Re: minute performance gain
| From: | Greg Beaver | Date: | Mon, 18 Oct 2004 18:46:18 +0000 |
| Subject: | Re: minute performance gain | ||
| References: | 1 2 3 4 5 | Groups: | php.pear.dev php.pear.qa |
| Request: | Send a blank email to pear-dev+get-33947@lists.php.net to get a copy of this message | ||
[cross-posting qa, please reply to pear-qa]
Alexey Borzov wrote:
Hi, Greg Beaver wrote:If all goes as I'm planning, I should be able to fix some pressing bugs and release 1.3.2 this week. There are quite a few wild cards up in the air. If there is anyone with RPM experience who can verify the fix in http://pear.php.net/bugs/bug.php?id=1163 works, that will speed things up immensely. Also peek at 1870 (rpm dependencies) Those with pearweb access, please take a look at 2396 (pear.bat renamed) and 2070 (ptmp in windows) these affect go-pear. With these bugs fixed and some others that have already been fixed (need to merge from HEAD, I will do that) we can release 1.3.2. I will also upgrade PEAR_ErrorStack to beta, after taking a close look at some design choices. GregBTW, when will a version of PEAR containing this fix (not registering _PEAR as destructor) be released?That brings me to another issue. Since the PEAR class defines a destructor it kinda sucks to extend from PEAR needlessly. Obviously if you dont call the PEAR class constructor you get around this issue.As my last email says, the PEAR destructor is never automatically registered thanks to the strcasecmp('pear') check, you have to explicitly call it through $this->_PEAR() in your own destructor, and I believe Stig designed it to be for debugging purposes only.