On 23 Jul 2002, Bojan Smojver wrote:
> Before I do that, some questions about uriEnv in jk2_handler(), since
> that part is very different to mod_jk. The initial test involves asking:
>
> if (uriEnv==NULL || strcmp(r->handler,JK_HANDLER)!= 0)
uriEnv == null means no match was found.
It is set by either jk2_translate or by a "JkSet" in a <Location>
context.
> After the first test, there is the second test that goes:
>
> if( uriEnv == NULL )
Yes, that's some old code that tryed to support the case when
someone would do:
<Location>
SetHandler jakarta/servlet
</Location>
I'll remove it - as you mention, it's obviously not working,
and I'm not sure it is a good idea - doing a JkSet has the
same effect and it's not worth the extra complexity.
> How do I go through this mine field? What's going to be the value of
> uriEnv if r->handler is DIR_MAGIC_TYPE? If it's going to be NULL, I have
It's jk_translate who decides if a request is to be handled by
jk ( by mapping it to a uriEnv ).
You can add a test for r->handler==DIR_MAGIC_TYPE, but don't assume
any uriEnv is set.
uriEnv is the equivalent of a servlet mapping ( or the Container
in 3.3 ).
You should still do an internal mapping in handler and find
some per/context or per/vhost uriEnv.
Costin
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
