Re: cvs: php-src(PHP_5_2) / NEWS /main rfc1867.c rfc1867.h

From: Date: Sun, 23 Jul 2006 06:50:40 +0000
Subject: Re: cvs: php-src(PHP_5_2) / NEWS /main rfc1867.c rfc1867.h
References: 1 2 3 4 5  Groups: php.cvs php.internals 
Request: Send a blank email to php-cvs+get-39461@lists.php.net to get a copy of this message
On 23.07.2006 02:31, Stefan Esser wrote:
Hello,
well the policy is you commit to HEAD first so that HEAD and base do not get out of synch. For PDO we already have the situation and it will mos
Yeah we all know how well commiting to HEAD and then merging the same untested code to the branches work. We have seen this commit policy in action with the latest PHP 5.1.3 releases.
So this is the reason you think you can break the upcoming release now? Please revert it, as I don't want to see yet another broken release just because you decided you can do what you want.
At the moment there is no point in commiting stuff to HEAD first and then merging it to branch. HEAD is completely different and needs completely differnt code. Chances are high that the fileupload code doesn't work at all in PHP6 after the unicode changes. I rather like to test new code against stable code branches instead of testing it against unstable branches that could blow up in my face.
"Test new code against stable branche" ? NO, thank you! Please test your code somewhere else, STABLE branches should remain STABLE, they are not for testing.
The time you waste on questioning my development decisions you better spend on fixing your own bugs, that I had to find for you in the past.
Oh my, what a nice reply.
(f.e. checking return values of calls to base64_decode() in 'data' URL shemes...)
Please do what everybody else does: 1) follow the policy and keep HEAD and 5.2 synchronized. 2) avoid committing experimental code to stable branches "just for testing" - this is what UNstable branches are for. -- Wbr, Antony Dovgal

« previous php.cvs (#39461) next »