perl-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Philippe M. Chiasson" <go...@ectoplasm.org>
Subject Re: cvs commit: modperl-2.0/todo release
Date Mon, 09 Aug 2004 17:41:12 GMT


Stas Bekman wrote:
> Philippe M. Chiasson wrote:
> [...]
> 
>>>If the macro is supposed to suggest the allowed context and PerlModule 
>>>is not allowed in DIR, then DIR is not appropriate. I see no problem 
>>>with long names if they require no extra source code reading to figure 
>>>out what they really do. These are internal macros and there are used 
>>>in one place, so what's the big deal with choosing a clear name, even 
>>>if it's long?
>>
>>
>>Well, there are 2 macros that must work together, so one possibility 
>>would be so call
>>
>> MP_CMD_ANYWHERE_RAW_ARGS("Perl", perldo, "Perl Code"),
>>
>>And clearly at the top of the perldo implementation:
>> MP_CMD_SRV_DECLARE(requires)
>> {
>>     MP_dSCFG(parms->server);
>>     MP_PERL_DECLARE_CONTEXT;
>>     MP_CMD_SERVER_OR_HTACCESS_ONLY;
>>
>>Would something like that be somewhat clearer ?
> 
> 
> We have the following contexts:
> 
> DIR
> SRV
> HTACCESS

Well, as far as httpd is concerned, we have:

SRV
DIR (HTACCESS)
ALL

> each directive can be either:
> 
> DIR
> SRV
> SRV + HTACCESS
> DIR + HTACCESS
> 
> correct? So why not base the naming based on that?

More like

DIR
SRV
ALL
ALL -DIR (SRV & HTACCESS)
ALL -HTACCESS (SRV & DIR)

So, I'd suggest this naming

MP_CMD_DIR_...
MP_CMD_SRV_...
MP_CMD_ALL_...

with 2 new macros :
   MP_CMD_NO_DIR
   MP_CMD_NO_HTACCESS

To add to the top of a directive handler ?
-- 
--------------------------------------------------------------------------------
Philippe M. Chiasson m/gozer\@(apache|cpan|ectoplasm)\.org/ GPG KeyID : 88C3A5A5
http://gozer.ectoplasm.org/     F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 88C3A5A5

Mime
View raw message