You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by André Malo <nd...@perlig.de> on 2002/12/18 23:51:13 UTC

Re: cvs commit: apache-1.3/src/modules/standard mod_autoindex.c

* wrowe@apache.org wrote:

>   +    /*
>   +     * If there's a header file, send a subrequest to look for it.  If it's
>   +     * found and html do the subrequest, otherwise handle it
>   +     */
>   +    r_accept = ap_table_get(hdrs, "Accept");
>   +    r_accept_enc = ap_table_get(hdrs, "Accept-Encoding");
>   +    ap_table_setn(hdrs, "Accept", "text/html, text/plain;q=.5, text/*;q=.1");
>   +    ap_table_unset(hdrs, "Accept-Encoding");

by the way. In 1.3 it's perhaps not interesting, but in 2.x this causes the 
problem, that mod_deflate won't be invoked, if a HeaderName file will be 
inserted. We delete the Accept-Encoding header, put the contents of the 
header file onto the filter stack, mod_deflate doesn't recognize an 
accepted encoding, it removes itself from the stack, we re-add the 
Accept-Encoding, now it's too late.

However, is there no possibility to modify the accept values only for the 
subrequest?

nd
-- 
> [...] weiß jemand zufällig, was der Tag DIV ausgeschrieben bedeutet?
DIVerses. Benannt nach all dem unstrukturierten Zeug, was die Leute da
so reinpacken und dann absolut positionieren ...
                           -- Florian Hartig und Lars Kasper in dciwam

Re: cvs commit: apache-1.3/src/modules/standard mod_autoindex.c

Posted by André Malo <nd...@perlig.de>.
* William A. Rowe, Jr. wrote:

> The request headers_in are only modified for the duration of
> looking up the potential subrequests.
> 
> So I'm confused why (for that brief time) it has been messing
> up mod_deflate's filter?

mod_autoindex.c (2.1) line 1000 removes the accept-encoding.
Then it does the sub_req_uri_lookup, if ok, it runs either ap_run_sub_req 
or do_emit_plain. If I understand it correctly, both put already data onto 
to filter stack.
And *after* the first data (i.e. the header file) is emitted, mod_autoindex 
re-adds the accept-encoding. That seems to be too late for mod_deflate.

the conclusion is:
AddOutputFilterByType DEFLATE httpd/unix-directory
doesn't work in the case, HeaderName is used (in my tests).

nd
-- 
die (eval q-qq[Just Another Perl Hacker
]
;-)
# André Malo, <http://www.perlig.de/> #

Re: cvs commit: apache-1.3/src/modules/standard mod_autoindex.c

Posted by "William A. Rowe, Jr." <wr...@apache.org>.
The request headers_in are only modified for the duration of
looking up the potential subrequests.

So I'm confused why (for that brief time) it has been messing
up mod_deflate's filter?

Bill

At 04:51 PM 12/18/2002, André Malo wrote:
>* wrowe@apache.org wrote:
>
>>   +    /*
>>   +     * If there's a header file, send a subrequest to look for it.  If it's
>>   +     * found and html do the subrequest, otherwise handle it
>>   +     */
>>   +    r_accept = ap_table_get(hdrs, "Accept");
>>   +    r_accept_enc = ap_table_get(hdrs, "Accept-Encoding");
>>   +    ap_table_setn(hdrs, "Accept", "text/html, text/plain;q=.5, text/*;q=.1");
>>   +    ap_table_unset(hdrs, "Accept-Encoding");
>
>by the way. In 1.3 it's perhaps not interesting, but in 2.x this causes the 
>problem, that mod_deflate won't be invoked, if a HeaderName file will be 
>inserted. We delete the Accept-Encoding header, put the contents of the 
>header file onto the filter stack, mod_deflate doesn't recognize an 
>accepted encoding, it removes itself from the stack, we re-add the 
>Accept-Encoding, now it's too late.
>
>However, is there no possibility to modify the accept values only for the 
>subrequest?
>
>nd
>-- 
>> [...] weiß jemand zufällig, was der Tag DIV ausgeschrieben bedeutet?
>DIVerses. Benannt nach all dem unstrukturierten Zeug, was die Leute da
>so reinpacken und dann absolut positionieren ...
>                           -- Florian Hartig und Lars Kasper in dciwam