On Sun, 2007-09-23 at 12:50 +0800, Russell Keith-Magee wrote:
[..]
> Another approach would be to only send the failure message to the
> committer that stimulated the build. i.e., you break it, you get a
> nastygram in your inbox :-)
Whilst well intentioned, this tends not to work for community-o
On 9/23/07, Joseph Heck <[EMAIL PROTECTED]> wrote:
>
> I'm also a little concerned that the noise level would be high for
> those messages - that ultimately people would ignore them because of
> frequency.
The current state of the tree is a little misleading; Post-sprint, we
have had a few test f
On Sat, 2007-09-22 at 04:31 -0700, Andrew Durdin wrote:
[...]
> I'm confident now that the patch is correct, as I now understand why
> it works:
>
> string_concat returns a __proxy__ that will call _string_concat when
> __unicode__ is called, so we expect _string_concat to return a unicode
> inst
During the sprint, Malcolm informed me about #2438, and that it was
the reason some models were disappearing from the admin when using
dbsettings. I've spent some time trying to figure it out, since I had
a setup that was exhibiting the problem, but so far I haven't come up
with anything.
So, in
On 9/22/07, Tristan King <[EMAIL PROTECTED]> wrote:
> A bit more progress today.
>
> I think i've solved the __unicode__ problems which leo has been solving
> with django workaround (i think it's the same problem anyways),
> internally to jython.
>
> (see obj-unicode.patch)
[More patches to Jytho
On 9/21/07, Brian Harring <[EMAIL PROTECTED]> wrote:
> The only spot I can see this potentially causing issues is for
> the 'literal' fast pathing, since it just reuses what was passed in-
> forcing that to unicode in __init__ ought to resolve that. If it's
> something other then that, definitely
We can certainly enable something like that, but I don't think we're
stable enough in our configuration to make that worthwhile at the
moment.
I'm also a little concerned that the noise level would be high for
those messages - that ultimately people would ignore them because of
frequency.
Maybe
On 9/21/07, Malcolm Tredinnick <[EMAIL PROTECTED]> wrote:
> Periodically checking the buildbot output
> might be useful for when we don't have time to run every combination of
> backend and Python version locally.
Perhaps the buildbot could send build failures to django-updates; I've
even seen se
On 9/22/07, Russell Keith-Magee <[EMAIL PROTECTED]> wrote:
> Yes, there _could_ be a clash of commands, and a user application
> _could_ override a default Django command. However, I put it to you
> that this is the 1% case, not the 99% case. Most applications won't be
> registering commands, so t
> a new command is registered. If a clash is identified, we can either:
> 1) Raise an error, stopping manage.py
> 2) Raise a warning, and
> a) adopt the new command
> b) ignore the new command
> 3) Register the duplicate command with an auto-aliased name - the
> appname.command syntax you prop
> The recommended approach for something like this is to first create it
> as a third-party app so that people can try it out and we can harvest
> some real-world experiences. Then it can be migrated to django/contrib,
> if necessary. So, if no core changes are necessary (and I can't see that
> t
On 9/21/07, Brian Harring <[EMAIL PROTECTED]> wrote:
> The only spot I can see this potentially causing issues is for
> the 'literal' fast pathing, since it just reuses what was passed in-
> forcing that to unicode in __init__ ought to resolve that. If it's
> something other then that, definitely
On 9/22/07, Martin Winkler <[EMAIL PROTECTED]> wrote:
>
> Sorry, please ignore my previous post. I did not understand what you
> really meant. :-/
>
> SmileyChris wrote:
> >> Common use is now href="{{ MEDIA_URL }}images/a.png"
> >> so it seems to me like you should *always* put a trailing slash i
I'm now tuning my project at work and resolve_variable shows up
alarmingly high in profiled data. I'm thus very eager to look at this
fix and will try it this Monday or Tuesday and will try to fix any
unicode problems.
> Examples would be lovely; that particularly patch proceeded the
> unicode me
On 9/22/07, Todd O'Bryan <[EMAIL PROTECTED]> wrote:
>
> On Sat, 2007-09-22 at 11:52 +0800, Russell Keith-Magee wrote:
> > Adding user-space commands is something that only a subset of
> > applications will need to do. As a result, there is even less
> > potential for clashes here than there is in
On Sep 22, 2:16 am, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
>
> Thanks, Andrew. That patch looks "intuitively" correct; it's in the
> right area of code that I would have expected to find the problem, etc.
> I'll have a think about it through today to see if there are any
> technical problem
Yuri Baburov wrote:
> I can't subclass ModelAdmin, because:
>
> class ModelAdmin(BaseModelAdmin):
> def __init__(self, model, admin_site):
> self.model = model
> self.opts = model._meta
> ...
>
> I have no model and have no _meta, have no QuerySet subclass (I have
> v
On Sat, 2007-09-22 at 11:52 +0800, Russell Keith-Magee wrote:
> On 9/22/07, Todd O'Bryan <[EMAIL PROTECTED]> wrote:
> >
> > I have a couple of reservations about the version of app commands that
> > got committed and just wanted to bring them up here.
> ...
> > I'm worried that we're setting ourse
Sorry, please ignore my previous post. I did not understand what you
really meant. :-/
SmileyChris wrote:
>> Common use is now href="{{ MEDIA_URL }}images/a.png"
>> so it seems to me like you should *always* put a trailing slash in
I'm very much +1 on this too. It makes a lot of sense to always
A bit more progress today.
I think i've solved the __unicode__ problems which leo has been solving
with django workaround (i think it's the same problem anyways),
internally to jython.
(see obj-unicode.patch)
the changes to the __unicode__ function, check if the object has a
__unicode__ function
Am Fri, 21 Sep 2007 20:25:09 -
schrieb SmileyChris <[EMAIL PROTECTED]>:
>
> Currently the docs [1] say "Note that this should have a trailing
> slash if it has a path component."
Since Django uses urlparse.urljoin for concatenating, you can omit the
trailing slash, if - and only if - the M
21 matches
Mail list logo
