Honza Král wrote:
> I get a ProgrammingError, the SQL command is missing quotes around the
> value in question. I worked around this using str(
> form.clean()['form_field'] ), but that doesn't strike me as very nice.
As far as I remember psycopg2 does correctly accept unicode strings. You
can ju
On 11/30/06, Honza Král <[EMAIL PROTECTED]> wrote:
> great work on those forms, I really like them, thanks.
> I am currently using them for a few projects and am delighted with them.
>
> I am also facing some UNICODE issues, when saving into a database
> (postgres), the backend fails to quote the
I'm hoping to inspire some interest in this issue...
http://code.djangoproject.com/ticket/2288 (Provide an equivalent for
'ON DELETE' and friends)
Currently, there's no one way to fully specify how foreign keys are
handled by the database. The "ON UPDATE" and "ON DELETE" behaviors
must be modifi
(Tried submitting a ticket, blocked by Akismet.)
When creating users with User.objects.create_user, email is a required
parameter and fails when None is passed because create_user assumes it
will be a string by calling .strip.lower(). However, email is an
optional attribute on User. Passing the e
On 11/30/06, gabor <[EMAIL PROTECTED]> wrote:
> but if i want to render it in a template and it contains non-ascii text,
> it fails in /home/gabor/src/django/django/template/__init__.py,
> line 745, UnicodeEncodeError.
>
> the code there calls str() on the form, and it's a problem because it
> cal
Hi Adrian,
great work on those forms, I really like them, thanks.
I am currently using them for a few projects and am delighted with them.
I am also facing some UNICODE issues, when saving into a database
(postgres), the backend fails to quote the values from the form
correctly (field is a charfie
On 11/30/06, James <[EMAIL PROTECTED]> wrote:
>
>
> Brian, splendid stuff :) Seems to work so far, at least in terms of
> keeping database integrity and actually doing what it's supposed to.
> Thanks very much for taking the time to post this solution.
>
> Now how do I write a regression test? Any
Adrian Holovaty wrote:
> Ah, you're talking about the *old* form system...django.newforms has a
> way of dealing with this -- check out
> CheckboxSelectMultiple.value_from_datadict(), which is automatically
> called by Form.full_clean().
Ah... I was in fact just looking at HTML and just recalle
Adrian Holovaty wrote:
> I like the output,
> but I can definitely appreciate the argument for using s instead.
I would say not using tables for layout may count as a good practice
like encouraging clean URLs, proper distinction of GET and POST etc.
--~--~-~--~~~---
On 11/30/06, Adrian Holovaty <[EMAIL PROTECTED]> wrote:
> > Instead of this it can look like this (note the value attribute which
> > replaces default "on"):
> >
> >
> >John
> >Paul
> >
> >
> > and a POST would be exactly as with without the need
> > of `prepare`.
>
>
On 11/30/06, Ivan Sagalaev <[EMAIL PROTECTED]> wrote:
> Currently using checboxes it looks like:
>
>
>John
>Paul
>
>
> and a POST is:
>
> beatlesJ=on&beatlsP=on
>
> And then you should remeber to call manipulator.prepare(new_data) to
> convert these names into a li
Adrian Holovaty wrote:
> How would CheckboxSelectMultiple work otherwise, and how is the
> compatibility broken?
Consider :
John
Paul
If you select both items in a POST you'll get:
beatles=John&beatles=Paul
... which you will then nicely get as request.POST.getl
On 11/30/06, Ivan Sagalaev <[EMAIL PROTECTED]> wrote:
> A new Form class' default representation (that is goes out of `print f`)
> is an HTML table. For those who cares about that proverbial 'semantics'
> thing this looks wrong. From the practical point of view it's also not
> very convenient sinc
On 11/30/06, Ivan Sagalaev <[EMAIL PROTECTED]> wrote:
> CheckboxSelectMultiple is in effect just an alternative representation
> of . Ideally one would be able to just switch between
> these widget types not having to touch any other code. However
> CheckboxSelectMultiple uses different names for
Hi!
I'm now 'having a look' at newforms upon Adrian's request.
I noticed a thing I was long forgotten to file as a ticket. So may be
it's a good time to fix it.
CheckboxSelectMultiple is in effect just an alternative representation
of . Ideally one would be able to just switch between
these
Hi!
A new Form class' default representation (that is goes out of `print f`)
is an HTML table. For those who cares about that proverbial 'semantics'
thing this looks wrong. From the practical point of view it's also not
very convenient since it (for example) doesn't allow to layout the form
i
Adrian Holovaty wrote:
> Hi all,
>
> The django.newforms library is getting more solid by the day, and I'm
> using it for some personal and work projects. Real-world use has
> helped me find and fix problems with it, and I'm pretty happy with how
> things have turned out.
>
> So, then, I'd encou
Hi all,
The django.newforms library is getting more solid by the day, and I'm
using it for some personal and work projects. Real-world use has
helped me find and fix problems with it, and I'm pretty happy with how
things have turned out.
So, then, I'd encourage everybody to play around with djan
On 11/30/06, James Bennett <[EMAIL PROTECTED]> wrote:
>
> On 11/27/06, Joseph Kocherhans <[EMAIL PROTECTED]> wrote:
> > I'm not sure that this should be a function however. It may also make
> > sense as a Manager and/or QuerySet method.
>
> It *feels* like something that'd make sense as a manager
Waylan Limberg wrote:
> Tests or not, be sure to post the fixes to the ticket (and perhaps a
> link to this thread) so is doesn't get lost.
It is done.
--
James
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"D
[EMAIL PROTECTED] wrote:
> Are you really sure to fork a php interpreter instance for every single
> request that hits your web server?
> (In reference to
> http://firestats.cc/browser/trunk/firestats/integration/django/firestats.py)
Hi,
I am the FireStats developer.
While this is not ideal, it'
thanks for your reply Jacob.
Am 30.11.2006 um 15:27 schrieb Jacob Kaplan-Moss:
>
> On 11/30/06 6:51 AM, patrickk wrote:
>> we had this problems recently:
>> 1. it´s currently not clear which columns are searched when using the
>> search-field in the admin-list. that´s sometimes very strange,
>>
On 11/30/06 6:51 AM, patrickk wrote:
> we had this problems recently:
> 1. it´s currently not clear which columns are searched when using the
> search-field in the admin-list. that´s sometimes very strange,
> especially when using model-methods.
> 2. it´s not possible to have multiple search-f
On 11/30/06, James <[EMAIL PROTECTED]> wrote:
>
> Brian, splendid stuff :) Seems to work so far, at least in terms of
> keeping database integrity and actually doing what it's supposed to.
> Thanks very much for taking the time to post this solution.
Tests or not, be sure to post the fixes to the
Brian, splendid stuff :) Seems to work so far, at least in terms of
keeping database integrity and actually doing what it's supposed to.
Thanks very much for taking the time to post this solution.
Now how do I write a regression test? Any docs available?
--
James
--~--~-~--~~--
Are you really sure to fork a php interpreter instance for every single
request that hits your web server?
(In reference to
http://firestats.cc/browser/trunk/firestats/integration/django/firestats.py)
--~--~-~--~~~---~--~~
You received this message because you ar
I´ve just been looking at the admin-generator in symfony and it has
the possibility to add multiple search-filters.
we had this problems recently:
1. it´s currently not clear which columns are searched when using the
search-field in the admin-list. that´s sometimes very strange,
especially
That verbose_name attribute is very important if we think about
Internationalization.
Something like this:
agent_phone = DateField(verbose_name=_("Agent's phone number"))
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Gr
On 11/30/06, James <[EMAIL PROTECTED]> wrote:
>
> > Any help you can provide in tracking the problem (and/or fixing
> > it) would be greatfully accepted.
>
> I had a brief look at the code, and (despite my limited Python
> knowledge) was going to pitch in if I could. Given Brian's detailed
> respo
On 11/30/06, Brian Beck <[EMAIL PROTECTED]> wrote:
>
> Anyway, the key point is
> that anything using get_accessor_name() should take None into account,
> or None should never be returned.
The former approach would be correct. get_accessor_name() shouldn't
return a value if the m2m relation is sy
> Any help you can provide in tracking the problem (and/or fixing
> it) would be greatfully accepted.
I had a brief look at the code, and (despite my limited Python
knowledge) was going to pitch in if I could. Given Brian's detailed
response, I shall probably apply the fixes he suggests and spend
31 matches
Mail list logo
