git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff King <peff@github.com>
To: "René Scharfe" <rene.scharfe@lsrfire.ath.cx>
Cc: git@vger.kernel.org, git-dev@github.com
Subject: Re: [PATCH 2/2] archive: support gzipped tar files
Date: Tue, 14 Jun 2011 16:14:33 -0400	[thread overview]
Message-ID: <20110614201433.GB1567@sigill.intra.peff.net> (raw)
In-Reply-To: <4DF7B90B.9050802@lsrfire.ath.cx>

On Tue, Jun 14, 2011 at 09:39:55PM +0200, René Scharfe wrote:

> > However, when running git-archive against a remote site,
> > having the remote side do the compression can save
> > considerable bandwidth. Service providers could always wrap
> > git-archive to provide that functionality, but this makes it
> > much simpler.
> 
> That's a good point and one that was overlooked when this topic came up
> earlier (see http://kerneltrap.org/mailarchive/git/2009/9/10/11507 and
> http://kerneltrap.org/mailarchive/git/2009/9/11/11577).

Hmph, I should have done my homework better. I totally missed that
thread.

Yeah, I am unsurprised that doing it in a single process is actually
slower. I do think because of the remote issue that we should provide
something like this. But we could implement it by piping to an external
gzip. That would make us just slightly less portable, but would give us
the multi-processor speedup, or even allow using something like pigz.

> > +static void output_gz(const char *buf, unsigned long len)
> > +{
> > +	if (!gzwrite(gz_file, buf, len))
> > +		die("unable to write compressed stream: %s",
> > +		    gzerror(gz_file, NULL));
> > +}
> 
> Does this do the right things when faced with interrupted writes or
> truncated pipes? I ask because the earlier attempt had a
> gzwrite_or_die() which did that, but I don't know anymore if that is
> strictly needed.

No, I blindly assumed that gzwrite was a little bit smart, but looking
at the zlib code, it really is just propagating whatever it got from
fwrite. I need to handle both errors and short writes myself. So we do
need gzwrite_or_die.

> Oh, and bridging the gap between unsigned long and int
> was certainly another reason for the existence of this function.

Ugh. I correctly saw that it took an unsigned long, but it actually
returns the number of bytes written as an int! Nice interface.

All of this can go away, though, if we switch to an external process.
It's tempting.

-Peff

  reply	other threads:[~2011-06-14 20:14 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-14 18:17 [PATCH 1/2] archive: factor out write phase of tar format Jeff King
2011-06-14 18:18 ` [PATCH 2/2] archive: support gzipped tar files Jeff King
2011-06-14 19:25   ` J.H.
2011-06-14 19:30     ` Jeff King
2011-06-14 19:39   ` René Scharfe
2011-06-14 20:14     ` Jeff King [this message]
2011-06-14 20:45       ` Jeff King
2011-06-15 22:30         ` [RFC/PATCH 0/7] user-configurable git-archive output formats Jeff King
2011-06-15 22:31           ` [PATCH 1/7] archive: reorder option parsing and config reading Jeff King
2011-06-15 22:33           ` [PATCH 2/7] archive: add user-configurable tar-filter infrastructure Jeff King
2011-06-15 23:33             ` Junio C Hamano
2011-06-16  0:29               ` Jeff King
2011-06-15 22:33           ` [PATCH 3/7] archive: support user tar-filters via --format Jeff King
2011-06-15 22:33           ` [PATCH 4/7] archive: advertise user tar-filters in --list Jeff King
2011-06-15 22:34           ` [PATCH 5/7] archive: refactor format-guessing from filename Jeff King
2011-06-15 23:48             ` Junio C Hamano
2011-06-16  0:34               ` Jeff King
2011-06-15 22:34           ` [PATCH 6/7] archive: match extensions from user-configured formats Jeff King
2011-06-15 22:35           ` [PATCH 7/7] archive: provide builtin .tar.gz filter Jeff King
2011-06-15 23:55             ` Junio C Hamano
2011-06-15 23:57               ` Junio C Hamano
2011-06-16  0:38               ` Jeff King
2011-06-16  6:27                 ` Junio C Hamano
2011-06-16  6:51                   ` Jeff King
2011-06-16  7:56                     ` Chris Webb
2011-06-16 17:46                       ` Jeff King
2011-06-16 18:02                         ` Junio C Hamano
2011-06-16 18:21                           ` Jeff King
2011-06-16 18:27                             ` John Szakmeister
2011-06-16 18:42                             ` Junio C Hamano
2011-06-16 18:57                               ` Jeff King
2011-06-18 14:52           ` [RFC/PATCH 0/7] user-configurable git-archive output formats René Scharfe
2011-06-18 15:28             ` Jakub Narebski
2011-06-20 15:58             ` Junio C Hamano
2011-06-22  1:19               ` [PATCHv2 0/9] configurable tar compressors Jeff King
2011-06-22  1:20                 ` [PATCHv2 1/9] archive: reorder option parsing and config reading Jeff King
2011-06-22  1:22                 ` [PATCHv2 2/9] archive-tar: don't reload default config options Jeff King
2011-06-22  1:23                 ` [PATCHv2 3/9] archive: refactor list of archive formats Jeff King
2011-06-23 17:05                   ` Thiago Farina
2011-06-23 17:30                     ` Jeff King
2011-06-22  1:24                 ` [PATCHv2 4/9] archive: pass archiver struct to write_archive callback Jeff King
2011-06-22  1:24                 ` [PATCHv2 5/9] archive: move file extension format-guessing lower Jeff King
2011-06-22  1:25                 ` [PATCHv2 6/9] archive: refactor file extension format-guessing Jeff King
2011-06-22  1:26                 ` [PATCHv2 7/9] archive: implement configurable tar filters Jeff King
2011-06-22  1:45                   ` Jeff King
2011-06-22  6:09                   ` René Scharfe
2011-06-22 14:59                     ` Jeff King
2011-06-22  1:27                 ` [PATCHv2 8/9] archive: provide builtin .tar.gz filter Jeff King
2011-06-22  1:35                 ` [PATCHv2 9/9] upload-archive: allow user to turn off filters Jeff King
2011-06-22  3:17                   ` Jeff King
2011-06-21 16:01             ` [RFC/PATCH 0/7] user-configurable git-archive output formats Jeff King
2011-06-18 15:40           ` René Scharfe
2011-06-14 20:30   ` [PATCH 2/2] archive: support gzipped tar files Junio C Hamano
2011-06-14 20:49     ` Jeff King
2011-06-14 23:40       ` Miles Bader
2011-06-15 22:46         ` Jeff King

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110614201433.GB1567@sigill.intra.peff.net \
    --to=peff@github.com \
    --cc=git-dev@github.com \
    --cc=git@vger.kernel.org \
    --cc=rene.scharfe@lsrfire.ath.cx \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).