[pacman-dev] Should pacman show just one download progress when a mirror uses redirection?

[Alexandre Filgueira]

>
> I'm more inclined to believe that this isn't due to the redirect so much
> as the size of the response headers. Pulling any random package off the
> arch website (which has a 302 redirect) doesn't show this behavior:


You're right, my headers are bigger, mirrorbrain include many things:

[faidoc at Delorean ~]$ curl -sD- -o/dev/null
> http://mirrors.cinnarch.com/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz
> HTTP/1.1 302 Found
> Date: Thu, 10 Jan 2013 18:49:17 GMT
> Server: Apache/2.2.16 (Debian)
> X-MirrorBrain-Mirror: softlibre.unizar.es
> X-MirrorBrain-Realm: country
> Link: <
> http://mirrors.cinnarch.com/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz.meta4>;
> rel=describedby; type="application/metalink4+xml"
> Link: <
> http://softlibre.unizar.es/cinnarch/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz>;
> rel=duplicate; pri=1; geo=es
> Link: <
> http://www.mirrorservice.org/sites/repo.cinnarch.com/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz>;
> rel=duplicate; pri=2; geo=gb
> Link: <
> http://cinnarch.limun.org/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz>;
> rel=duplicate; pri=3; geo=de
> Digest: MD5=rmURnbMAoRYoNhtja1VQgA==
> Digest: SHA=Hu0X2ym8sBEUQGaXjvFQHdIctgw=
> Digest: SHA-256=nfFkPNZd8vq05MUCxOVqlwE3YMrAJzT9R0xga6P0DNY=
> Location:
> http://softlibre.unizar.es/cinnarch/cinnarch-core/x86_64/cinnarch-keyring-20120717-3-any.pkg.tar.xz
> Vary: Accept-Encoding
> Content-Length: 369
> Content-Type: text/html; charset=iso-8859-1




2013/1/8 Dave Reisner <d at falconindy.com>

> On Tue, Jan 08, 2013 at 12:39:22AM +0100, Alexandre Filgueira wrote:
> > Hi
> >
> > I have a repo using redirections to let the server get the closest mirror
> > for the client using Mirrorbrain software. When configuring pacman to
> that
> > mirror, he follows the redirection as expected by using FOLLOWLOCATION in
> > curl opts, but he shows 2 download progress, I guess that one is for the
> > redirection and the last one for the file itself. So my question is...
> > should be look like that? or would be correct to detect redirections and
> > show just one download progress?
>
> I'm more inclined to believe that this isn't due to the redirect so much
> as the size of the response headers. Pulling any random package off the
> arch website (which has a 302 redirect) doesn't show this behavior:
>
>  $ sudo pacman -U
> https://www.archlinux.org/packages/extra/x86_64/aalib/download/
>   alpmtmp.q1pXbL           144.7 KiB   591K/s 00:00
> [----------------------] 100%
>   loading packages...
>   resolving dependencies...
>   looking for inter-conflicts...
>
>   Packages (1):
>
>   Name   New Version  Net Change
>
>   aalib  1.4rc5-9       0.75 MiB
>
>   Total Installed Size:   0.75 MiB
>
>   Proceed with installation? [Y/n]
>
> The header size is a full 100 bytes smaller than your example:
>
>  $ curl -sD- -o/dev/null
> https://www.archlinux.org/packages/extra/i686/aalib/download/ | wc -c
>  287
>
> >
> > This is an example of what is writting pacman:
> >
> > Proceed with installation? [Y/n] y
> > > :: Retrieving packages from cinnarch-repo...
> > >  faenza-icon-theme-1...   382.0   B  6.51M/s 00:00
> > > [######################] 100%
> > >  faenza-icon-theme-1...    10.8 MiB   333K/s 00:16
> > > [##############--------]  66%
> >
>
> Uggh. IMO, curl really shouldn't be doing this and should only be
> triggering the progress callback for the response body.
>
> d
>
>