Re: checklink 4.4 release? (Was: [Bug 5648] Handle gzip content-encoded content)

On Thursday 24 April 2008, Olivier Thereaux wrote:
> On Mon, Apr 21, 2008, bugzilla@farnsworth.w3.org wrote:
> > ------- Comment #1 from ville.skytta@iki.fi  2008-04-21 07:13 -------
> > Done in CVS, will be in 4.4.
> > http://dev.w3.org/cvsweb/perl/modules/W3C/LinkChecker/bin/checklink.diff?
> >r1=4.98&r2=4.99
>
> Ville,
>
> Thanks a lot for this fast patch. Do you have an idea of the things left
> to clear and clean before we can push a 4.4 release? I haven't had a lot
> of bandwidth for checklink in the past few weeks but I don't recall
> there was much in the way of a release.

I just had a look, and seems that the thing needing most work is HTML output 
in command line mode.  I'm working on a patch, it should be ready pretty 
soon.

The other thing I found is that the "Checking link" javascript callbacks 
printed from perl don't get the URL to display in HTML escaped form (in 
check_uri()).  HTML escaping them is trivial but doing so makes my Firefox 
and Konqueror display &'s as &amp's in the status widget.  I'm a bit slow 
today and maybe I'm missing something obvious there, but doesn't that sound 
like a browser bug?

> It may also be worth looking again at where the project is heading.
> Checklink is now a venerable service, working well, and it seems from
> the feedback I receive now and then that the only issue is that its good
> behavior (following robots.txt and waiting a second between requests) is
> annoying in an age of faster web servers. It may be worth looking at
> whether to trigger the robotUA only in recursive mode, or whether we
> want to plug in a faster ajax interface (not for the recursive mode),
> etc.

Yep.  And for significant new developments, also check whether Perl is still 
the desired implementation language for it in the first place...

Received on Thursday, 1 May 2008 12:01:19 UTC