Re: [Reproducible-builds] GSoC 2015 Week 9: Move forward reproducible builds

2015-07-25 Thread Mattia Rizzolo
On Sat, Jul 25, 2015 at 05:24:50PM +0200, Holger Levsen wrote:
> On Samstag, 25. Juli 2015, Dhole wrote:
> [...]
> > I have also uploaded the package in our APT repository.
> [...]
> > For next week I plan to send the ghostscript patches to debian and probably
> > upstream[...]
> 
> I'm aware that some people only want to file bugs with working and verified 
> patches and that uploading to our repo is a good way to achieve that, but at 
> the same time I'm worried that such work might get lost if no tracking bug is 
> filed from the start...
> 
> Or do you (all) think it's enough to track such work via patches in a git 
> repo? (And then file a bug once the patch is ready - and should the driving 
> person of this go MIA we will notice and have the git repo?!?)

I do believe that testing the patches before sending out the bug is somehow
important. As we have this awesome policy to strive to attach patches to bugs
from us i very like it, and as a maintainer I'd really prefer getting a single
email with patch than several with maybe days between.

IMO the better approach would be:
 1) working on it and get a patch
 2) test it out (uploading to our repo + schedule something is good+enough)
 3) if that works fine go ahead, otherwise back to point 1.
 4) file a bug (always file it in the debian BTS)
 5) edit the package our repos to add the bug # to the patch header (see DEP-3)
and the changelog entry

As i seen it this usually takes few days to go from 1 to 5, imho we don't risk
too much to lost track of it. and anyway looks like some of us ciclely look at
our repo state farly often.

> Another option would be to file a tracking bug against qa.d.o (usertagged 
> reproducible.d.n) for patch development + tracking...?!?

nah, that would annoy people on -qa@ who are not interested in reproducible
stuff, for a start.

>   Holger, who really loves bug #s but maybe a bit too much... ;-) 

;)

Be assured, I agree that once we have something on our hands attaching it to a
bug is a must! :)

-- 
regards,
Mattia Rizzolo

GPG Key: 66AE 2B4A FCCF 3F52 DA18  4D18 4B04 3FCD B944 4540 .''`.
more about me:  http://mapreri.org : :'  :
Launchpad user: https://launchpad.net/~mapreri `. `'`
Debian QA page: https://qa.debian.org/developer.php?login=mattia `-


signature.asc
Description: Digital signature
___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds

Re: [Reproducible-builds] GSoC 2015 Week 9: Move forward reproducible builds

2015-07-25 Thread Holger Levsen
Hi Dhole,

thanks for your continious reports and cheers and kudos for all the nice work! 
Besides thinking "yay" this sparked one question:

On Samstag, 25. Juli 2015, Dhole wrote:
[...]
> I have also uploaded the package in our APT repository.
[...]
> For next week I plan to send the ghostscript patches to debian and probably
> upstream[...]

[general comment, more or less]

I'm aware that some people only want to file bugs with working and verified 
patches and that uploading to our repo is a good way to achieve that, but at 
the same time I'm worried that such work might get lost if no tracking bug is 
filed from the start...

Or do you (all) think it's enough to track such work via patches in a git 
repo? (And then file a bug once the patch is ready - and should the driving 
person of this go MIA we will notice and have the git repo?!?)

Another option would be to file a tracking bug against qa.d.o (usertagged 
reproducible.d.n) for patch development + tracking...?!?


cheers,
Holger, who really loves bug #s but maybe a bit too much... ;-) 


signature.asc
Description: This is a digitally signed message part.
___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds