RE: [PHP-DEV] 4.0.6

At 11:36 PM 4/29/2001 +0100, James Moore wrote:

<snip> K I have a list of bugs that need to be at least reviewed by the appropraite developers, this list needs to be added to/altered etc can you please send feedback on which issues should be fixed before 4.0.6, there are some there that will not be some that are a 2 second fix etc... Could the QA TEam also look at them and where possible provide scripts that reproduce the problem and/or just add an and me note. =================== List of bugs ======================================= List of iteresting bugs so far: ===============================

I hope you keep good notes :)

Zend Related ------------ 6491 (Incorrect setting of PHP_SELF under certain circumstances)

Someone who knows this better should state his opinion. Maybe Sascha who knows all the RFC's by heart :) And I guess it depends what the PHP manual says too.

8130 (Shallow Copy problems)

For now this is going to have to stay a "feature".

8414 (set_timout_limit problem

     (very weird not the normal set_timeout_limit bogus report)

8889 (Memory consumption.. decent discussion included)

I mailed Brian and am waiting for a response from him. I'd like to work with him together on this one.

9289 $argv/$argc weirdness (unverifed) 9462 Include/Require need to be binary safe (see report for example)

I don't understand why this is a bug. He should code better :) This is how the OS works or am I missing something?

9505 (Patch included OS400 specific)

No idea why this was a bug but I changed the code to how he wanted because it doesn't make a difference to us. Commited to CVS.

10299 Same as 8889. To be verified in Zend ---------------------- 10029 Not sure about this one

      but its here due to my lack of understanding of Zend :)

It's probably not a problem but I mailed Zeev to look at it.

Build Related ------------- 8045 Configuration order of ccvs and mcrypt CGI Related ----------- 9041 #! at top of script problem. (this one really needs fixing!)

I see where the code needs to be changed. The question is if we should really skip this in CGI or not. I think we shouldn't. We don't want to and won't do it for Apache, ISAPI and so on...

Enviroment Related ------------------ 8725 (putenv problems (see report)) Can anyone verify this? ini_* funcs ----------- 10431 ini_alter eats the include_path (unverified) Interbase Related ----------------- 10458 Bugs #9257 and #10292 located and fixed - see diff

      (can someone check the fix please)

Sockets Related --------------- 9427 (PHP blocks waiting for packets (needs to be verified)) Time Related ------------ 9640 strtotime behaving weirdly (derick did you get to the bottom of this)? 9878 gmmktime doesnt work with daylight saving

     (can anyone verify this?) (test script included)

URL Related ----------- 1249 url_parse() is a bit too strict To be verified -------------- 9526 Can anyone verify this? (safmode copy problems) 9780 Seems like the dirname etc confusion due to standards

The standard we took is dirname on Linux. From a Linux shell do dirname /tmp/ Andi