On Sat, Aug 25, 2007, Petr Baudis wrote:
quoted text
> On Wed, Jul 25, 2007 at 08:39:43PM CEST, Jakub Narebski wrote:

quoted text
>> Check if PerlIO::gzip is available, and if it is make it possible to
> 
> It doesn't really check if the require succeeded. Either the description
> or (preferrably, but not a showstopper, IMO) the code should be
> adjusted.

It does not check if require succeeded (I could do that this way),
but instead checks if $PerlIO::gzip::VERSION is defined (if it is true).

our $enable_transparent_compression = !! $PerlIO::gzip::VERSION;
 
quoted text
>> enable (via 'compression' %feature) transparent compression of HTML
>> output.  Error messages and any non-HTML output are excluded from
>> transparent compression.
>> 
>> Signed-off-by: Jakub Narebski <jnareb@gmail.com>
> 
> Acked-by: Petr Baudis <pasky@suse.cz>

By the way, this was more "proof of concept" than solution of an itch.
 
quoted text
> I'd put it on repo.or.cz... too bad that there I value CPU much more
> than the bandwidth. ;-)
> 
> Why did you exclude non-HTML output from transparent compression? Me and
> I guess other people too sometimes download rather large chunks of raw
> data over gitweb.

Because it was easiest. We have single point of entry for HTML output
(the git_header_html subroutine), but we don't have anything similar for
non-HTML output. And we most certainly wouldn't want to enable transparent
compression for snapshots and 'blob_plain' view for compressed files,
including png, gif, jpeg, zip, mp3, ogg,...

-- 
Jakub Narebski
Poland
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html