Re: [PATCH 00/46] A re-styling of Patchwork

2014-11-10 Thread Brian Norris
On Sat, Nov 08, 2014 at 07:08:18PM +, Damien Lespiau wrote:
> A small introduction first:
> 
> I'm part of the i915 driver team and we'd like to use patchwork. However, 
> there
> are a few things getting in the way: the bigger item is that patchwork doesn't
> scale very well above a certain threshold of patches.
> 
> We have a number of ideas to improve the situation, mainly being able to
> manipulate a higher level object, the "series", instead of "patches". There's 
> a
> a somewhat unstructured todo list at:
> 
>   https://github.com/dlespiau/patchwork/wiki/TODO

+1 for most of the "series" TODO items. Several of the other items have
been tripping points for me too. And I know that there are other users
who would appreciate thread-based / series-based handling of patches
("bundles" don't really cut it in their current form).

Some of your series- and message-ID-related comments remind me of this
report I made, which highlights another problem with the existing
fuzziness in Patchwork:

  https://lists.ozlabs.org/pipermail/patchwork/2014-May/001019.html

Anyway, thanks for the work so far. Maybe I'll set up my own Patchwork
instance and give some of your code a run.

Brian
___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 26/46] tests: Fix the submitter completion tests

2014-11-10 Thread Jeremy Kerr

Hi Damien,


Because we've changed the format of the json file we return to be a
"proper" object rather than the db dump django does by default,
there's no 'fields' sub-object anymore.


If you're doing a v2, could you roll this into the original change?

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 25/46] filters: Rewrite the submitter autocompletion code

2014-11-10 Thread Jeremy Kerr

Hi Damien,


We now have a nice(r) autocompletion widget that will popup the list of
options directly underneath the input fild.


woot!


A slight change in behaviour is that we now don't allow "free form"
submitter search, ie we need a valid completion to search with that
specific person (through its primary key). I didn't remove the backend
logic that allows the "free form" mechanism, but maybe we should.


Originally, I've included that for the case where we have no client-side 
javascript, and hence no completion. It looks like that'll still work 
with this change though - is that right?


Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 40/46] patch: Single out the commit message

2014-11-10 Thread Jeremy Kerr

Hi Damien,


All 'Comments' are stored the same way in the db, but I believe it's
worth making the distinction between introducing what the patch does and
eventual review comments.

Signed-off-by: Damien Lespiau 
---
  apps/patchwork/models.py   | 11 ++-
  templates/patchwork/patch.html | 18 +-
  2 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/apps/patchwork/models.py b/apps/patchwork/models.py
index 4bed9b8..b9c7794 100644
--- a/apps/patchwork/models.py
+++ b/apps/patchwork/models.py
@@ -188,7 +188,16 @@ class Patch(models.Model):
  return self.name

  def comments(self):
-return Comment.objects.filter(patch = self)
+comments = Comment.objects.filter(patch = self)
+# len() will trigger a queryset evaluation. That's what we want as
+# we need the full list of comments. That the list is then sliced into
+# (commit message, list of followup comments) won't hit the db twice.
+n_comments = len(comments)
+if n_comments > 2:
+return (comments[0], comments[1:])
+if n_comments == 1:
+return (comments[0], ())
+return (None, ())


This doesn't guarantee that you'll pick the correct comment for the 
commit message - you need to find the comment that has the same 
message-id as the patch.


Also, could we have this as a separate function? I'd like to 
Patch.comments() for the complete set as a single list. Your template 
might be a bit cleaner if have separate functions for 
commit-message-comments and follow-up-comments too.


Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 00/46] A re-styling of Patchwork

2014-11-10 Thread Jeremy Kerr

Hi Damien,


I'm part of the i915 driver team and we'd like to use patchwork. However, there
are a few things getting in the way: the bigger item is that patchwork doesn't
scale very well above a certain threshold of patches.

We have a number of ideas to improve the situation, mainly being able to
manipulate a higher level object, the "series", instead of "patches".


That's something that I'd definitely like to see. Being able to work 
with an entire series would make things a lot more manageable in this 
situation.




But I'm getting ahead of myself here. This is mostly a spare time project, and
this re-styling series is only a step on the road.

Belén, a designer working in my office on Yocto (www.yoctoproject.org), has
been kind enough to spend some of her spare time doing a pass on the patchwork
HTML/CSS and I've sprinkled a few ideas of my own as well. We've tried to
respect what was there and offer a slight refresh for the first pass. It's
possible to see the result at:

   http://patchwork.lespiau.name/


Wow, looks great!

Since you don't touch the models there, I'd like to set up a beta site, 
(which would use the same database as the regular site) and use your 
changes on that. Once we've got some feedback from core patchwork users, 
we can merge these to the main site.


So, I'll get the reworked A/R/T code whipped into shape, apply your 
regesign series on top of that, and we can go from there.


Thanks for the contributions!

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 1/2] settings: Add django.contrib.staticfiles to the project

2014-11-10 Thread Jeremy Kerr

Hi Damien,


This is needes to serve the admin static files.


Hm, not really. In production, the static files should be served 
directly through the web server, not through the django app. By all 
means, we should include instructions for configuring django to do so, 
but this shouldn't be in the default installation setup.


Same applies to the django toolbar - this shouldn't be in the default 
settings.py file, as this'll get likely be used as-is in a production 
environment.


Cheers,


Jeremy
___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 1/2] docs: Add a note about how to run tests

2014-11-10 Thread Jeremy Kerr

Hi Damien,


Given that hacking on patchwork is spaced in time,


You too huh? :)

Thanks, applied.


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH] jquery: Fix jqeury typo

2014-11-10 Thread Jeremy Kerr

Hi Damien,


@@ -1,7 +1,7 @@
  This directory contains the jQuery Javascript library, and jQuery
  Table Drag & Drop plugin. These are available from:

-  http://jqeury.com/
+  http://jquery.com/


That seems a pretty common typo for me. Thanks! Applied.


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH] docs: Add the Debian package name for virtualenv/mysql

2014-11-10 Thread Jeremy Kerr

Hi Damien,


Signed-off-by: Damien Lespiau 
---
  docs/HACKING | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)


Thanks, applied.

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH] parsemail: Honour existing PYTHONPATH

2014-11-10 Thread Jeremy Kerr

Hi Damien,


For setups where the patchwork dependencies are not part of the system
packages, one needs to indicate where to find them. parsemail.sh was
overriding PYTHONPATH without leaving us a chance to join in.


OK, makes sense. Applied.

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 1/2] pwclient: add output format option to list/search command

2014-11-10 Thread Jeremy Kerr

Hi Jani,


Make scripting easier by letting the user specify the output format. The
format string may contain tag references to fields, such as %{id} and
%{msgid}.


Looks good. I've applied both in this series.

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH] COPYING: Add the text of the GPLv2 license

2014-11-10 Thread Jeremy Kerr

Hi Damien,


I missed having that file when I first looked at the git repository to
figure out the license of the project. So add one.


Thanks! Applied.

Cheers,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Re: [PATCH 1/4] pwclient: add --help / -? to all sub commands

2014-11-10 Thread Jeremy Kerr

Hi Berhnard,

Thanks for the changes, all look good. I've applied them all.

Regards,


Jeremy

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork


Proxy transport support

2014-11-10 Thread Wodkowski, PawelX
Hi,
Is there a way to use pwclient from behind a proxy?

Paweł Wódkowski

___
Patchwork mailing list
Patchwork@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/patchwork