep 19 00:49:18 CDT 2004


Yes, that is the major question we had during our meeting.

shawn.c.carroll added t/tree.t in r40, which failed tests 7 and 13 out
of the box (they are the exact same test, just one is read from a file
and the other is read from a string). The _test_ case says that it
should not find the word "comment" because all instances are in a
comment, or _recursive_ comment. This test fails because currently the
_module_ does not deal with recursive comments, so they are converted
to escaped HTML. (This paragraph is all at revision 40!)

We need the author to decide on what the final solution to this should
be, then change the tests appropriately. Jon has made an additional
test for simple comments in t/tree.t. That test is run twice for
new_from_content and new_from_file. Now, t/tree.t fails the exact same
tests as before, but they are now tests #7 and #14.

So, as of now, someone needs to fake a final call on how nested
comments should be handled.


On a more general note, the test coverage is up (and mercilessly
refactored) since our meeting today. You can see the two pages here on
my website:
Before:
http://s89790145.onlinehome.us/random/phalanx_chicago/cover_dbr42/coverage.html
and now....
http://s89790145.onlinehome.us/random/phalanx_chicago/cover_dbr59/coverage.html

So the net coverage increased by... 3%!

Now, that not too impressive, but a lot of the tests were refactored,
a test case found an (architecture) bug, and a module that previously
had 0% test coverage has 100%! Also, a lot was learned about
subversion and the annoyance of a commit setting the trunk on fire.
This was done a few times during our session (oops) and it is still
burning, so somebody make the contacts needed to put it out!

I need to go to bed right now, but I'll have detailed results on all
that happened at the meeting up on the wiki in the afternoon tomorrow,
once I get a plug for my laptop again, and go over the commit logs.


On Sat, 18 Sep 2004 21:32:06 -0500, Jon Amundsen
<jamundsen at jamundsen.dyndns.org> wrote:
> 
> 
> On Sat, Sep 18, 2004 at 08:57:52PM -0500, Andy Lester wrote:
> >
> > So what happened with HTML::Tree today?  I've seen a lot of updates, but
> > t/tree.t fails.
> >
> > xoxo,
> > Andy
> >
> > --
> > Andy Lester => andy at petdance.com => www.petdance.com => AIM:petdance
> > _______________________________________________
> > Chicago-talk mailing list
> > Chicago-talk at mail.pm.org
> > http://mail.pm.org/mailman/listinfo/chicago-talk
> 
> We think it may be an actual bug in HTML::Parse or somewhere.  It doesn't seem to handle nested html comments like:
> 
> <!-- outside comment
> <!-- inside comment
> -->
> -->
> 
> I refactored the html test data this evening and split the 'comment' test into 2 tests.  One for simple comments <!-- simple --> and one for nested.  No one has tackled the actual code or standard yet.  It may be that nested comments aren't valid html, but may be supported in the spirit of 'be lenient in what you receive'?
> 
> Leland volunteered to update the kwiki... I suspect you'll see more info on our days activities by tomorrow.  I'll probably send something more to the list also...
> 
> --
> 
> Jon Amundsen
> jamundsen at jamundsen.dyndns.org
> 
> A "No" uttered from deepest conviction is better and greater than a
> "Yes" merely uttered to please, or what is worse, to avoid trouble.
>                 -- Mahatma Gandhi
> 
> 
> 
> _______________________________________________
> Chicago-talk mailing list
> Chicago-talk at mail.pm.org
> http://mail.pm.org/mailman/listinfo/chicago-talk
> 
> 
> 
> 



-- 
Leland Johnson
http://protoplasmic.org


More information about the Chicago-talk mailing list