Quoting Scott MacVicar :
It could be related to the following patch to do with evaluation order
of per directory configuration
http://svn.php.net/viewvc?view=revision&sortby=date&revision=269640
Again it's another security issue that was resolved.
I'm not sure if this is the cause either? Th
On 4 Aug 2009, at 05:36, dan...@zoltak.com wrote:
Quoting dan...@zoltak.com:
Quoting Scott MacVicar :
[snip]
Do you know the first version it stopped working? 5.2.7 - 5.2.10 is
about 15 months worth of bug fixes. Narrowing it down to a
particular
version would help.
I will determine
Quoting dan...@zoltak.com:
Quoting Scott MacVicar :
[snip]
Do you know the first version it stopped working? 5.2.7 - 5.2.10 is
about 15 months worth of bug fixes. Narrowing it down to a particular
version would help.
I will determine the first version it stopped working with and get back
Quoting Scott MacVicar :
On 4 Aug 2009, at 04:49, dan...@zoltak.com wrote:
Quoting Scott MacVicar :
[snip]
You might be able to use sapi_getenv("DOCUMENT_ROOT",
sizeof("DOCUMENT_ROOT"))
This method dosen't appear to work.
Any chance of a clarification on "doesn't appear to work"? Does
On 4 Aug 2009, at 04:49, dan...@zoltak.com wrote:
Quoting Scott MacVicar :
1. Is it safe to lookup the DOCUMENT_ROOT in fopen_wrappers. If
not is there an alternative?
Not at the point the function is being executed on a fresh child
start
PHP hasn't always been fully started by the time
Quoting Scott MacVicar :
1. Is it safe to lookup the DOCUMENT_ROOT in fopen_wrappers. If not
is there an alternative?
Not at the point the function is being executed on a fresh child start
PHP hasn't always been fully started by the time it's doing the config
merging. None of the SAPI variabl
On 4 Aug 2009, at 03:14, dan...@zoltak.com wrote:
Quoting Jani Taskinen :
Yes. You shouldn't really do it like that. Unpack it in separate
directory. And build outside the sources:
# tar zxfv php5.2-x.tar.gz
# mkdir php_5_2
# cd php_5_2
# ../php5.2-x/configure --disable-all --enable-d
Quoting Jani Taskinen :
Yes. You shouldn't really do it like that. Unpack it in separate
directory. And build outside the sources:
# tar zxfv php5.2-x.tar.gz
# mkdir php_5_2
# cd php_5_2
# ../php5.2-x/configure --disable-all --enable-debug
# make
# make install
I couldn't get the dir
dan...@zoltak.com wrote:
> Quoting Jani Taskinen :
>> # tar zxfv php5.2-x.tar.gz
>> # mkdir php_5_2
>> # cd php_5_2
>> # ../php5.2-x/configure --disable-all --enable-debug > of relevant options here>
>> # make
>> # make install
>
> Well I have flex-2.5.4 installed. Not sure why it requires
Quoting Jani Taskinen :
# tar zxfv php5.2-x.tar.gz
# mkdir php_5_2
# cd php_5_2
# ../php5.2-x/configure --disable-all --enable-debug
# make
# make install
Well I have flex-2.5.4 installed. Not sure why it requires this
version when the release versions work with flex-2.5.3x?
I have
On 4 Aug 2009, at 00:08, dan...@zoltak.com wrote:
Quoting Jani Taskinen :
Yes. You shouldn't really do it like that. Unpack it in separate
directory. And build outside the sources:
# tar zxfv php5.2-x.tar.gz
# mkdir php_5_2
# cd php_5_2
# ../php5.2-x/configure --disable-all --enable-de
Quoting Jani Taskinen :
Yes. You shouldn't really do it like that. Unpack it in separate
directory. And build outside the sources:
# tar zxfv php5.2-x.tar.gz
# mkdir php_5_2
# cd php_5_2
# ../php5.2-x/configure --disable-all --enable-debug
# make
# make install
I've tried to compile f
On 24.07.2009, at 12:48, David Zülke wrote:
Yes, thanks, I realize that, but I need to test it in a .phpt unit
test, which is a bit trickier. But as I said, I already have an
idea. Will do it this later and open a ticket.
I finally got around to writing a test case and opening a ticket
(w
On 24.07.2009, at 12:48, David Zülke wrote:
Yes, thanks, I realize that, but I need to test it in a .phpt unit
test, which is a bit trickier. But as I said, I already have an
idea. Will do it this later and open a ticket.
Issue: http://bugs.php.net/49144
.phpt: http://pastie.org/569897 (tri
On Tue, Aug 4, 2009 at 2:47 AM, Stanislav Malyshev wrote:
>> And yes, it's worth providing separate conversion system. You might
>> not be aware of it, but there are several sets of different character
>> sets, each of which is often represented with a specific encoding
>> scheme. Shift_JIS is o
Hi!
It can be done through conversion error handlers. You can append an
encoded form of a codepoint for such unassigned characters to the
buffer within the handler.
OK, if so we may want to add implementation of this behavior to our ICU
support.
And yes, it's worth providing separate conve
On Thu, Jul 2, 2009 at 3:54 AM, Michael A. Peters wrote:
> Nick Cooper wrote:
>>
>> Does anyone have any further information on the PECL Binaries for 5.3,
>> will
>> they be released?
>
> I don't, but I suspect it is just a matter of compile + test.
We develop on Windows and deploy on Linux, so wi
On 24.07.2009, at 12:48, David Zülke wrote:
Yes, thanks, I realize that, but I need to test it in a .phpt unit
test, which is a bit trickier. But as I said, I already have an
idea. Will do it this later and open a ticket.
Issue: http://bugs.php.net/49144
.phpt: http://pastie.org/569897 (als
Uwe Schindler wrote:
> Thanks Jani,
>
> I thought it is still optional to do this.
>
> The question is, how to do that on Windows with TortoiseSVN using the GUI?
> We had a similar discussion on the Lucene-Java Mailinglist about updating
> the backwards-compatibility tests branch with one commit
Le Fri, 31 Jul 2009 11:35:45 +0200, Hannes Magnusson a écrit :
> I just realized we don't link to that function from the simplexml docs.
>
> I'll add some "see also" links and add an "errors" section mentioning it
> too.
>
> -Hannes
Thank you for the documentation :)
--
PHP Internals - PHP Ru
On 08/03/2009 03:19 PM, dan...@zoltak.com wrote:
Quoting Jani Taskinen :
Well, this is using PHP 5.2.10. And you were supposed to test the
snapshot. How about the backtrace using latest PHP_5_2 checkout?
I downloaded the latest snap and tar.bz2 it in the php-5.2.10 folder so
I could compiled
Quoting Jani Taskinen :
Well, this is using PHP 5.2.10. And you were supposed to test the
snapshot. How about the backtrace using latest PHP_5_2 checkout?
I downloaded the latest snap and tar.bz2 it in the php-5.2.10 folder
so I could compiled it using the php-5.2.10 ebuild file in gentoo.
On 08/03/2009 02:52 PM, dan...@zoltak.com wrote:
Quoting Jani Taskinen :
On 08/03/2009 07:52 AM, dan...@zoltak.com wrote:
Jani Taskinen wrote:
[snip]
Obviously you haven't compiled using --enable-debug OR most likely are
not using the compiled libphp5.so. Check your phpinfo() that you
actua
Quoting Jani Taskinen :
On 08/03/2009 07:52 AM, dan...@zoltak.com wrote:
Jani Taskinen wrote:
[snip]
Obviously you haven't compiled using --enable-debug OR most likely are
not using the compiled libphp5.so. Check your phpinfo() that you
actually use the one you compiled first..
--Jani
Her
PHP 6 Bug Database summary - http://bugs.php.net/
Num Status Summary (86 total -- which includes 40 feature requests)
===[*Unicode Issues]==
48265 Open Source and result of database have different encodings.
==
PHP 5 Bug Database summary - http://bugs.php.net/
Num Status Summary (1460 total -- which includes 902 feature requests)
===[*General Issues]==
48597 Open Unclosed array keys break space escaping in $_GET/POST/REQUEST
48778 Assign
On Sat, Aug 1, 2009 at 9:02 AM, Stanislav Malyshev wrote:
> Hi!
>
>> They calculate the total width of a string based on "east asian width"
>> property, which is still valid to give a rough measurement of the
>> rendered string.
>
> OK, I guess if it's some kind of special calculation that doesn't
On 08/03/2009 07:52 AM, dan...@zoltak.com wrote:
Jani Taskinen wrote:
[snip]
You obviously do not have the correct sources then.
Try get them directly using SVN:
# svn co http://svn.php.net/repository/php/php-src/branches/PHP_5_2/
And if those don't work either, you need to provide a GDB back
28 matches
Mail list logo
