Quickly checked out the patch and first thing to catch my eye
was the use of '//' comments. That's a big no-no. (this is not C++)
Please read the CODING_STANDARDS couple of times. :)
(Use of these C++ comments causes compile failure with some compilers)
--Jani
I guess the way the implementation works is quite fine. But i guess leaving
the calls in a random order gives the use of the destructors a quite bad
taste. The current implementation leaves many room for 'try and error'
writing of code as you are almost unable to make things work the way
you would
Hello,
I've talked to a few people on the team about this issue, on and off.
It seems that every time I have tried compiling php since about 4.0.7, the
resulting cli binary does not behave properly when used with -a. This problem
has persisted with every version of php since then. I have only tri
Er, either the php-internals MARC archive doesn't show all attachments, or
my diff file got stripped en route for not having a .txt extension. I've
attached it again, just to make sure.
Paul
--
-Original Message-
From: Paul Hudson [mailto:[EMAIL PROTECTED]
Sent: 10 February 2004 00:30
Hi there,
Bug #24064 (submitted by [EMAIL PROTECTED]) requests a standard deviation
function for PHP. I realise that any of you could implement this in 10
minutes, but according to the bug database it is still Open so I figured I
would give it a try myself!
There are probably a dozen errors in t
At 03:55 PM 2/9/2004 +, Stephane Drouard wrote:
== Quote from Andi Gutmans ([EMAIL PROTECTED])'s article
> He doesn't necessarily own a reference but tries to access it in the
> destructor.
IMO, this is bad programming. If an object wants to access another object
(global or not), it has to own
Hi Greg,
Do you have a short script to reproduce it or at least the last working
line in phpdocumentor?
I'm working a bit in ZE2 these days for a project, should be a good
exercise, aka a 2nd bugfixe :)
hth
pierre
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit:
== Quote from Eric Daspet ([EMAIL PROTECTED])'s article
> Be carreful, this does not resolve all things :
>
> $source1 = new stdClass ;
> $source2 = new stdClass ;
> $source3 = new stdClass ;
> $refLevel1 = new stdClass ;
> $refLevel2 = new stdClass ;
>
> $source1->ref = $refLevel1 ;
> $source2->re
>
> revision 1.76
> date: 2002/08/17 20:32:26; author: zeev; state: Exp; lines: +2 -2
> Use mtime instead of atime, as we always update the session file anyway.
>
>
> --Jani
Sorry... I hadn't see that. After submitting the patch I'
On Mon, 2004-02-09 at 18:28, Andrey Hristov wrote:
> Hello internals,
> what's your opinion on whether the constructors must follow one of the
> principles of the polymorphism - "The visibility in extending classes must be
> the same or better".
[...]
> Currently this code emits a compilation error
On Mon, 9 Feb 2004, Nuno Lopes wrote:
>Hi,
>
>As I don't have sufficient karma, here is a patch to fix bug #23432 (PHP
>using mtime instead of atime to clean session files).
revision 1.76
date: 2002/08/17 20:32:26; author: zeev; state: Exp; lines: +2 -2
Use mtime i
On Mon, 9 Feb 2004, Pierre-Alain Joye wrote:
>On Mon, 9 Feb 2004 15:14:46 +0100 (CET)
>Derick Rethans <[EMAIL PROTECTED]> wrote:
>
>> > So this is bug is bogus, and this is the expected behavior. Unset
>> > (or set the var null) frees the memory only if memory_limit is
>> > enabled and not equal t
Hello internals,
what's your opinion on whether the constructors must follow one of the
principles of the polymorphism - "The visibility in extending classes must be
the same or better".
Example :
class parent {
public function __construct() {
}
}
class child extends parent {
private function __
Stephane Drouard wrote:
Couldn't PHP destroy global objects by taking their refcount into
account? All objects with refcount=1 are first destroyed, then all
objects with refcount=2, and so on... Objects with a same refcount
are destroyed in a random "order".
Be carreful, this does not resolve all
On Mon, 2004-02-09 at 17:09, Nuno Lopes wrote:
> Hi,
>
> As I don't have sufficient karma, here is a patch to fix bug #23432 (PHP
> using mtime instead of atime to clean session files).
Might be more correct, but what about filesystems mounted with
"noatime"?
- Timm
--
PHP Internals - PHP Runt
== Quote from Stephane Drouard ([EMAIL PROTECTED])'s article
> == Quote from Andi Gutmans ([EMAIL PROTECTED])'s article
> > He doesn't necessarily own a reference but tries to access it in the
> > destructor.
>
> IMO, this is bad programming. If an object wants to access another object (global or
I decided to delete everything and do a new cvs checkout, and now the
segfault has disappeared. Not sure what was going on.
Thanks,
Greg
Greg Beaver wrote:
Hi,
When I try to run phpDocumentor with HEAD in win32/apache 1.3 checked
out a few minutes ago, I get a segfault, here's a stack trace g
Hi,
As I don't have sufficient karma, here is a patch to fix bug #23432 (PHP
using mtime instead of atime to clean session files).
Nuno
Index: mod_files.c
===
RCS file: /repository/php-src/ext/session/mod_files.c,v
retrieving revisi
== Quote from Andi Gutmans ([EMAIL PROTECTED])'s article
> He doesn't necessarily own a reference but tries to access it in the
> destructor.
IMO, this is bad programming. If an object wants to access another object (global or
not), it has to own a reference on it, to guaranty the referenced obje
Hi,
When I try to run phpDocumentor with HEAD in win32/apache 1.3 checked
out a few minutes ago, I get a segfault, here's a stack trace grabbed
from VS:
php5ts_debug.dll!execute(_zend_op_array * op_array=0x0109f020, void * *
* tsrm_ls=0x010925d8) Line 1270 + 0x15 C
php5ts_debug.dll!execute_ne
On Mon, 9 Feb 2004 15:14:46 +0100 (CET)
Derick Rethans <[EMAIL PROTECTED]> wrote:
> > So this is bug is bogus, and this is the expected behavior. Unset
> > (or set the var null) frees the memory only if memory_limit is
> > enabled and not equal to -1 (unlimited) for both php5 and php4.
>
> memory
Hi internals,
I have created a small patch for debug_print_backtrace() that adds the
possibility the string representation of the backtrace in printed form to be
returned as string (as it is with print_r() currently).
The patch is available at :
http://andrey.hristov.com/projects/php_stuff/patches/
On Mon, 9 Feb 2004, Pierre-Alain Joye wrote:
> Hello,
>
> While reading the bugs reports I came up with something I did not
> expect with the bug 27714. Maybe I have had too much ram in the past
> years ;)
>
> So this is bug is bogus, and this is the expected behavior. Unset
> (or set the var null
Hello,
While reading the bugs reports I came up with something I did not
expect with the bug 27714. Maybe I have had too much ram in the past
years ;)
So this is bug is bogus, and this is the expected behavior. Unset
(or set the var null) frees the memory only if memory_limit is enabled
and not e
At 01:36 PM 2/9/2004 +, Stephane Drouard wrote:
== Quote from Zeev Suraski ([EMAIL PROTECTED])'s article
> They're not locked to 1, and nothing in a symbol table will ever be with a
> refcount of less than 1... But generally, all global variables (or for
> that matter, all variables period) ha
== Quote from Zeev Suraski ([EMAIL PROTECTED])'s article
> They're not locked to 1, and nothing in a symbol table will ever be with a
> refcount of less than 1... But generally, all global variables (or for
> that matter, all variables period) have a refcount of 1, unless you do
> something 'speci
At 12:38 09/02/2004, Stephane Drouard wrote:
== Quote from Zeev Suraski ([EMAIL PROTECTED])'s article
> If you're talking about destruction that honors reference counts (which has
> nothing to do with order, it's still randomly-ordered), then yes, it's
> *generally* ok.
Yes, I was talking about tha
== Quote from Zeev Suraski ([EMAIL PROTECTED])'s article
> If you're talking about destruction that honors reference counts (which has
> nothing to do with order, it's still randomly-ordered), then yes, it's
> *generally* ok.
Yes, I was talking about that point.
> But that what we had before, an
At 21:33 08/02/2004, Derick Rethans wrote:
On Thu, 5 Feb 2004, Andi Gutmans wrote:
> Hi Ananth,
>
> I'm sorry but your patches are kind of messy. There are far too many #ifdef
> NETWARE's in there. It should be possible to improve on this by creating
> abstraction for things such as stat.
> I don'
At 12:03 07/02/2004, Stephane Drouard wrote:
== Quote from Zeev Suraski ([EMAIL PROTECTED])'s article
> To make a long story short, no. Circular references are hardly the
> problem, even a one-sided reference can be problematic. Just one example
> (there are many others) - you have $container and
> Adding it to the distribution is +1 from me, enabling it by default is a
> -1 personally as I would turn others on by default before this one.
I agree 100% with Derick
Georg
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
31 matches
Mail list logo
