Re: CS fixes
| From: | Klaus Guenther | Date: | Wed, 25 Feb 2004 13:23:51 +0000 |
| Subject: | Re: CS fixes | ||
| References: | 1 2 3 | Groups: | php.pear.qa |
| Request: | Send a blank email to pear-qa+get-656@lists.php.net to get a copy of this message | ||
Lukas Smith wrote:
> Tobias Schlitt wrote:
>
> > Hi all!
> >
> >
> >>I will maintain a current list of maintainers that have agreed to this
> >>and will make it available online.
> >
> >
> > That are the issues I wanted to avoid by getting up a QA group which
sets up
> > an official annoncement of QA's tasks and powers. IMHO the list solution
is
> > not very comfortable for QA. It's a better way to tell people "That's
the
> > way to go, if someone has objections scream now" instead of "Scream if
you
> > accept the way to go"...
>
> I dont think it is feasible to have QA commit whitespace fixes if the
> maintainers dont want this to happen. The internals people are having
> frequent flame wars about this topic and I want to evade it by making
> this an optional service instead of a PITA.
I agree that this should be optional, but I also agree with Toby that the
list should be maintained by QA and the rules as to what is permissible,
etc. should also be handled by QA. What if the maintainer is working on the
code and someone submits CS changes? It's generally a lot nicer to provide a
patch to the maintainer. However, if the code hasn't been touched for a
number of weeks, then I can understand committing directly.
By the way, why not include this information in the pear package database?
How about automatically asking the question the next time a release is
uploaded? Once that has been done, the list can be automatically generated.
If a maintainer later decides to opt out, he could do that on a special
"Package properties" panel that could be accessed directly from the package
page -- just like the form for adding and removing developers.
I think it's getting time to have a QA section of the website, with
instructions for testing, the list proposed in this thread, etc.
Klaus