hi!
I like the idea to improve this area. However I'm with Stas here. I do
not feel comfortable enough with this RFC to agree to get into 5.4 at
this stage. It is a sensible area and we are already late (because of
the tests) with the 5.4 release plan.
Next will begin in spring next year, not too
Hi!
On 9/12/11 3:45 PM, Gustavo Lopes wrote:
It's a change in behavior so it makes sense. Those operations are wrapper
operations and by their nature they are static operations, meaning a
stream instance is not required. See the difference between php_stream_ops
and php_stream_wrapper_ops. It wo
On Mon, 12 Sep 2011 18:00:13 +0100, Hannes Magnusson
wrote:
On Mon, Sep 12, 2011 at 16:56, Gustavo Lopes
wrote:
Ah, right. stat() and unlink() should have been static in the first
place.
Given the circumstances, it might be a good idea, to document url_stat,
unlink, rename, mkdir and rmdi
On Mon, Sep 12, 2011 at 16:56, Gustavo Lopes wrote:
> Em Mon, 12 Sep 2011 15:40:20 +0100, Christian Kaps
> escreveu:
>
>> On Mon, 12 Sep 2011 14:43:33 +0100, Gustavo Lopes wrote:
>>>
>>> Em Mon, 12 Sep 2011 12:53:13 +0100, Sebastian Bergmann
>>> escreveu:
>>>
Regarding state it is importa
Em Mon, 12 Sep 2011 15:40:20 +0100, Christian Kaps
escreveu:
On Mon, 12 Sep 2011 14:43:33 +0100, Gustavo Lopes wrote:
Em Mon, 12 Sep 2011 12:53:13 +0100, Sebastian Bergmann
escreveu:
Regarding state it is important to notice that PHP does *not* execute
the constructor on all low level
On Mon, 12 Sep 2011 14:43:33 +0100, Gustavo Lopes wrote:
Em Mon, 12 Sep 2011 12:53:13 +0100, Sebastian Bergmann
escreveu:
Regarding state it is important to notice that PHP does *not*
execute
the constructor on all low level calls when instantiating the
wrapper
class - for whatever reason
Em Mon, 12 Sep 2011 12:53:13 +0100, Sebastian Bergmann
escreveu:
Regarding state it is important to notice that PHP does *not* execute
the constructor on all low level calls when instantiating the wrapper
class - for whatever reason that is the case. Changing that behaviour
would cause qu
Am 12.09.2011 00:26, schrieb Gustavo Lopes:
> A patch against trunk (or 5.4) would have been nicer. Other than that:
> * This patch has a huge BC:
Johannes already said that he is updating his patch. It is not our
intention to break BC, we only want to add additional/optional
behaviour.
> Coul
Hi,
On Sun, 2011-09-11 at 23:26 +0100, Gustavo Lopes wrote:
> On Sun, 11 Sep 2011 20:26:20 +0100, Sebastian Bergmann
> wrote:
>
> > [...]
> > --
> > [1] https://wiki.php.net/rfc/streamwrapper-factory
> > [2] http://schlueters.de/~johannes/php/stream_factory.diff
> >
>
> A patch against tr
On Sun, 11 Sep 2011 20:26:20 +0100, Sebastian Bergmann
wrote:
[...]
--
[1] https://wiki.php.net/rfc/streamwrapper-factory
[2] http://schlueters.de/~johannes/php/stream_factory.diff
A patch against trunk (or 5.4) would have been nicer. Other than that:
* This patch has a huge BC:
PHP
Hi,
Am 11.09.2011 23:40, schrieb Anthony Ferrara:
So, based on the rfc, does "must return an instance" mean that it can't
throw an exception (ex: if a resource couldn't be opened)?
A resource is opened on `streamWrapper::dir_opendir()` or
`streamWrapper::stream_open()` anyway, but not on Wrap
So, based on the rfc, does "must return an instance" mean that it can't
throw an exception (ex: if a resource couldn't be opened)?
On Sep 11, 2011 3:26 PM, "Sebastian Bergmann" wrote:
> We (Arne Blankerts, Stefan Priebsch, Benjamin Eberlei, and I) have
> worked on/with code where a factory for str
We (Arne Blankerts, Stefan Priebsch, Benjamin Eberlei, and I) have
worked on/with code where a factory for stream wrappers would be very
helpful.
Earlier this year we talked to Ilia about this at ConFoo and he was open
to the idea but did not have the time to implement it. At FrOSCon we
talk
13 matches
Mail list logo
