ANN: Django 0.90 is available

2005-11-15 Thread Jacob Kaplan-Moss
Hey folks -- We've wrapped up a 0.90 release of Django; let the Subversion-less rejoice! http://www.djangoproject.com/weblog/2005/nov/16/ firstrelease/ has the full details, but suffice to say that we've now got a tarball (and and egg for those who are into that sort of thing). Enjoy! J

"startapp" now creates views.py, not views package

2005-11-15 Thread Adrian Holovaty
Earlier today I made a small change to the output of "django-admin.py startapp" -- it creates a file "views.py" instead of a directory "views". This is for simplicity's sake. It means fewer files called (in the case of the tutorial example) "polls.py", and a lighter directory structure in general.

Re: 'block' tag with name 'userlinks' appears more than once

2005-11-15 Thread Bryan Murdock
On 11/15/05, Bryan Murdock <[EMAIL PROTECTED]> wrote: > I just did an svn up and now I get this: > > 'block' tag with name 'userlinks' appears more than once. Um, that's when I go to admin. Shoulda said that the first time, sorry. > > I've tried to follow all the backwards incompatible changes.

'block' tag with name 'userlinks' appears more than once

2005-11-15 Thread Bryan Murdock
I just did an svn up and now I get this: 'block' tag with name 'userlinks' appears more than once. I've tried to follow all the backwards incompatible changes. Any ideas? Bryan

Re: Broken site since svn update

2005-11-15 Thread Jakub Labath
This may sound silly, but have you tried renaming that property in your model? Or better yet using db_column parameter. Would that help you? DMI_id = meta.SlugField(db_index = True, unique = True, prepopulate_from=("title",),db_column="dmi_id") Wild guess: maybe escaping functions made somethin

Re: Broken site since svn update

2005-11-15 Thread Adrian Holovaty
On 11/15/05, Laurent Rahuel <[EMAIL PROTECTED]> wrote: > Since the beginning, one of my models (folder) contains a property defined > like this : > > DMI_id = meta.SlugField(db_index = True, unique = True, > prepopulate_from=("title",)) > > Note the uppercase DMI. > > My app is now crashing compla

Re: Broken site since svn update

2005-11-15 Thread Laurent Rahuel
Yes and I did not find anything relating any upper/lower case issue. Le Mercredi 16 Novembre 2005 01:00, Eugene Lazutkin a écrit : > Did you check if something from "Backwards-incompatible changes" list > (http://code.djangoproject.com/wiki/BackwardsIncompatibleChanges) applies > to you? > > > "L

Re: Broken site since svn update

2005-11-15 Thread Eugene Lazutkin
Did you check if something from "Backwards-incompatible changes" list (http://code.djangoproject.com/wiki/BackwardsIncompatibleChanges) applies to you? "Laurent Rahuel" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] > > Hi, > > I was using a "rather" old version of django svn wit

Re: DRY templates

2005-11-15 Thread oggie rob
Holy moly! That's exactly it! Thanks for the change - I had been looking for a while and my head was starting to swim!

Broken site since svn update

2005-11-15 Thread Laurent Rahuel
Hi, I was using a "rather" old version of django svn with no problems. I'm using Postgresql as backend. I updated sources a now I got a strange problem. Since the beginning, one of my models (folder) contains a property defined like this : DMI_id = meta.SlugField(db_index = True, unique = Tru

Re: DRY templates

2005-11-15 Thread Jacob Kaplan-Moss
On Nov 15, 2005, at 5:09 PM, oggie rob wrote: But when I get to the template, no scalable option is obvious. I would expect something like: (% for field in form.fields %) [snip] As of revision [1253] this is now possible. Enjoy! Jacob

DRY templates

2005-11-15 Thread oggie rob
Hi, There is one part of the design that puzzles me. Considering DRY and simple design goals, it is odd that you have to define a very specific template to cover even a generic view. For example, I have a Client(Person) model (Person is a subclass as it holds general info like names, addresses, et

Re: understanding regroup

2005-11-15 Thread Sean Perry
Robert Wittams wrote: Would: {% regroup user.get_thing_list|dictsort:"category_id" by category_id as grouped %} {% for category in grouped %} {{ category.list.0.get_category.description }} {% for entry in category.list %} {% endfor %}

Re: understanding regroup

2005-11-15 Thread Robert Wittams
Sean Perry wrote: > > I have the following classes in my model: > > class Category: > description = charField() > > class Thing: > self.description = charField() > self.category(ForeignKey, Category) > > I would like a page layed out like: > > Category1 > * foo > * bar > * baz >

Re: understanding regroup

2005-11-15 Thread Sean Perry
Jacob Kaplan-Moss wrote: On Nov 15, 2005, at 2:15 AM, Sean Perry wrote: The {% regroup %} tag looked promising, but I can not get it to work. {% regroup user.get_thing_list|dictsort:"category" by category as grouped %} You have to group by actual fields (and not relationship names),

Re: understanding regroup

2005-11-15 Thread Jacob Kaplan-Moss
On Nov 15, 2005, at 2:15 AM, Sean Perry wrote: The {% regroup %} tag looked promising, but I can not get it to work. {% regroup user.get_thing_list|dictsort:"category" by category as grouped %} You have to group by actual fields (and not relationship names), so to Things by Category yo

Re: Error With /media In URL?

2005-11-15 Thread JA
In case this helps, entering /media in the url produces the error everytime, even though /media is not in any of my url files. Thanks J

Re: Creating new tables

2005-11-15 Thread Kenneth Gonsalves
On Tuesday 15 Nov 2005 2:42 pm, Jeremy Dunck wrote: > It sounds to me like you're complaining that a relational database > isn't the right tool for your job. thats what i was trying to say -- regards kg http://www.livejournal.com/users/lawgon tally ho! http://avsap.org.in ಇಂಡ್ಲಿನಕ್ಸ வாழ்க!

Re: Creating new tables

2005-11-15 Thread Jeremy Dunck
On 11/15/05, panos <[EMAIL PROTECTED]> wrote: > > Well I have thought about that but there are some issues. > First of all the table "logdetails" will become huge. Imagine logging > values of 5000 variables > every 15minutes! > Secondly not all variables are of the same time. They can be boolean,

Re: Creating new tables

2005-11-15 Thread panos
Well I have thought about that but there are some issues. First of all the table "logdetails" will become huge. Imagine logging values of 5000 variables every 15minutes! Secondly not all variables are of the same time. They can be boolean, int and real. If you have just one table for all of them t

OneToOne relations broken (#527), any workarounds ?

2005-11-15 Thread plisk
Hi, i'm trying to use OneToOne relation on new-admin branch and as of rev 1238 i'm getting this error when updating UserDetail entry in auto-admin Request Method: POST Request URL:http://192.168.1.156:8000/admin/intranet/userdetails/1/ Exception Type: TypeError Exception Value

Re: Creating new tables

2005-11-15 Thread Jeremy Dunck
On 11/15/05, panos <[EMAIL PROTECTED]> wrote: > > I don't know if it is strange or not, but it came to me as a natural > choice for my application. > Think of a logging application where some variables need to be logged > and we need a table for > each variable holding a great number of "timestamp

Re: Creating new tables

2005-11-15 Thread Kenneth Gonsalves
On Tuesday 15 Nov 2005 1:39 pm, panos wrote: > I don't know if it is strange or not, but it came to me as a natural > choice for my application. > Think of a logging application where some variables need to be logged > and we need a table for > each variable holding a great number of "timestamp -

understanding regroup

2005-11-15 Thread Sean Perry
I have the following classes in my model: class Category: description = charField() class Thing: self.description = charField() self.category(ForeignKey, Category) I would like a page layed out like: Category1 * foo * bar * baz CategoryX * woo * war * waz etc. The {% regr

Re: Creating new tables

2005-11-15 Thread panos
I don't know if it is strange or not, but it came to me as a natural choice for my application. Think of a logging application where some variables need to be logged and we need a table for each variable holding a great number of "timestamp - value" entries. The thing is we don't know beforehand w