On Thu, 24 Nov 2005 23:45:45 -0800 [EMAIL PROTECTED] (Rasmus Lerdorf) wrote:
> Helgi Þormar wrote: > > Rasmus mentioned that no PEAR person tested the final RC and all > > that and thus this issue wasn't found ... Correct me if I'm wrong, > > but wasn't that change done between the final RC and the official > > release ? > > Nope, this change was in RC6: > > http://cvs.php.net/diff.php/php-src/ext/date/php_date.c?r1=1.43.2.20&r2=1.43.2.21&ty=u > > I missed it too, but then I don't use pear/Date anywhere. It is not only about date, such changes in 5.1 are the worst idea we can have, and doing so late in the release process is even worst. Wez and you told me that I do propose solutions, for the 5.1 release, I see 2 possibilities: 1) Drop this 5.1.0 release and repack everything without the class date 2) Release 5.1.0pl1 as soon as possible to fix the problem 1) is, as far as I know, not possible, so we have to go with 2). Marcus, arguing such way is missing the minimum informations one may need to comment this problem. Can you seriously think that adding such class in a minor release is the way to go? What is obvious is that there is not enough people having a real influence on what should go in or what should not go. As a reminder, the last releases with such non obvious breaks was our last two... I let you wonder who wrote the fix. About ext/date, it was a bad idea from day #1 to do it this way. I proposed to add a new date/time APIs in php5 before it was even available, a 1st proposal was done and the 1st codebase was commited in pecl (where all new codes should first go for tests and approbations). I asked for feedbacks and for people willing to help to define a clean api and how it should work (both for OO and procedural). Then Derick asked me to stop to work on this extension until he is ready with the rewrite of strtotime with a new TZ, it was his only idea, for what he said. We first agreed first to do nothing in 5.1, it was too late as 5.1 was already in a bug fixes only period, who cares? he commited and replace strtotime... Then comes the replacement of date() and some other functions, and now the addition of a date """object""". Can we really say it is a mistake? a lack of attention from QA people? from users? from PEAR? I do not think so. --Pierre -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
