On 4/4/06, Luke Plant <[EMAIL PROTECTED]> wrote:
> OK, so I tried this out on a local Apache2 instance and got exactly the
> same thing. I found an oddity in the source code, and simplifying it
> (see below) fixed it for me in both instances, though I don't really
> know why.
>
> I'm nervous abou
On Tuesday 04 April 2006 21:52, Luke Plant wrote:
> I have an app that is live and running m-r (yeah, I know!), and I
> just got a "connection limit exceeded for non-superusers" error. It
> appears that connections are not being closed at all. With every hit
> I'm getting another entry in the pr
For me the same, hitting refresh a couple of times creates more and
more of those '[local] idle in transaction' things.
Rudolph
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Django developers" group.
To post to
I have an app that is live and running m-r (yeah, I know!), and I just
got a "connection limit exceeded for non-superusers" error. It appears
that connections are not being closed at all. With every hit I'm
getting another entry in the process table, mainly of this form:
$ps -ef
postgres 20
Hi Eugene,
thanks for the pointer. I suspected it's not been the first
discussion--but it's hard to find. Now, I'll need some time to digest ;-)
Michael
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Django deve
Michael Radziej wrote:
>
> You're right. We should go from settings.ROOT_URLCONF downward, anyway,
> and not take any url module as parameter, which wouldn't work out in
> reality.
>
> I'm not saying this is the solution. But I think this is the way to go.
>
> Anybody else interested in this
On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> I just realized that both the calendar and clock widgets were within a
> fieldset with 'classes': 'collapse'
http://files.shtuff.us/admin-dojo.tgz
Fixed now :)
--
"May the forces of evil become confused on the way to your house."
-- George Ca
On 4/4/06, James Bennett <[EMAIL PROTECTED]> wrote:
> On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> > I just realized that both the calendar and clock widgets were within a
> > fieldset with 'classes': 'collapse'
OK, now I'm able to replicate it, and I've found the source of the
problem but
On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
>
> James Bennett schrieb:
> > Then why did the original email say that it would "tie the urls to the
> > field names in the model"?
>
> Let's suppose you have an url config like this (current syntax, urls.py):
>
>
> urlpatterns = patterns('mai
James Bennett schrieb:
> Then why did the original email say that it would "tie the urls to the
> field names in the model"?
Let's suppose you have an url config like this (current syntax, urls.py):
urlpatterns = patterns('mailadmin.views',
(r'^(?P[1-9]\d*), 'edit_foobar')),
# ...
)
Hi,
I'm sorry if my reply was a little bit too short.
The bottom line of my proposal is to tie get_absolute_url() to the url
configuration in urls.py. This will need a little twisting of urls.py,
but I hope to do it in a quite and unobtrusive way. It should still be
overwriteable; I'm aware t
On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
> Have you read through the proposal? I don't tie the URL structure to the
> internals of of the application's code. Quite to the opposite.
Then why did the original email say that it would "tie the urls to the
field names in the model"?
--
"
James Bennett schrieb:
> Tying the URL structure to the internals of the application's code,
> which is what you seem to be saying your proposal would do, limits a
> developer's flexibility to choose the URLs that are most appropriate
> to the actual content, and often forces the use of URLs which
On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> I just realized that both the calendar and clock widgets were within a
> fieldset with 'classes': 'collapse'
I've got a test app running with date/time fields inside a collapsed
fieldset, and it's rendering just fine in all the browsers I can thr
On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
> I feared this might turn out a problem. I haven't so much experience
> with different projecs, can you explain a litte about the problems with
> this?
Tying the URL structure to the internals of the application's code,
which is what you seem
limodou schrieb:
> I think is not very simple. And I don't see that how to resolve apps
> url prefix. Or you just want to resolve get_absolue_url.
You're right. We should go from settings.ROOT_URLCONF downward, anyway,
and not take any url module as parameter, which wouldn't work out in
reality
On Apr 3, 2006, at 7:28 PM, Malcolm Tredinnick wrote:
> Jacob: is there something wrong with this, or is it just unloved?
Looks like Russell beat me to it :)
Sorry I've been a bit on the flaky side lately: I just bought a house
(!) so as you can imagine I'm a bit distracted :)
Jacob
--~--~--
On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
>
> James Bennett schrieb:
> > On 4/4/06, limodou <[EMAIL PROTECTED]> wrote:
> >> [...]
> > You've taken this completely out of context; as I understand it, Jacob
> > and Adrian were not saying that the *entire* URL system should be
> > replac
> On 4/4/06, James Bennett <[EMAIL PROTECTED]> wrote:
>
> On 4/4/06, limodou <[EMAIL PROTECTED]> wrote:
> > And there are many people except me consider that: django need a good
> > urls system:
>
> You've taken this completely out of context; as I understand it, Jacob
> and Adrian were not saying
Some after-thoughts:
Michael Radziej schrieb:
> The problem is, urls.py knows all about the prefixes where all the model
^^^ read: point
> If you want to use this stuff somewhere in a view or model, do:
>
> {{{
> from yourapp import urls
> from django.jabberwocky import get_absolute
James Bennett schrieb:
> On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
>
>>* I want to do url configuration in exactly one place
>
>
> One place for each project? One place for each application? One place
> for each model? The problem is that each of these has perfectly valid
> use case
On 4/4/06, Michael Radziej <[EMAIL PROTECTED]> wrote:
> * I want to do url configuration in exactly one place
One place for each project? One place for each application? One place
for each model? The problem is that each of these has perfectly valid
use cases for wanting to specify its own URL st
James Bennett schrieb:
> On 4/4/06, limodou <[EMAIL PROTECTED]> wrote:
>> [...]
> You've taken this completely out of context; as I understand it, Jacob
> and Adrian were not saying that the *entire* URL system should be
> replaced, only that the 'get_absolute_url' method for models doesn't
> fee
On 4/4/06, Malcolm Tredinnick <[EMAIL PROTECTED]> wrote:
>
> Jacob: is there something wrong with this, or is it just unloved?
I can't speak for Jacob, but I'm pretty sure its just been lost in the
flood. I know it is frustrating, but quite often the rate of patch
production exceeds the rate at w
On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> I just realized that both the calendar and clock widgets were within a
> fieldset with 'classes': 'collapse'
Hmm. I'll try that tomorrow and see if I can replicate it and find out
what causes it.
--
"May the forces of evil become confused on the
Caches cleared, same problem in Firefox on both Debian Linux and
Windows.
I just realized that both the calendar and clock widgets were within a
fieldset with 'classes': 'collapse'
I removed the classes and the problem went away.
Note that it works fine in the non-Dojo admin with 'classes':
'co
On 4/4/06, limodou <[EMAIL PROTECTED]> wrote:
> And there are many people except me consider that: django need a good
> urls system:
You've taken this completely out of context; as I understand it, Jacob
and Adrian were not saying that the *entire* URL system should be
replaced, only that the 'ge
On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> Tested on Debian Sarge (Firefox from backports.org) and Windows XP.
That's extremely weird... I just set up a fresh copy and tested on FF
1.5 on Ubuntu, and it looks fine. It also was fine on FF 1.5 on a Mac
at work. Could you completely clear yo
Firefox 1.5
Tested on Debian Sarge (Firefox from backports.org) and Windows XP.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Django developers" group.
To post to this group, send email to django-developers@goog
On 4/4/06, Bryan Chow <[EMAIL PROTECTED]> wrote:
> One issue I noticed is that the calendar and clock widgets do not
> render properly on Firefox -- they're positioned above the top of the
> page instead of beside the field.
What version of Firefox are you using? I'm unable to replicate that
with
On 4/4/06, binaryfeed <[EMAIL PROTECTED]> wrote:
>
> This has been solved in Aquarium. http://aquarium.sf.net/
>
But Aquarium is not Django, it's another web framework.
--
I like python!
My Blog: http://www.donews.net/limodou
My Django Site: http://www.djangocn.org
NewEdit Maillist: http://grou
Very cool!
I tried installing the Dojo integration package into magic-removal from
svn (latest revision = 2607). First impressions are that generally
things seem to work without issue, although admittedly I haven't spent
much time looking through the code yet.
One issue I noticed is that the cal
This has been solved in Aquarium. http://aquarium.sf.net/
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Django developers" group.
To post to this group, send email to django-developers@googlegroups.com
To unsubs
33 matches
Mail list logo
