Re: No comment about serious bug in error handling
| From: | Greg Beaver | Date: | Mon, 29 Mar 2004 21:45:39 +0000 |
| Subject: | Re: No comment about serious bug in error handling | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-26863@lists.php.net to get a copy of this message | ||
Hi Markus,
Your points are very valid. PEAR_Error is flawed in its design. This is why I wrote PEAR_ErrorStack, which will be in the next version of PEAR. The issues you raise are among a huge number of flaws. Fixing them forces hacks, and so this requires more than a casual consideration. Also, I know for a fact that right now Pierre-Alain is completely swamped with school, I'm out of town in an intense residency from my day job, Tomas has fallen off of the face of the earth and Stig is basically inactive. Don't take it personally, I return next Friday :).
Greg
Markus Kalkbrenner wrote:
Hi! Three weeks ago, I posted a bug regarding PEAR's error handling: http://pear.php.net/bugs/bug.php?id=949 From my point of view it's a serious one which already caused a lot of trouble here. If you have an existing object that inherits PEAR_Error (like an instance of DB_Error) it's impossible to raise it again using PEAR::raiseError(), if the constructor of the object doesn't support messages. I feel very disapointed about the support from the PEAR developers. I didn't just post the bug but also tried to find the reason of it and provided a patch in order to enable you to easlily get an idea of the problem. But I didn't get a single comment regarding this issue for three weeks now! Please don't get me wrong. I guess that most people don't work full time on PEAR but spending their free time on it. But I'm of the opinion that a simple message like "you're right, I'll have a look at it" or "you're wrong, it's forbidden to raise an error this way" must be possible. Regards, Markus Kalkbrenner