PEAR break some quality asurance?
| From: | Pablo Roman | Date: | Tue, 24 Aug 2004 17:45:48 +0000 |
| Subject: | PEAR break some quality asurance? | ||
| References: | 1 2 3 | Groups: | php.pear.general |
| Request: | Send a blank email to pear-general+get-14259@lists.php.net to get a copy of this message | ||
following the thread:
"[PEAR] can I depend on stable PEAR classes to not change?"
then,
PEAR break some quality asurance?
Greg Beaver wrote:
Chuck Hagenbuch wrote:-- Atte Pablo E. Román A. ----------------- F: 6784248 Jefe de Desarrollo Area de Desarrollo de Info Tecnologías Escuela de Ingeniería y Ciencias Universidad de Chile ----------------- "Find the set of all sets that doesn't contain itself." B. RussellQuoting Miles Keaton <mileskeaton@gmail.com>:Chuck is right, history has not been good. Some differences from history: 1) there is a standard defining how BC-breaking releases should be handled, and it is much more strict than the past "do whatever" situation. 2) there is a QA group that has the power and the desire to enforce BC - any package that releases a BC-breaking release will have its release pulled within a few hours. I still wouldn't put absolute trust in any changing code - always have a backup plan in place as I described in my email. GregIf I write some apps based on PEAR features, but my system is constantly updating my /usr/local/share/pear directory, is it safe to assume that my code can depend on these PEAR classes for a long time?Judging by history, no.