On 21.05.2007 11:55, Lester Caine wrote:
Well, there's a reason - not every developer watches bugs closely.
Thanks to great work by Antony, there's somebody who does that, but we
can't realistically expect every maintainer to do that, so assigning
bugs is the means to alert the developer that this bug is (at least
potentially) in his domain.
Assigning blindly is not the right procedure. There needs to be a proper
REFERRAL process so that other people who are working in the area are made
aware of the problem. If the bug is flagged with the correct area or package
name, then anybody can pick up the baton. Simply 'assigning' a bug to say Wez
when AS HE SAYS - there are other developers - is a pointless waste of time.
Blindly?
I'm afraid you don't understand what you're talking about.
There is EXTENSIONS file (in the sources root dir) which lists extensions along
with their maintainers.
Do you see any maintainers for PDO? No? Me neither.
Then there are any maintainers for com_dotnet except Wez? Nope..
Or maybe you know any streams maintainers except Wez and Sara? Tell me.
If the file is wrong and there are other maintainers for these extensions,
then I've been totally blind for years, since I follow the CVS list pretty close
and I can distinguish between a person committing occasional fixes and maintainers
looking for the extension on a permanent basis.
I think assignment is request to look - and if possible fix, if not -
one should indicate he can't accept it. Of course, as we know, in theory
practice follows theory, in practice it doesn't - so there are cases
where bugs are assigned but not really looked at.
HOW do you 'request' a group of developers look at a bug?
Personally I scan the bug summary each week to see what has popped up on my
areas of interest ( The 'feature requests' should be killed from that !!! )
and that would seem to be the most productive method possible. Antony and
others just needs to make sure that bugs are correctly classified and
duplicates flagged and removed. There are assigned bugs there that have not
moved in months and it would be nice to see the total number of BUGS going down.
Want to help me with that (or maintainers with bugs in their extensions)?
I'd gladly accept both, even though bugfixes are much more welcome =)
--
Wbr,
Antony Dovgal
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
