chdir

[David Welton <da...@gmail.com>]

Hi,

I just wanted to add a note about chdir.  The reason it's not in the
apr_ stuff is that Apache 2 has the potential to be run 'threaded' -
which means that having a cwd is not really possible.

However... I think that it's better to get something working rather
than worrying about it being perfect - that's the big mistake I made
with Apache 2 in the past, worrying about all the threading stuff
instead of getting something working and then improving it.  Do keep
it in mind as something to work on in the future, though.

-- 
David N. Welton
http://www.welton.it/davidw/

---------------------------------------------------------------------
To unsubscribe, e-mail: rivet-dev-unsubscribe@tcl.apache.org
For additional commands, e-mail: rivet-dev-help@tcl.apache.org

Re: chdir

[Massimo Manghi <ma...@unipr.it>]

David Welton wrote:
> Hi,
>
> I just wanted to add a note about chdir.  The reason it's not in the
> apr_ stuff is that Apache 2 has the potential to be run 'threaded' -
> which means that having a cwd is not really possible.
>
>   
> However... I think that it's better to get something working rather
> than worrying about it being perfect - that's the big mistake I made
> with Apache 2 in the past, worrying about all the threading stuff
> instead of getting something working and then improving it.  Do keep
> it in mind as something to work on in the future, though.
>
>   
I hope I'm not being rebuked too hard by David for having wasted time
trying to understand how to approach this problem in a new way:
I  was simply doing some practice in order to get familiar with this stuff.

Yesterday night I was playing around with the request_rec fields. I 
could not
figure out in what cases the 'filename' and 'canonical_filename' are 
different...
Apache2 docs simply say that 'filename' gets canonicalized into 
'canonical_filename'
if they don't match. I thought they could be different when a URI makes 
a reference
to an alias or the path contains a reference to an aliased directory but 
I always get
the real file path for both fields.

My attempt was seeking to understand if we could use Tcl_  calls in 
order to extract
the full path to the current page, put this information into the  
TclWebRequest and
use it later on in the 'parse' and 'source' commands. Just for the 
benefit of the time
when the problem will be addressed in our original way.

I wrote this message as a future reference for that day...

 -- Massimo


---------------------------------------------------------------------
To unsubscribe, e-mail: rivet-dev-unsubscribe@tcl.apache.org
For additional commands, e-mail: rivet-dev-help@tcl.apache.org