Ray Van Dolson wrote:

Both blocks work, but I'm thinking I could potentially end up calling
gzdeflate on data that wasn't originally compressed.  Maybe someone can
clarify:

Yep, your config looks OK to me.  Where's the problem compressing
data that didn't start out compressed (or vice versa)?

  - Does each member of a FilterChain get called regardless?  For
    example:

      text/html (uncompressed) doesn't match on the gzinflate line, but
      would one he next two.  I'd end up with compressed output that
      wasn't originally compressed.

Yep.

  - If so, should I set an environment variable if Content-Encoding =
    gzip and then have my gzdeflate provider check for it instead of
    operating blindly on all text/* ?

Why?

    Or maybe some way to check if the request we're responding to had
    Accept-Encoding gzip set?

Aha!  That's a different question.  mod_deflate has that logic built in
(and has done since before mod_filter existed).  See force-gzip in the
docs.

Couple of other points.  First, inflate - substitute - deflate is
a lot of computation, so if you can cache it (e.g. with mod_cache)
you could get much better performance.  Second, you may want to
look at expression evaluation in apache: mod_filter in 2.3 and
up (but not in 2.2.x) can evaluate complex expressions to
determine whether to run a filter.

--
Nick Kew

---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
  "   from the digest: users-digest-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org

Reply via email to