--- Andi Gutmans <[EMAIL PROTECTED]> wrote:
> Please checkout the CVS and let me know if it works
> for you.
>
Hello Andi,
It seems to be working.
I´m going to make more tests. If anything goes wrong
I´ll tell you.
If somebody else would like to test, would help i
think.
regards,
Eduard
> --- Timm Friebe <[EMAIL PROTECTED]> wrote:
> > Andi Gutmans wrote:
> > Remember if anyone has posted a patch for this in
> > the past?
>
> I did:
> http://zend.com/lists/engine2/200306/msg00056.html
How about Marcus ideia http://zend.com/lists/engine2/200306/msg00078.html
___
On Tue, 2003-11-18 at 23:01, Andi Gutmans wrote:
> Yeah, this is something we need to tackle. i think we're going to have to
> keep the pretty case version but it's going to be a slight hit in performance.
> Remember if anyone has posted a patch for this in the past?
I did: http://zend.com/lists/
On Wed, 19 Nov 2003, Andi Gutmans wrote:
> Yes, this is definitely a problem. Today, when possible, we lowercase the
> class at compile-time so that we save time during execution. Supporting
> this will lead to a small slow down.
> I tend to agree that __autoload() should support this but I'm not
On Wed, 19 Nov 2003, Andi Gutmans wrote:
>At 11:09 AM 11/19/2003 -0800, Eduardo R. Maciel wrote:
>>Well,
>>
>>At the point the classname is passed to __autoload
>>is it already in lowercase???
>
>Yes it's already lowercase.
>We will need to keep it case-sensitive until this stage. Might not be
At 11:09 AM 11/19/2003 -0800, Eduardo R. Maciel wrote:
Well,
At the point the classname is passed to __autoload
is it already in lowercase???
Yes it's already lowercase.
We will need to keep it case-sensitive until this stage. Might not be too
bad because I have thought of a way of improving t
Well,
At the point the classname is passed to __autoload
is it already in lowercase???
If not, maybe you could just do a single stat in the
file the way it is defined in the __autoload()
function, with no sanity checks (to avoid performance
loss) and include it.
And warn users about it o
Yes, this is definitely a problem. Today, when possible, we lowercase the
class at compile-time so that we save time during execution. Supporting
this will lead to a small slow down.
I tend to agree that __autoload() should support this but I'm not sure if
we should change everything to run-time
Yeah, this is something we need to tackle. i think we're going to have to
keep the pretty case version but it's going to be a slight hit in performance.
Remember if anyone has posted a patch for this in the past? If not I'll try
and look into fixing it tomorrow.
Andi
At 09:42 AM 11/18/2003 -080