On 11/29/2006 11:52 AM, Rob Richards wrote:
rrichards Wed Nov 29 08:52:09 2006 UTC
Modified files:
/php-src/ext/simplexml simplexml.c
/php-src/ext/simplexml/tests bug39662.phpt
Log:
MFB: correct fix for bug #39662 to fix leak and tree corruption
fix
Hello Rob,
actually the comment was wrong/premature. This part of todays changes
only allows the URI to be specified. The prefix stuff comes later. And
well the prefix stuff is only a convenient stuff if you know what you do.
And yes you are right only the URI is unique but not all people know t
I think using namespace URI will also help us out with default
namespaces because they don't have a ns prefix.
-adam
On Fri, 14 Apr 2006, Rob Richards wrote:
> Hi Marcus,
>
> I think it would be better to speficy namespace URI rather than prefix.
> Prefix can be anything, so unless you know what
Hi Marcus,
I think it would be better to speficy namespace URI rather than prefix.
Prefix can be anything, so unless you know what was used, you need to
use the new namespace methods (which are primarily debug functions). It
is more common to known the URI than prefix.
I havent tried the cha
nal Message-
> From: Jani Taskinen [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, December 07, 2005 8:23 PM
> To: Dmitry Stogov
> Cc: php-cvs@lists.php.net
> Subject: Re: [PHP-CVS] cvs: php-src /ext/simplexml
> simplexml.c /ext/simplexml/tests000.phpt 001.phpt 002.phpt
> 003
All these macros around and you still duplicate the stuff?
Instead of doing:
ZVAL_U_STRING(ZEND_U_CONVERTER(UG(runtime_encoding_conv))...
Why not use the ZVAL_RT_STRING macro? (What is RT supposed to be anyway?)
--Jani
On Wed, 7 Dec 2005, Dmitry Stogov wrote:
dmitry
Hi Marcus,
You broke tests in unicode mode.
Dmitry.
> -Original Message-
> From: Marcus Boerger [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, November 29, 2005 5:46 AM
> To: php-cvs@lists.php.net
> Subject: [PHP-CVS] cvs: php-src /ext/simplexml simplexml.c
>
>
> helly Mon Nov 28
helly Sun Nov 20 08:03:19 2005 EDT
Modified files:
/php-src/ext/simplexml simplexml.c
Log:
- Extend SimpleXmlElement::__construct() with two parameters
. int options: pass options to 2.6.0+ parsing functions
. bool data_is_url: whether first parameter is a url rather then xml
data
Hello Rob,
Tuesday, November 1, 2005, 12:59:02 PM, you wrote:
> Marcus Boerger wrote:
>>>When doing a recursive get, if the namespace is redefined further down
>>>in the subtree, the redefinition is what ends up in the array rather
>>>than the closest namespace to the element. Just to be clear
Marcus Boerger wrote:
When doing a recursive get, if the namespace is redefined further down
in the subtree, the redefinition is what ends up in the array rather
than the closest namespace to the element. Just to be clear this also
only adds namespaces that are used, not those defined but not
Hello Rob,
Tuesday, November 1, 2005, 11:49:22 AM, you wrote:
> Marcus,
> There are a few problems with this approach. A crash will occur if
> adding a default namespace as there is no prefix.
> The namespace should probably be the key and prefix the value or default
> namespaces should not be
Marcus,
There are a few problems with this approach. A crash will occur if
adding a default namespace as there is no prefix.
The namespace should probably be the key and prefix the value or default
namespaces should not be added.
When doing a recursive get, if the namespace is redefined furth
never mind. i thought you were changing a PHP function :)
At 06:05 AM 8/16/2005 +, Rolland Santimano wrote:
rolland Tue Aug 16 02:05:00 2005 EDT
Modified files:
/php-src/ext/simplexml simplexml.c
/php-src/ext/standard http_fopen_wrapper.c
Log:
Updated calls
my reaction was two-fold:
a) I'd already reverted this exact change beforehand (or at least
vetoed it), so i was surprised to see it without being consulted first,
to which I would of said: "I think this is utter bollocks, but I'm done
arguing the issue." Same holds true now :)
b) It should be
Sterling,
We're moving to studlyCaps. And don't say it makes a difference in SimpleXML :)
Andi
At 01:28 PM 3/29/2004 -0800, Sterling Hughes wrote:
wtf?
-sterling
On Mar 29, 2004, at 11:58 AM, Marcus Boerger wrote:
helly Mon Mar 29 14:58:01 2004 EDT
Modified files:
/php-src/ex
wtf?
-sterling
On Mar 29, 2004, at 11:58 AM, Marcus Boerger wrote:
helly Mon Mar 29 14:58:01 2004 EDT
Modified files:
/php-src/ext/simplexml simplexml.c
/php-src/ext/simplexml/tests001.phpt 002.phpt 003.phpt 004.phpt
007.phpt 008.p
Yeah, that's a good idea.
At 09:35 PM 2/11/2004 +0100, Marcus Boerger wrote:
Hello Zeev,
Wednesday, February 11, 2004, 8:15:34 PM, you wrote:
> zeev Wed Feb 11 14:15:34 2004 EDT
> Modified files:
> /php-src/ext/simplexmlsimplexml.c
> Log:
> Prevent SimpleXML from silently
Hello Zeev,
Wednesday, February 11, 2004, 8:15:34 PM, you wrote:
> zeev Wed Feb 11 14:15:34 2004 EDT
> Modified files:
> /php-src/ext/simplexmlsimplexml.c
> Log:
> Prevent SimpleXML from silently modifying types of variables that are
> assigned to its obje
Index: php-src/ext/simplexml/tests/009.phpt
+++ php-src/ext/simplexml/tests/009.phpt
--TEST--
SimpleXML and XPath
--SKIPIF--
--FILE--
$val) {
var_dump($name);
var_dump(get_class($val));
var_dump(trim((string)$val));
}
?>
===DONE===
--EXPECT--
string(5) "elem1"
string(17) "s
Do these changes (the ones you made later too) fix any of the
bugs reported at bugs.php.net ?
(all 3 assigned to you)
--Jani
On Mon, 3 Nov 2003, Sterling Hughes wrote:
>sterling Mon Nov 3 08:36:24 2003 EDT
>
> Modified files:
>/php
Should the same fix be applied to new com stuff?
My guess is yes :)
--Wez.
> - f->type = ZEND_OVERLOADED_FUNCTION;
> + f->type = ZEND_OVERLOADED_FUNCTION_TEMPORARY;
--
PHP CVS Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
At 18:10 18/08/2003, Rob Richards wrote:
From: Zeev Suraski
> PZVAL_UNLOCK() should *never* be used by anything outside the engine. The
> leak is somewhere else.
If that should never be called outside the engine, then the issue seems to
be from any extension that needs to override zend_std_read_p
From: Zeev Suraski
> PZVAL_UNLOCK() should *never* be used by anything outside the engine. The
> leak is somewhere else.
If that should never be called outside the engine, then the issue seems to
be from any extension that needs to override zend_std_read_property. DOM has
the same issue, as it c
At 15:47 18/08/2003, Rob Richards wrote:
Without PZVAL_UNLOCK(return_value), sxe_property_read seems to leak every
time it is called.
PZVAL_UNLOCK() should *never* be used by anything outside the engine. The
leak is somewhere else.
Even with it there, I see leaks when returning a single value si
Without PZVAL_UNLOCK(return_value), sxe_property_read seems to leak every
time it is called.
Even with it there, I see leaks when returning a single value since it was
only called when returning an array.
What was it crashing on?
Rob
> zeev Mon Aug 18 07:26:12 2003 EDT
>
> Modified files:
>
Cool! This is what I meant when I said it would reduce at least 80% of
the code in pear::soap. :)
Shane
Sterling Hughes wrote:
sterling Fri Jul 4 17:58:09 2003 EDT
Modified files:
/php-src/ext/simplexml simplexml.c
Log:
fix memory cverrun when accessing an empty xml
Sterling,
Extensions shouldn't really be using stuff like PZVAL_UNLOCK(). Can you
explain what you are trying to accomplish?
Andi
At 12:33 AM 5/7/2003 +, Sterling Hughes wrote:
sterlingFri Jul 4 20:33:13 2003 EDT
Modified files:
/php-src/ext/simplexml simplexml.
uhh, other way around. :)
-Sterling
On Fri, 2003-07-04 at 20:33, Sterling Hughes wrote:
> sterling Fri Jul 4 20:33:13 2003 EDT
>
> Modified files:
> /php-src/ext/simplexmlsimplexml.c
> Log:
> not a temp var unless its a singular var
>
>
> Index: p
28 matches
Mail list logo
