Re: Patch for "HTML5" support

Bjoern Hoehrmann <[email protected]> Sun, 15 Jan 2012 23:37:10 +0100
Newsgroups gmane.comp.web.html-tidy.devel
Message-ID <[email protected]>
* Andy Lester wrote:
>If this code was posted on GitHub, the process would involve:
>
>* Bjoern makes a fork of the main tidy repository
>* Bjoern commits his changes to his fork
>* Bjoern announces that his changes have been made, and gives the location=
 of his fork
>* Anyone wanting to test out the changes can either fork Bjoern's fork
>and see the changes directly, or else merge Bjoern's fork with their
>own repository
>
>All of this dealing with patch formats and diffs and whatnot disappears.  =
His changes Just Work.

This happend two months ago already https://github.com/w3c/tidy-html5
and there has been only one relevant bug report (meta in noscript in
head is moved out of noscript because HTML4 et al. did not allows this
kind of nesting but "HTML5" does; no patch attached). I am unlikely to
fix such problems, either there are only very few such issues, then we
simply apply the patch, or there are very many, then we would need a
volunteer to investigate and fix them, or alternatively let Tidy rest
in peace. I could also have made a CVS branch for this, but people are
more likely, in my opinion, to find the patch in HEAD than finding a
different branch; we could also move things over to GitHub, I have two
dozen repositories there, but if there is no point if there is no in-
terest in actually working on the code there in some coordinated way.

I want to know whether the patch is good or not, and the highly publi-
cised GitHub fork above has not produced anything resembling an answer
to that. You are quite probably right with your description above, but
right now I am a maintainer trying to decide whether to apply a patch
that's been contributed, whether it's a good patch, and I do not have
a good way to find out.

(I will also note that there are downsides with this kind of forking,
the fork above for instance now contains changes that are not in line
with how we have developed Tidy so far, like not reporting accessibi-
lity problems because they are not conformance errors in "HTML5".)
-- =

Bj=F6rn H=F6hrmann =B7 mailto:[email protected] =B7 http://bjoern.hoehrma=
nn.de
Am Badedeich 7 =B7 Telefon: +49(0)160/4415681 =B7 http://www.bjoernsworld.de
25899 Dageb=FCll =B7 PGP Pub. KeyID: 0xA4357E78 =B7 http://www.websitedev.d=
e/ =


---------------------------------------------------------------------------=
---
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2