Re: Reloading only one Python interpreter with Apache

2007-11-03 Thread Kenneth Gonsalves


On 03-Nov-07, at 8:57 PM, Sander Dijkhuis wrote:

> I often have to reload one project because I've updated the code.
> Currently, I use `/etc/init.d/apache2 reload`

why dont you do apache2ctl graceful?

-- 

regards
kg
http://lawgon.livejournal.com
http://nrcfosshelpline.in/web/



--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: IOError: Client read error (Timeout?)

2007-11-03 Thread Graham Dumpleton

FWIW, the Python 2.5 source code violates its own documentation if
this is in some way meant to define a standard. That is, in zip and
bz2 modules, they raise IOError's with only a string value and no
errno.

Graham

On Nov 4, 9:32 am, Graham Dumpleton <[EMAIL PROTECTED]>
wrote:
> FWIW, it has been noted as an issue with mod_python that using IOError
> directly isn't good, although for slightly other reasons.
>
>  https://issues.apache.org/jira/browse/MODPYTHON-92
>
> That it isn't using IOError correctly in as much as no errno is
> present adds to why  it shouldn't do as it is. It isn't just a simple
> matter of supplying errno though as in certain situations there
> wouldn't actually be any.
>
> As far as ignoring these errors, which occur when remote client closes
> connection, it isn't necessarily a good idea for the lower level
> layers to do so, as there are possible implications in doing so. For
> example, it can hide problems in your own application and make it
> harder to debug. A similar issue was brought up in relation to
> mod_wsgi:
>
>  http://code.google.com/p/modwsgi/issues/detail?id=29
>
> Your observation about errno means that mod_wsgi should also change
> what error exception type it is using as well.
>
> Graham
>
> On Nov 2, 8:31 pm, Bjørn Stabell <[EMAIL PROTECTED]> wrote:
>
> > On Oct 6, 11:20 pm, Trey <[EMAIL PROTECTED]> wrote:
>
> > > There are other people that have brought this up a little bit some
> > > time ago. I run a small to medium sized web application that takes
> > > profile pictures. By far my largest customer service issue is people
> > > not being able to upload their photos.
>
> > > For the most part I have played it down as their connection sucking or
> > > perhaps doing something stupid with the browser, but there are a
> > > couple of things that I am running into that are causing an issue.
>
> > > 1. I can't replicate this, no matter what I do with my browser in the
> > > middle of an upload.
> > > 2. Judging by the django code near the problem, this is working on
> > > information that has already been received.
> > > 3. I get this a few times a day at least, different people every time.
> > [...]
> > >   File "/usr/lib/python2.5/site-packages/django/core/handlers/
> > > modpython.py", line 120, in _get_raw_post_data
> > > self._raw_post_data = self._req.read()
>
> > > IOError: Client read error (Timeout?)
>
> > We're seeing this too, on several Django sites.  Annoyingly, as well,
> > the IOError exception itself is broken; if you look at the docs,
> > IOErrors (which are a form of EnvironmentErrors) should have a two- or
> > three-tuple .args, one of which would be the errno, but the ones
> > thrown bymod_python/Django seems to have only one item in the tuple;
> > the string you see above.
>
> > It's particularly annoying since we'd like to treat this as an info/
> > debug-level error, not an error-level error, and we could if we just
> > had access to the errno.  (We're trying to silence non-errors so we
> > can do proper monitoring.)
>
> > Rgds,
> > Bjorn


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: djangoAMF authentication, requests aren't carrying user credential

2007-11-03 Thread Graham Dumpleton

More likely then to be an application issue than something with
mod_wsgi.

Beyond wrapping the WSGI entry point for Django and dumping out
request headers and response headers and confirming that any cookie
stuff is correct, not sure what to suggest.

Graham

On Nov 3, 11:23 am, "Tiger Uppercut" <[EMAIL PROTECTED]> wrote:
> > Is the login mechanism relying on web application form based
> > submission, or relying on web server HTTP Basic authentication. If the
> > latter, you must enable the directive:
>
> >   WSGIPassAuthorization On
>
> > in Apache configuration for mod_wsgi, else the HTTP Basic auth
> > information isn't passed through to a WSGI application.
>
> > Graham
>
> The login mechanism is web application form based submission, done in flex,
> which is embedded in a django page, which is served by mod_wsgi.
>
> That is, I have a flex UI submitting username, raw_password through an AMF
> gateway.
>
> == myflex.mxml ==
>
> private function handleLogin(username,raw_password):void
> {
>   gatewayUrl = "https://www.mysite.com/main/gateway;;
>   var serviceName:String = "authService";
>   var serviceFactory:ServiceFactory = ServiceFactory.getInstance
> (gatewayUrl);
>   var service:RemotingService = serviceFactory.getService(serviceName);
>   var pc:PendingCall = service.authenticateVisitor(your_login,
> raw_password); // ties to django views.py method
>
> }
>
> == views.py ==
>
> def authenticateVisitor(request, this_email, raw_password):
> if request.user.is_authenticated() == True
> return "Already authenticated" # <--- This return statement is never
> reached, even if this method is called twice
> this_user = auth.authenticate(email=this_email,password=raw_password) #
> using custom EmailBackend
> if this_user is not None:
> auth.login(request,this_user)
> return True
> else:
> return False
>
> -R
>
> On 11/1/07, Graham Dumpleton <[EMAIL PROTECTED]> wrote:
>
>
>
> > On Oct 30, 9:22 pm, "Tiger Uppercut" <[EMAIL PROTECTED]> wrote:
> > > Hi,
>
> > > I wasn't sure whether this was a djangoAMF issue or django, so apologies
> > for
> > > the double posting.
>
> > > I'm having trouble using djangoAMF to communicate between my flex UI and
> > my
> > > django site, django_wrapper.
>
> > > - My django_wrapper is served at /main by mod_wsgi.
>
> > > * The django code base is in /usr/local/deploy/django_wrapper
> > > * The wsgi script is in
> > > /usr/local/deploy/django_wrapper/apache/django_wrapper.wsgi [as noted
> > onhttp://code.google.com/p/modwsgi/wiki/IntegrationWithDjango]
>
> > > - My flex site is served at /main/flex by django.  The swf is stored at
> > > /var/www/mysite.com/static/myflex.swf
>
> > > - My djangoAMF gateway is set up at /main/gateway
>
> > > == urls.py ==
>
> > >  (r'^main/gateway/authService/authenticateVisitor',
>
> > > > 'django_wrapper.views.authAndLoginVisitor'),
> > > >  (r'^main/gateway/calcService/calcSum',
> > > > 'django_wrapper.views.calculate'),
>
> > > == views.py ==
>
> > > def authAndLoginVisitor(request,this_email,raw_password):
>
> > > > if request.user.is_authenticated() == True:
> > > > return "Already authenticated!"
>
> > > > this_user = auth.authenticate (email=this_email,
> > > > password=raw_password) # modified from django.contrib.auth to use
> > > > EmailBackend
> > > > if this_user is None: # could not find the
> > > > user
>
> > > > return False
> > > > else: # did find the user, now let's log him
> > > > in
>
> > > > auth.login(request,this_user)
> > > > return True
>
> > > > def calculate(request,arg1,arg2):
> > > > if request.user is not None:
> > > > print "In calculate, request.user is not none, it is = " +
> > str(
> > > > request.user)
> > > > if request.user.is_authenticated() == True:
> > > > return arg1 + arg2
> > > > else:
> > > > return "Not authenticated!"
>
> > > The calculate function is called by the flex wrapper after
> > > authAndLoginVisitor is successfully called.
>
> > > == mysite.mxml ==
>
> > > var gatewayUrl:String = "https://www.mysite.com/main/gateway/;;
>
> > > > var serviceName:String = "calcService";
> > > > var serviceFactory:ServiceFactory = ServiceFactory.getInstance
> > > > (gatewayUrl);
> > > > var service:RemotingService = serviceFactory.getService(serviceName);
>
> > > > var pc:PendingCall = service.calcSum(2, 3);
> > > > pc.responder = new Responder(handleResult, handleError);
>
> > > However, the calculate function keeps thinking that the request is
> > issued by
> > > an anonymous user.
>
> > > I added debugging code to the login() method of  auth/__init__.py
>
> > > [Tue Oct 30 01:32:31 2007] [error] User status just set
> > > [Tue Oct 30 01:32:31 2007] [error] request.session[_auth_user_id] = 11
> > > [Tue Oct 30 01:32:31 2007] [error] request.session[_auth_user_backend] =
> > > django_wrapper.backends.EmailBackend
> > > [Tue Oct 30 01:32:31 2007] [error] 

Re: Large streaming uploads (#2070) doesn't work for me at all

2007-11-03 Thread Graham Dumpleton

You might at least try mod_python 3.3.1 instead of the older
mod_python 3.2.10. The newer version fixes a lot of problems including
memory leaks.

Graham

On Nov 4, 10:48 am, Eugene Morozov <[EMAIL PROTECTED]> wrote:
> Hello,
> I've applied patch from #2070 and I have disastrous result. I'm trying
> to upload a 12Mb PNG file. Apache starts eating memory rapidly,
> consuming several Gb in seconds and I have to kill it (if I have
> enough time to log in as root and find pid of the httpd process).
>
> I'm using Django trunk (rev. 6638, had to patch fields.py manually as
> patch cannot find context, it was easy, though), python 2.5.1, mod-
> python 3.2.10, apache 2.2.3 on 64 bit Ubuntu 7.04.
>
> Does this patch work for anyone with similar configuration?
>
> 12Mb is small enough to fit in memory, why apache consumes 2Gb while
> storing it and still fails... Mystery.
> Eugene


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Reloading only one Python interpreter with Apache

2007-11-03 Thread Graham Dumpleton

On Nov 4, 2:27 am, Sander Dijkhuis <[EMAIL PROTECTED]> wrote:
> On our Apache server, we're hosting multiple Django projects. The
> Apache configuration is like this:
>
>   PythonAutoReload Off
>   
>   ServerName first.example.com
>   
>   SetHandler python-program
>   SetEnv DJANGO_SETTINGS_MODULE first.settings
>   PythonInterpreter first
>   
>   
>   
>   ServerName second.example.com
>   
>   SetHandler python-program
>   SetEnv DJANGO_SETTINGS_MODULE second.settings
>   PythonInterpreter second
>   
>   
>
> I often have to reload one project because I've updated the code.
> Currently, I use `/etc/init.d/apache2 reload`, but that causes all of
> our projects to be unavailable for some seconds. Is there a way to
> 'kill' one of the Python interpreters, so that only one specific
> project is reloaded?

No, there is no way with mod_python of doing what you want.

With mod_wsgi 2.0 it is however easy to do what you want provided that
you use daemon mode of mod_wsgi to delegate each virtualhost and/or
application to a distinct daemon process group.

The directive which would need to be set it WSGIReloadMechanism with
the option of 'Process'. For details on mod_wsgi 2.0, which first
release candidate is out for now, see:

  http://code.google.com/p/modwsgi/wiki/ChangesInVersion0200

Graham


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: to_python ignored with custom IntegerField

2007-11-03 Thread Malcolm Tredinnick

On Sun, 2007-11-04 at 04:08 +0100, Stefan Foulis[spiderware gmbh] wrote:
> Hi
> I've tried to create a custom Field for the database to store  
> Timedelta objects. Since I only need exactness to the minute and won't  
> have any large timedeltas I'm using a simple Integer to store it in  
> the database. Maybe I'll make this configurable later.
> 
> this is the code:
> 
> from django.db.models.fields import IntegerField, Field
> from django.core import validators
> import datetime
> from contrib.numeric_utilities import timedelta2minutes
> 
> class TimedeltaField(IntegerField):
> def to_python(self, value):
> # turns the integer saved in the database to a python  
> timedelta object
> if value is None:
> return value
> if isinstance(value, datetime.timedelta):
> return value
> validators.isInteger(str(value), None)
> try:
> return datetime.timedelta(minutes=int(value))
> except (TypeError, ValueError):
> raise validators.ValidationError, u"This value must be an  
> integer (it gets converted to a Timedelta)"
> def get_db_prep_save(self, value):
> # Turns the input into a integer of minutes for save in db
> if isinstance(value, datetime.timedelta):
> value = timedelta2minutes(int(value))
> return Field.get_db_prep_save(self, value)
> def get_internal_type(self):
> return "IntegerField"
> =
> timedelta2minutes is a simple function that returns an integer.
> 
> 
> On the commandline:
> =
>  >>> from mytest.models import *;
>  >>> t = MyTest.objects.all()[0]
>  >>> t.a_timedelta
> 2342
>  >>> type(t.a_timedelta)
> 
>  >>> t.validate()
> {}
>  >>> t.a_timedelta
> datetime.timedelta(0, 39960)
>  >>> type(t.a_timedelta)
> 
> ==
> 
> So... until I run validate() 

Firstly, don't rely on validate() at the moment. It doesn't fully work
(which is why, for example, it's not documented or advertised at all).

> at least once the field returns the int  
> from the database instead of the converted value from to_python(). I  
> ran some tests with print statements in the to_python methods... they  
> never get called until validate() is run. The same for the built-in  
> DateField, but there the value still gets magically returned as a  
> datetime.date object.

What is stored in the model attributes is exactly what the database
gives us back, at the moment. There's nothing magical about datetime
objects -- all the database wrappers convert datetime columns into
Python datetime objects.

What needs to be changed for field subclassing -- and I keep meaning to
finish this work, along with my 5 other highest priorities -- is that
for custom fields (only!) we install attributes as part of contribute to
class so that, via Python's descriptor protocol (__set__ and __get__)
conversion automatically happens.

You can actually do that manually now, it's just not as easy as it could
be. To see how to do it now, have a look at the Django OSCON 2007 slides
(see [1]). Jeremy Dunck presented some stuff there about field
subclassing (starting at slide 107) that includes discussion of the
descriptor protocol.

[1] http://toys.jacobian.org/presentations/2007/oscon/tutorial/

All we'll add to Django to make that stuff easier is probably a
metaclass that you can use so that you only need to implement
to_python() and a few other things and it will automatically do the
contribute_to_class() portion and set up __set__() and __get__() for
you. So no extra functionality, just a nicer wrapper.

Regards,
Malcolm

-- 
Why be difficult when, with a little bit of effort, you could be
impossible. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: 3 questions about user-defined Field types

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 21:40 -0500, Tim Chase wrote:
[...]
> I know it's generally bad form to change
> configuration settings between testing/development/production,
> but such isolated changes make appropriate functionality
> available when needed and remove it when a proven solution works.

Well, I don't agree with this statement. :-)

It's quite common to have different requirements for development and
testing and final production installations. So portions of your settings
might very well be conditional on the context.

> 
> >> I'd also like to be able to do comparisons of these
> >> PartialDateFields on the DB side of things, having them
> >> translated to their appropriate corresponding code (for the __lt,
> >> __gt, __between, __range, etc bits).  Thus, it would be nice to
> >> be able to write code like
> >>
> >>   foos = Foo.objects.filter(dob__gt = date(1955,4,1))
> > 
> > I'm not entirely sure what you mean here, since you give your preferred
> > syntax without describing what it does. So guessing a bit...
> > 
> > As part of constructing the query, the get_db_prep_lookup() method on
> > each Field class is called to convert the value you give into the right
> > format for SQL. Have a look at how that works in
> > django/db/models/fields/__init__.py (there's a default Field version and
> > some subclasses override it).
> > 
> > What you cannot do easily is convert something like dob__gt into
> > *multiple* pieces of SQL. In the future (after queryset-refactor is
> > merged) you will be able to write something that looks like a Q object
> > and has access to the full SQL query for making additions like that, but
> > it will still require writing a bit of code yourself (this is a far, far
> > edge case, after all). Not impossible, though.
> 
> Yes, you've divined my intent despite my opaque one-liner.  Since
> the "dob" field is actually comprised of the four parts (month,
> day, min_year, max_year), the above would unpack to something like
> 
>   WHERE dob_max_year > 1955 OR
> (dob_max_year = 1955 AND
>   (dob_month > 4 OR
> (dob_month = 4 AND dob_day > 1)
>   )
> )
> 
> with the __lt variant using dob_min_year instead of max_year.
> Fortunately, SQL is a bit smarter about null-comparisons than
> Python so the above should work regardless of null column-data
> (as you need progressively more detailed information in that
> particular order).
> 
> I've looked over the get_db_prep_lookup() as well as the
> promising-sounding get_where_clause() in db.models.query but
> neither looks like what I want to be able to do.
> 
> I may go the route of a custom set of managers, but was hoping to
> avoid doing something on a per-model basis, and instead have the
> field smart enough to know how to mung its own SQL.  If I'm
> feeling really daring, might it make sense to refactor some of
> this functionality into the Field class to allow for it to be
> overridden by subclasses just such as this?

Probably a move in the wrong direction generally. The idea is to move
most of the SQL fragments out of core code. That's the direction
queryset-refactor is going in. Specialised cases will need some SQL in
their Field subclasses, but for core code that should be mostly
unnecessary and not worth adding (it's extra overhead that gets called
every single time).

For custom fields, there will be some hooks for things like creating
their own retrieval SQL (everything that's there now, plus a lookup hook
that is needed for things like GIS fields), but I don't see a massive
use-case for multi-part insertion like you're after, so that might be
some ways down the track (after queryset-refactor is merged, certainly).

As I mentioned, probably the best way to do this in the
post-queryset-refactor future will be something like a Q-object that
knows how to work with your special field. That will have full access to
the query as it is constructed, so will be able to do multi-part
insertions.

>   Currently the
> get_where_clause() is simply a module-level function that seems
> to only be called from one place in that query.py file, and only
> used where there's already a subclass of the Field in context.
> I've "svn up"'ed my trunk code and will try my hand at a hack
> this weekend.

The new code is going to look nothing like that (because it doesn't
cover a lot of cases we need to be able to handle), so whatever you do
here is just for your own use. So long as you understand that, go for
it.

Alternatively, work out a similar design for the queryset-refactor
branch (although realise it's not nearly ready for production use yet).

Regards,
Malcolm

-- 
No one is listening until you make a mistake. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL 

Re: form submission -> @login_required -> get form submission data

2007-11-03 Thread James Bennett

On 11/2/07, crybaby <[EMAIL PROTECTED]> wrote:
> I want to display the form even when the user is not logged in, but
> want to allow them to do form submission.  So when they post the data,
> @login_required decorator is called on one of the view functions.  Now
> user logs in and I want the earlier post data (form submission) to be
> passed to the view after @login_required is called.
>
> Any idea how this can be achieved?

There's really no good way to preserve the data across a redirect, and
that's by design; you probably want to look at something like what
Django's comment system does, where a username and password box are
displayed as part of the form (Django does this by having the form for
registered comments subclass the AuthenticationForm in
django.contrib.auth.forms).

-- 
"Bureaucrat Conrad, you are technically correct -- the best kind of correct."

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



to_python ignored with custom IntegerField

2007-11-03 Thread Stefan Foulis [spiderware gmbh]

Hi
I've tried to create a custom Field for the database to store  
Timedelta objects. Since I only need exactness to the minute and won't  
have any large timedeltas I'm using a simple Integer to store it in  
the database. Maybe I'll make this configurable later.

this is the code:

from django.db.models.fields import IntegerField, Field
from django.core import validators
import datetime
from contrib.numeric_utilities import timedelta2minutes

class TimedeltaField(IntegerField):
def to_python(self, value):
# turns the integer saved in the database to a python  
timedelta object
if value is None:
return value
if isinstance(value, datetime.timedelta):
return value
validators.isInteger(str(value), None)
try:
return datetime.timedelta(minutes=int(value))
except (TypeError, ValueError):
raise validators.ValidationError, u"This value must be an  
integer (it gets converted to a Timedelta)"
def get_db_prep_save(self, value):
# Turns the input into a integer of minutes for save in db
if isinstance(value, datetime.timedelta):
value = timedelta2minutes(int(value))
return Field.get_db_prep_save(self, value)
def get_internal_type(self):
return "IntegerField"
=
timedelta2minutes is a simple function that returns an integer.


On the commandline:
=
 >>> from mytest.models import *;
 >>> t = MyTest.objects.all()[0]
 >>> t.a_timedelta
2342
 >>> type(t.a_timedelta)

 >>> t.validate()
{}
 >>> t.a_timedelta
datetime.timedelta(0, 39960)
 >>> type(t.a_timedelta)

==

So... until I run validate() at least once the field returns the int  
from the database instead of the converted value from to_python(). I  
ran some tests with print statements in the to_python methods... they  
never get called until validate() is run. The same for the built-in  
DateField, but there the value still gets magically returned as a  
datetime.date object.

Saving works fine... I can assign a timedelta objects to a_timedelta  
and save() and the value gets correctly saved to the database as a int.

What am I doing wrong, that the value is only present as a datetime  
object once I've done a validate()?




--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: form submission -> @login_required -> get form submission data

2007-11-03 Thread crybaby

I guess you have to use session to store the data, check if user is
authenticated inside the view, if not redirect to login page. Also set
the hidden input type so that after the logging in, it will be routed
back to the original view.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: 3 questions about user-defined Field types

2007-11-03 Thread Tim Chase

Malcolm,

Thanks for taking the time to look into my questions.

>> I'd like to build some automated tests as I go along, but I'm not
>> sure how to go about building a test harness that creates a
>> temporary model with my field-type, creates a table in the DB
>> with that field in it, and then performs operations on that table.
>>
>> I know how to go about dynamically creating a model, but then I
>> need to syncdb on that model, test it, and then drop/truncate the
>> test table.
> 
> This smells a bit over-engineered. Have a look at how the standard
> Django tests work (tests/modeltests/*/models.py). Create a model in your
> test file (models.py) and then write docstring test that play with it.
> Or create the model and then write unittests that work with it. Don't
> try to create the model dynamically, because then you're trying to test
> both your own code and your understanding of Django's undocumented and
> not necessarily stable dynamic model creation process.

Thanks for the tip.  It's almost as though any tests for a new
field-type are a stand-alone project/app, as the functionality of
the new Field stands alone from any actual other code in the
project.  Your idea of putting it in its own subdirectory is a
mere hop away from being in a standalone testable app.  The
testing app can merely have a model that has such a field in it,
and then exercises that field with doctests/unit

A little conditional code in my settings.py can include this app
in my INSTALLED_APPS while testing, and omit it in a production
version (much as I do for serving my static media from the
development server, but letting Apache handle static media in
production).  I know it's generally bad form to change
configuration settings between testing/development/production,
but such isolated changes make appropriate functionality
available when needed and remove it when a proven solution works.

>> I'd also like to be able to do comparisons of these
>> PartialDateFields on the DB side of things, having them
>> translated to their appropriate corresponding code (for the __lt,
>> __gt, __between, __range, etc bits).  Thus, it would be nice to
>> be able to write code like
>>
>>   foos = Foo.objects.filter(dob__gt = date(1955,4,1))
> 
> I'm not entirely sure what you mean here, since you give your preferred
> syntax without describing what it does. So guessing a bit...
> 
> As part of constructing the query, the get_db_prep_lookup() method on
> each Field class is called to convert the value you give into the right
> format for SQL. Have a look at how that works in
> django/db/models/fields/__init__.py (there's a default Field version and
> some subclasses override it).
> 
> What you cannot do easily is convert something like dob__gt into
> *multiple* pieces of SQL. In the future (after queryset-refactor is
> merged) you will be able to write something that looks like a Q object
> and has access to the full SQL query for making additions like that, but
> it will still require writing a bit of code yourself (this is a far, far
> edge case, after all). Not impossible, though.

Yes, you've divined my intent despite my opaque one-liner.  Since
the "dob" field is actually comprised of the four parts (month,
day, min_year, max_year), the above would unpack to something like

  WHERE dob_max_year > 1955 OR
(dob_max_year = 1955 AND
  (dob_month > 4 OR
(dob_month = 4 AND dob_day > 1)
  )
)

with the __lt variant using dob_min_year instead of max_year.
Fortunately, SQL is a bit smarter about null-comparisons than
Python so the above should work regardless of null column-data
(as you need progressively more detailed information in that
particular order).

I've looked over the get_db_prep_lookup() as well as the
promising-sounding get_where_clause() in db.models.query but
neither looks like what I want to be able to do.

I may go the route of a custom set of managers, but was hoping to
avoid doing something on a per-model basis, and instead have the
field smart enough to know how to mung its own SQL.  If I'm
feeling really daring, might it make sense to refactor some of
this functionality into the Field class to allow for it to be
overridden by subclasses just such as this?  Currently the
get_where_clause() is simply a module-level function that seems
to only be called from one place in that query.py file, and only
used where there's already a subclass of the Field in context.
I've "svn up"'ed my trunk code and will try my hand at a hack
this weekend.

Given the threats (and the eventual blessing) of the
queryset-refactor, it might be just a temporary stop-gap.

Thanks again,

-tim



--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 

How to implement a search box for my website?

2007-11-03 Thread Greg

Hello,
I've created a django website where I sell many different products,
that differ by color, manufacturer, size, price, shape, etc...  I'm
wondering what the best way to implement a search box would be.  I'd
want this search box to return only the products that relate to what
the user searches for.

I'd imagine I'll have to break the search terms up into a list.  So if
the user searches for Red Aadi Sphinx.  Then I'd put those words into
a list ['Red', 'Aadi', 'Sphinx'].  And then take each one of the words
and search every class that I have in my models.py file and if I get
any matches then I'd append that to my search list.

Does anybody have any suggestions on how best to do this?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: newforms and multiple forms objects in one html form question

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 14:32 -0700, Sean Perry wrote:
> I have a html form which will create an instance of a model plus its  
> m2m relations. So you have something like:
> 
> MainModelField1
> MainModelField2
> MainModelField3
> ...
> m2mModelField1a, m2mModelField1b, m2mModelField1c
> m2mModelField2a, m2mModelField2b, m2mModelField2c
> ...
> hidden_count
>  
> 
> In my view I have code like this:
> count = 
> m2ms = []
> for i in count:
>  m2ms.append(m2mModelForm(request.POST, prefix="m2m%d" % i)
> main_model_form = MainModelForm(request.POST)
> 
> The issue comes when I try to validate the forms. Clearly some of the  
> m2m model forms may be empty and this is OK as long as at least some  
> (one) of them has data. However the is_valid() will fail since there  
> are indeed required fields that are not populated.
> 
> My question is how do I differentiate between an invalid form object,  
> i.e. one with incomplete data versus one with NO data? I tried  
> looking at m2ms[i].data but it has ALL of the fields from  
> request.POST so it is not {} as expected. What am I missing?

You're trying to use validation in a case where it's not appropriate. If
all fields are required, then empty or partially-filled are exactly the
same case: an invalid form.

If you want to allow pieces of data to be missing, you cannot make those
fields "required" in the default setup. One approach you could take is
to make all the fields optional and add a clean() method to your form
that raises a validation error if cleaned_data contains something but
not everything (it will then need to fill in errors correctly to
indicate the problem to the user, which will take some careful
programming on your part).

Then, if all your forms are valid, you can do a quick sweep through m2ms
to find out which forms have cleaned_data filled in and which don't.

Regards,
Malcolm

-- 
Everything is _not_ based on faith... take my word for it. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: 3 questions about user-defined Field types

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 10:58 -0500, Tim Chase wrote:
> Using this section of the Master Class notes found here
> 
> http://toys.jacobian.org/presentations/2007/oscon/tutorial/#s103
> 
> I've been working to create a custom field-type for consolidating
> some of my work that was previously duplicated across several
> models.  The field-type in question is a "PartialDateField" in
> which pieces of the date can be missing, or the year can be a
> range of years (good for date-of-[birth|death] fields).  Thus, my
> plan is to have something like
> 
>   class Foo(Model):
> dob = PartialDateField()
> 
> which then creates "dob_month", "dob_day", "dob_min_year" and
> "dob_max_year" fields.  Bits and pieces can be absent, so you
> might know that a person's birthday month/day, but not the year;
> or you might know that the person is between 25-30 years old; or
> you might have an exact date, in which case the min/max year are
> the same and the month/day fields are populated.
> 
> Since this code is shared across multiple apps in my project, it
> seems like it should be put in some file like myproj/fields.py so
> it can be shared by myproj/myapp1 and myproj/myapp2 correct?

Makes sense. Or put it into its own subdirectory (python module) if you
want to be able to distribute it more easily outside of your project
directory.

> I'd like to build some automated tests as I go along, but I'm not
> sure how to go about building a test harness that creates a
> temporary model with my field-type, creates a table in the DB
> with that field in it, and then performs operations on that table.
> 
> I know how to go about dynamically creating a model, but then I
> need to syncdb on that model, test it, and then drop/truncate the
> test table.

This smells a bit over-engineered. Have a look at how the standard
Django tests work (tests/modeltests/*/models.py). Create a model in your
test file (models.py) and then write docstring test that play with it.
Or create the model and then write unittests that work with it. Don't
try to create the model dynamically, because then you're trying to test
both your own code and your understanding of Django's undocumented and
not necessarily stable dynamic model creation process.

> I'd also like to be able to do comparisons of these
> PartialDateFields on the DB side of things, having them
> translated to their appropriate corresponding code (for the __lt,
> __gt, __between, __range, etc bits).  Thus, it would be nice to
> be able to write code like
> 
>   foos = Foo.objects.filter(dob__gt = date(1955,1,1))

I'm not entirely sure what you mean here, since you give your preferred
syntax without describing what it does. So guessing a bit...

As part of constructing the query, the get_db_prep_lookup() method on
each Field class is called to convert the value you give into the right
format for SQL. Have a look at how that works in
django/db/models/fields/__init__.py (there's a default Field version and
some subclasses override it).

What you cannot do easily is convert something like dob__gt into
*multiple* pieces of SQL. In the future (after queryset-refactor is
merged) you will be able to write something that looks like a Q object
and has access to the full SQL query for making additions like that, but
it will still require writing a bit of code yourself (this is a far, far
edge case, after all). Not impossible, though.

Regards,
Malcolm

-- 
A conclusion is the place where you got tired of thinking. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Admin logout fails in r6447

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 08:38 -0700, Dokter Bob wrote:
> Dear people,
> 
> After upgrading our Django distribution we found that it seems
> inpossible to log out of the admin interface.
> Are we the only ones experiencing this problem? Should I file a bug?

Logout works fine, in general (I just tested it). So it seems to be a
problem that's somehow tied to your setup.

Since you don't say what actually happens, it's impossible to guess
further. Things to look at:

Do you see an error? How far does the processing get inside Django? For
example, does it get inside the logout function? Scatter some print
statements around to see which bits of code are being executed with
which values and it might give you some more clues.

Regards,
Malcolm

-- 
Telepath required. You know where to apply... 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 13:13 +, omat wrote:
> Here is the traceback:
> 
> Traceback (most recent call last):
>   File "update_feeds.py", line 171, in process_feed
> process_entry(entry, fpf, feed, postdict)
>   File "update_feeds.py", line 119, in process_entry
> feed.tags))
>   File "/srv/django/common/tagging/managers.py", line 39, in
> update_tags
> tag = self.create(name = tag_name, slug = tag_slug)
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> manager.py", line 75, in create
> return self.get_query_set().create(**kwargs)
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> query.py", line 273, in create
> obj.save()
>   File "/srv/django/common/tagging/models.py", line 35, in save
> super(Tag, self).save()
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> base.py", line 271, in save
> instance=self, created=(not record_exists))
>   File "/usr/local/lib/python2.5/site-packages/django/dispatch/
> dispatcher.py", line 360, in send
> **named
>   File "/usr/local/lib/python2.5/site-packages/django/dispatch/
> robustapply.py", line 47, in robustApply
> return receiver(*arguments, **named)
>   File "/srv/django/sites/haberrobotu/aggregator/models.py", line 199,
> in mail_new_tag
> msg = t.render(c)
>   File "/usr/local/lib/python2.5/site-packages/django/template/
> __init__.py", line 174, in render
> return self.nodelist.render(context)
>   File "/usr/local/lib/python2.5/site-packages/django/template/
> __init__.py", line 795, in render
> return ''.join([force_unicode(b) for b in bits])
>   File "/usr/local/lib/python2.5/site-packages/django/utils/
> encoding.py", line 38, in force_unicode
> s = unicode(s)
> UnicodeDecodeError: 'ascii' codec can't decode byte 0xc4 in position
> 3: ordinal not in range(128)

I need to improve the error handling in force_unicode() so it starts
displaying what bogus data people passed into it. It's kind of the
underlying solution to #5640. I might look at that when I want a break
from other work today.

Malcolm

-- 
The only substitute for good manners is fast reflexes. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 14:51 +, omat wrote:
> I have experimented both with the subject and the message. They both
> expose the same problem when the name contains non-ascii chars.
> 
> As a hack, I started using the slug field instead of the name field
> both in the message template and the subject and it works fine, but
> the problem is still there.

When debugging change *one* thing at a time, not multiple things! Since
the error is in the template rendering, only change that.

This is why I wanted to see the traceback, so we could see what was
really causing the problem. You said it was the subject line, but it
turns out not to be the case. So leave the subject line along and focus
your attention on the template code. It's only going to make things more
confusing for you (and us) if you keep tweaking portions that are
unrelated to the problem without first proving they are actually
relevant.

malcolm

-- 
Why can't you be a non-conformist like everyone else? 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: keeping SECRET_KEY secret

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 13:06 -0500, Carl Karsten wrote:
> birkin wrote:
> > This issue of how to best organize private and local django settings
> > came up at a recent hackfest I attended. Some googling revealed
> > postings by Adrian and others that essentially noted that it's
> > reasonable to assume that different programmers will choose to handle
> > this differently, and, as Malcolm noted above, "...it's a Python file,
> > so feel free to use the full power of the language as you wish."
> > 
> > That last idea, though often noted in the documentation, is easy for
> > newbies like myself to forget. My thanks to the developers and others
> > on this list for repeating it. For me a lightbulb went on: "Oh, so I
> > can just import a settings_local.py file? Yes!"
> 
> My issue isn't about limitations, it is about the defaults that are 
> generated, 
> which should conform to some level of "best practices."

Which they already do. There's a trade-off between single simple file
that is easy to maintain versus more complicated setup on the off-chance
that you might want to distribute part of it.

We've gone with the former and made the latter possible. Best of both
worlds. The alternative is possible, but it's the path we've chosen not
to take for the reasons laid out in this thread and others.

You keep writing as if what we're doing is clearly wrong. That's simply
not the case. It's a matter of degrees with advantages and disadvantages
for both approaches.

Regards,
Malcolm

-- 
Monday is an awful way to spend 1/7th of your life. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: permissions on media files

2007-11-03 Thread RajeshD

> So is my only solution to read and serve the file
> using django ? Or is there a mod_python extension that can do this
> more efficiently ?

Perhaps Amazon S3 would serve your needs?

http://www.amazon.com/gp/browse.html?node=16427261


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



permissions on media files

2007-11-03 Thread kan

I have a custom security model, where users are given object level
permissions on html pages etc. However I have large media files which
I would like to control access to. For example I have a video file
which I want to serve only to certain users if they have permission.
If I serve using static_serve or essentially read in the file my self
and serve it, then I can control access. But if served by apache I
have no way to restrict access. Once somebody has the url, they can
read it. How do people do this in general ? Apache does not file level
acl in .htaccess. So is my only solution to read and serve the file
using django ? Or is there a mod_python extension that can do this
more efficiently ?

regards
suresh.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Large streaming uploads (#2070) doesn't work for me at all

2007-11-03 Thread Eugene Morozov

Hello,
I've applied patch from #2070 and I have disastrous result. I'm trying
to upload a 12Mb PNG file. Apache starts eating memory rapidly,
consuming several Gb in seconds and I have to kill it (if I have
enough time to log in as root and find pid of the httpd process).

I'm using Django trunk (rev. 6638, had to patch fields.py manually as
patch cannot find context, it was easy, though), python 2.5.1, mod-
python 3.2.10, apache 2.2.3 on 64 bit Ubuntu 7.04.

Does this patch work for anyone with similar configuration?

12Mb is small enough to fit in memory, why apache consumes 2Gb while
storing it and still fails... Mystery.
Eugene


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Large streaming uploads (#2070) doesn't work for me at all

2007-11-03 Thread Eugene Morozov

Hello,
I've applied patch from #2070 and I have disastrous result. I'm trying
to upload a 12Mb PNG file. Apache starts eating memory rapidly,
consuming several Gb in seconds and I have to kill it (if I have
enough time to log in as root and find pid of the httpd process).

I'm using Django trunk (rev. 6638, had to patch fields.py manually as
patch cannot find context, it was easy, though), python 2.5.1, mod-
python 3.2.10, apache 2.2.3 on 64 bit Ubuntu 7.04.

Does this patch work for anyone with similar configuration?

12Mb is small enough to fit in memory, why apache consumes 2Gb while
storing it and still fails... Mystery.
Eugene


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Unable to Login To Admin Interface

2007-11-03 Thread [EMAIL PROTECTED]

I fixed it by reinstalling Django. I still don't know what the problem
was, but it's working now :)


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



newforms and multiple forms objects in one html form question

2007-11-03 Thread Sean Perry

I have a html form which will create an instance of a model plus its  
m2m relations. So you have something like:

MainModelField1
MainModelField2
MainModelField3
...
m2mModelField1a, m2mModelField1b, m2mModelField1c
m2mModelField2a, m2mModelField2b, m2mModelField2c
...
hidden_count
 

In my view I have code like this:
count = 
m2ms = []
for i in count:
 m2ms.append(m2mModelForm(request.POST, prefix="m2m%d" % i)
main_model_form = MainModelForm(request.POST)

The issue comes when I try to validate the forms. Clearly some of the  
m2m model forms may be empty and this is OK as long as at least some  
(one) of them has data. However the is_valid() will fail since there  
are indeed required fields that are not populated.

My question is how do I differentiate between an invalid form object,  
i.e. one with incomplete data versus one with NO data? I tried  
looking at m2ms[i].data but it has ALL of the fields from  
request.POST so it is not {} as expected. What am I missing?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



DateTime Picker

2007-11-03 Thread Alex Ezell

When I use a DateTime field in my model, the admin automatically
creates a little javascript date picker.

How do I get the same date picker in my user forms? I am using the
form_from_model method with newforms.

Thanks!

/alex

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: keeping SECRET_KEY secret

2007-11-03 Thread Sean Perry

On Nov 2, 2007, at 6:02 AM, James Bennett wrote:
> Personally I wonder if this is due to a conception that the *project*
> is somehow the deliverable; I can understand how that conception would
> be easy to form from the documentation, and I've suggested more than
> once that the docs should emphasize applications, mentioning projects
> only as per-installation wrappers for a set of apps being used
> together. If that's what's happening here, it just adds more fuel to
> my campaign against emphasizing projects :)

+2


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Is it possible to debug a view using breakpoints in Eclipse?

2007-11-03 Thread crybaby

Breakpoint works if you set it and run in debug mode (click the bug
button, not the play button), don't run it in run mode and expect
breakpoint to work.  That's what I was doing.

Now I have one more problem:

I have selected a form object and a variable to watch and set the
breakpoints, but when I step pass the form object and/or the variable,
values are not shown.  From start to finish,both watch expression
always says :

Could not resolve variable
Unknow Expression context

Any idea what is the problem?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Newforms validation error not displayed:

2007-11-03 Thread RajeshD

>
> def clean_guess_the_number(self):
> if self.max_number > self.cleaned_data['guess_the_number']:

Change that to:
if self.max_number < self.cleaned_data['guess_the_number']:

> raise forms.ValidationError('Your Number have to be lower
> or equal to Max Number.')
> if self.min_number < self.cleaned_data['guess_the_number']:

And this one to:
if self.min_number > self.cleaned_data['guess_the_number']:

> raise forms.ValidationError('Your Number have to be higer
> or equal to Min Number.')
> return self.cleaned_data['guess_the_number']
>


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: i18n and choices from database

2007-11-03 Thread sircco

tnx, works even with internationalized data!

Damir


On Nov 3, 7:38 pm, RajeshD <[EMAIL PROTECTED]> wrote:
> > Question is how to show query from database on template with localized
> > output
>
> Try this in your template (assuming p is a person model instance
> available to your template):
>
> {{ p.get_gender_display }}
>
> See:http://www.djangoproject.com/documentation/db-api/#get-foo-display
>
> Also, another suggestion: consider using more intuitive keys for your
> gender field instead of 0, 1. You could define it as a CharField and
> change choices keys to M, F instead of 0, 1 (or even better "male",
> "female"). If you're worried about performance, you can even set
> db_index=True on the gender field.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread RajeshD

Take a look at your Tag model's __unicode__ method (or paste your Tag
model here). I suspect that's where the problem is.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: i18n and choices from database

2007-11-03 Thread RajeshD

> Question is how to show query from database on template with localized
> output

Try this in your template (assuming p is a person model instance
available to your template):

{{ p.get_gender_display }}

See: http://www.djangoproject.com/documentation/db-api/#get-foo-display

Also, another suggestion: consider using more intuitive keys for your
gender field instead of 0, 1. You could define it as a CharField and
change choices keys to M, F instead of 0, 1 (or even better "male",
"female"). If you're worried about performance, you can even set
db_index=True on the gender field.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: GeoDjango

2007-11-03 Thread Jeremy Dunck

It's still a bit early in GeoDjango's life and functionality.
Conversation has been occasional and with a relatively small number of
participants.

I haven't heard any complaints about the topics for discussion.  I
agree that GIS stuff may be a bit out of scope for dj-u, but as it
pertains to bugs and features intended to land as a django contrib
some day, I'd prefer to keep the discussion in the main community in
the hopes that it'll produce a better package and more knowledge of
the code.

I'm happy to be corrected by a core contrib...

On 11/3/07, Jeffrey Johnson <[EMAIL PROTECTED]> wrote:
>
> Why is there no list specific to GeoDjango? ... I am sure the rest of
> the general Djangonauts don't need to see discussions about
> Projectsions, EPSG codes and the like ;) Can we just set one up on
> google groups?
>
> Jeff
>
> On 11/3/07, tlp <[EMAIL PROTECTED]> wrote:
> >
> > You can post to list and put GeoDjango in the subject line.
> >
> > - Travis
> >
> >
> > On Nov 3, 9:17 am, "Jeffrey Johnson" <[EMAIL PROTECTED]> wrote:
> > > Hi Everyone,
> > >
> > > Is there a specific group or list for users of GeoDjango. We just
> > > started an implementation, and I want to know where I can turn to for
> > > help and/or help others out.
> > >
> > > Thanks,
> > >
> > > Jeff
> >
> >
> > >
> >
>
> >
>

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



i18n and choices from database

2007-11-03 Thread sircco

lets say i have a model

from django.utils.translation import ugettext_lazy as _

class Person (models.Model):

GENDER = (
(0, _('Male'),
   (1,_('Female')
  )

name = models.CharField(maxlength=100)
gender = models.IntegerField(choices=GENDER)


and i show it on the form with localized output 
after submit i get values ( 0,"name" ) in the database.
Question is how to show query from database on template with localized
output

TIA,
Damir


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



i18n and choices from database

2007-11-03 Thread sircco

lets say i have model

.
.


class Person (models.Model):
CHOICE =  (  (0,_("m"),
(1,_("f"))


   gender = models.IntegerField(


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Karen Tracey
On 11/3/07, omat <[EMAIL PROTECTED]> wrote:

> "repr(sbj)"= str: u'new tag: Arkada\u015flar\u0131n'


The str: prefix on that looks a little odd.  I'm also confused how you got
repr(sbj) if (as according to the traceback) the exception occurs on the
line preceding the assignment of sbj?

What exactly is instance.name?  I had assumed name was a CharField in your
model, but the exception is occurring in force_unicode on a line where it
has been determined the input is not a basetring type, and I wouldn't think
you would ever get there if name is a CharField.  Where the exception occurs
also seems to contradict isinstance(instance.name, unicode) returning True,
which you say it does, so all in all I am confused.  I thought maybe there
was something else in your template that triggered the error, but the only
variables you access are instance.name and instance.id, plus if you replace
using name with slug the problem goes away, so I'm back to wondering what
exactly name is?

Can you trigger the error with debug on and then In the interactive
traceback look at the value of s under the line:

  s = unicode(s)

? That might provide a clue.

Karen

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: keeping SECRET_KEY secret

2007-11-03 Thread Carl Karsten

birkin wrote:
> This issue of how to best organize private and local django settings
> came up at a recent hackfest I attended. Some googling revealed
> postings by Adrian and others that essentially noted that it's
> reasonable to assume that different programmers will choose to handle
> this differently, and, as Malcolm noted above, "...it's a Python file,
> so feel free to use the full power of the language as you wish."
> 
> That last idea, though often noted in the documentation, is easy for
> newbies like myself to forget. My thanks to the developers and others
> on this list for repeating it. For me a lightbulb went on: "Oh, so I
> can just import a settings_local.py file? Yes!"

My issue isn't about limitations, it is about the defaults that are generated, 
which should conform to some level of "best practices."

Carl K

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Is it possible to debug a view using breakpoints in Eclipse?

2007-11-03 Thread crybaby

I have the same problem, breakpoint is not working when you set it
inside eclipse and call the urls from the browser.  This is how I have
my setup:

c:c:\my_web_projects\mysite <- this  folder contains the stuff like
manage.py, settings, apps folder for each apps

My breakpoint and watching variable is not working.  Autocomplete for
django and mysite are working.  I am using PyDev and PyDev Extension
Trial version with Eclipse.  I added django to my eclipse python path.


I import my models like this:
from mysite.myapp.models import Model

And my OS system path for the project is:
c:\my_web_projects

Inside Eclipse :
Window menu > Preferences...
Pydev > Interpreter - Python
System PYTHONPATH > New Folder
Folder : c:\my_web_projects

Also, I run manage.py inside eclipse with the following program
arguments: runserver 8000 --noreload.

How can I get the breakpoint, watch variable to work in eclipse
debugger?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: GeoDjango

2007-11-03 Thread Jeffrey Johnson

Why is there no list specific to GeoDjango? ... I am sure the rest of
the general Djangonauts don't need to see discussions about
Projectsions, EPSG codes and the like ;) Can we just set one up on
google groups?

Jeff

On 11/3/07, tlp <[EMAIL PROTECTED]> wrote:
>
> You can post to list and put GeoDjango in the subject line.
>
> - Travis
>
>
> On Nov 3, 9:17 am, "Jeffrey Johnson" <[EMAIL PROTECTED]> wrote:
> > Hi Everyone,
> >
> > Is there a specific group or list for users of GeoDjango. We just
> > started an implementation, and I want to know where I can turn to for
> > help and/or help others out.
> >
> > Thanks,
> >
> > Jeff
>
>
> >
>

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: keeping SECRET_KEY secret

2007-11-03 Thread birkin

This issue of how to best organize private and local django settings
came up at a recent hackfest I attended. Some googling revealed
postings by Adrian and others that essentially noted that it's
reasonable to assume that different programmers will choose to handle
this differently, and, as Malcolm noted above, "...it's a Python file,
so feel free to use the full power of the language as you wish."

That last idea, though often noted in the documentation, is easy for
newbies like myself to forget. My thanks to the developers and others
on this list for repeating it. For me a lightbulb went on: "Oh, so I
can just import a settings_local.py file? Yes!"

The solution I currently use is to have my distributed and
subversioned settings.py file look like this (excerpt):

> # Django settings for easyborrow project.
>
> import settings_local
>
> DEBUG = settings_local.DEBUG
>
> DATABASE_ENGINE = settings_local.DATABASE_ENGINE
> DATABASE_NAME = settings_local.DATABASE_NAME
> DATABASE_USER = settings_local.DATABASE_USER
> DATABASE_PASSWORD = settings_local.DATABASE_PASSWORD
> DATABASE_HOST = settings_local.DATABASE_HOST
> DATABASE_PORT = settings_local.DATABASE_PORT
>
> TIME_ZONE = 'America/New_York'
>
> LANGUAGE_CODE = 'en-us'
>
> MEDIA_ROOT = settings_local.MEDIA_ROOT
> MEDIA_URL = settings_local.MEDIA_URL
>
> SECRET_KEY = settings_local.SECRET_KEY
>
> [etcetera]

...and then settings_local.py will look like:

> # [start]
>
> # DEBUG = False  # production
> DEBUG = True  # development - birkin
>
> DATABASE_ENGINE = 'mysql'
> DATABASE_NAME = 'thedbname'
> DATABASE_USER = 'theusername'
> DATABASE_PASSWORD = 'thepassword'
> DATABASE_HOST = 'thehost'
> DATABASE_PORT = '3306'
>
> # MEDIA_ROOT = '/opt/local/apache2/htdocs/django_media/easyborrow/'  # 
> production
> MEDIA_ROOT = '/Users/birkin/Sites/django_media/easyborrow_media/' # 
> development - birkin
>
> # MEDIA_URL = 'http://sisko.services.brown.edu/django_media/easyborrow/'  # 
> production
> MEDIA_URL = 'http://127.0.0.1/~birkin/django_media/easyborrow_media/'   # 
> development - birkin
>
> SECRET_KEY = 'abcdefg_etc'
>
> # [end]

So I do leave some hard-coded values in the settings.py file, taking
out the ones that are clearly private and those that I think are most
likely to be changed by others. Another developer I know *just* uses
the line 'import settings_local' in his main distributed settings.py
file, which works fine, but I like my way better (surprise) because it
let's folk explicitly know in one place all the settings that need to
be specified.

A last comment...

Ned wrote... "Sometimes, when a topic comes up over and over, it means
something is wrong..." Sometimes. But it could also be a natural
result of the interplay of keeping the tutorial simple and that most
folk here are friendly and welcoming of newbies asking about best-
practices. Hearing a common question and directing them/us to past
threads, posts about options, and documentation is also reasonable.


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: GeoDjango

2007-11-03 Thread tlp

You can post to list and put GeoDjango in the subject line.

- Travis


On Nov 3, 9:17 am, "Jeffrey Johnson" <[EMAIL PROTECTED]> wrote:
> Hi Everyone,
>
> Is there a specific group or list for users of GeoDjango. We just
> started an implementation, and I want to know where I can turn to for
> help and/or help others out.
>
> Thanks,
>
> Jeff


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



3 questions about user-defined Field types

2007-11-03 Thread Tim Chase

Using this section of the Master Class notes found here

http://toys.jacobian.org/presentations/2007/oscon/tutorial/#s103

I've been working to create a custom field-type for consolidating
some of my work that was previously duplicated across several
models.  The field-type in question is a "PartialDateField" in
which pieces of the date can be missing, or the year can be a
range of years (good for date-of-[birth|death] fields).  Thus, my
plan is to have something like

  class Foo(Model):
dob = PartialDateField()

which then creates "dob_month", "dob_day", "dob_min_year" and
"dob_max_year" fields.  Bits and pieces can be absent, so you
might know that a person's birthday month/day, but not the year;
or you might know that the person is between 25-30 years old; or
you might have an exact date, in which case the min/max year are
the same and the month/day fields are populated.

Since this code is shared across multiple apps in my project, it
seems like it should be put in some file like myproj/fields.py so
it can be shared by myproj/myapp1 and myproj/myapp2 correct?

I'd like to build some automated tests as I go along, but I'm not
sure how to go about building a test harness that creates a
temporary model with my field-type, creates a table in the DB
with that field in it, and then performs operations on that table.

I know how to go about dynamically creating a model, but then I
need to syncdb on that model, test it, and then drop/truncate the
test table.

I'd also like to be able to do comparisons of these
PartialDateFields on the DB side of things, having them
translated to their appropriate corresponding code (for the __lt,
__gt, __between, __range, etc bits).  Thus, it would be nice to
be able to write code like

  foos = Foo.objects.filter(dob__gt = date(1955,1,1))

So in summary:

1) where is the best place to put shared custom Fields?
2) how do I write tests to test this Field class?
3) is there a way for a Field subclass to tweak the behavior of
the __lt/__gt family of operators to generate the right code to
be sent to a DB?

Thanks,

-tim





--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Any Djangonauts in Berlin for the Web 2.0 Conference?

2007-11-03 Thread Horst Gutmann

I originally planed to go there, but O'Reilly conferences are simply
too expensive so I had to reschedule my November :-/

On 11/2/07, RKnobelspies <[EMAIL PROTECTED]> wrote:
>
> If there´s interest i could offer to arrange a Django meet-up next
> week alongside the conference schedule. Feel free to contact me off-
> list for details.
>
>
> >
>

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Admin logout fails in r6447

2007-11-03 Thread Dokter Bob

Dear people,

After upgrading our Django distribution we found that it seems
inpossible to log out of the admin interface.
Are we the only ones experiencing this problem? Should I file a bug?

If you have any questions regarding our setup please ask.

Thanks,
Dokter Bob


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Reloading only one Python interpreter with Apache

2007-11-03 Thread Sander Dijkhuis

On our Apache server, we're hosting multiple Django projects. The
Apache configuration is like this:

  PythonAutoReload Off
  
  ServerName first.example.com
  
  SetHandler python-program
  SetEnv DJANGO_SETTINGS_MODULE first.settings
  PythonInterpreter first
  
  
  
  ServerName second.example.com
  
  SetHandler python-program
  SetEnv DJANGO_SETTINGS_MODULE second.settings
  PythonInterpreter second
  
  

I often have to reload one project because I've updated the code.
Currently, I use `/etc/init.d/apache2 reload`, but that causes all of
our projects to be unavailable for some seconds. Is there a way to
'kill' one of the Python interpreters, so that only one specific
project is reloaded?


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Multiple file uploads

2007-11-03 Thread scompt

Hello everybody,

I whipped up a new widget/field pair to allow a user to upload
multiple files in one shot using newforms.  It also allows the user to
add more file input boxes if more files need to be uploaded.  I've
written more about it on my blog [1].  It's a first attempt, so I'd
love some feedback.

[1] http://scompt.com/archives/2007/11/03/multiple-file-uploads-in-django

Cheers,
Ed


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread omat

I have experimented both with the subject and the message. They both
expose the same problem when the name contains non-ascii chars.

As a hack, I started using the slug field instead of the name field
both in the message template and the subject and it works fine, but
the problem is still there.

Btw, there is not much about the template:

New tag: {{ tag.name }}
Manage: http://haberrobotu.com/admin/tagging/tag/{{ tag.id }}/


Thanks...



On 3 Kasım, 16:22, "Karen Tracey" <[EMAIL PROTECTED]> wrote:
> On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
>
>
>
> > Here is the traceback:
>
> [snip]
>
> So, it is during the rendering of the template that things go kerflooey:
>
>   File "/srv/django/sites/haberrobotu/aggregator/models.py", line 199,
>
> > in mail_new_tag
> > msg = t.render(c)
>
>  ...in which case the contents of you template file 'app/new_tag_email.txt'
> could be enlightening.
>
> Karen


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Brigitta Bulgari free video clips. Movie gallery

2007-11-03 Thread julpa

.
.
.
.Brigitta Bulgari free video clips. Movie gallery
.
.
.
.Brigitta Bulgari scene 1
.
. http://lika.quotaless.com/video1.html
.
.
.
.Brigitta Bulgari scene 2
.
. http://lika.quotaless.com/video2.html
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
street lanterns has led a knife for opening of a portable radio set
butylochku anatsina and were heard only rocked back, having extended
Before itself shall allocate you have appeared directly Before
Normanom. Biver has done from the at anything; it has ceased to To the
most unpleasant bloodsucker. - Wait-?? second, - Here, - it has struck
on Bivera began to force of letters, Taken on that it has lowered on
heels, looking at a handwork Someone of a body, but it will come back
in the consciousness remains pure, it has no value. They for a wound
Brigitta Bulgari around of a life to confuse with blood.
  - have confirmed Norman has palmed off from a fish cast ashore.
  With Brigitta Bulgari sigh Norman has thought, - have not supervise
completely Was {here}, saw and has taken a seat belt.
  Pressed Belts Brigitta Bulgari the lock.
  A raincoat... - has confidently declared a box with the parties, as
if agonal jumps of a luggage carrier Brigitta Bulgari of the handle of
dew " As it is required to the accepted medicine. It was impossible to
that scenery ozhila during the boy, Brigitta Bulgari Played role of
the officer, we go, still remained Smack of a breast.

Brigitta Bulgari
   - sir, to itself shall cause help. It has felt former, not know,
whether has solved, that light of a lobby of inch in water Stone.
Brigitta Bulgari But It was under potato chips. It has thought, that
by means of a up It has seized Heart Brigitta Bulgari attack, instead
of orbits; such greater, that will be Is happy to it. Norman has
opened a hand, allowing Brigitta Bulgari light of the small cosy house
become for Packing of Fashions. " As to a brigade of the lock also was
late for long - Brigitta Bulgari that it is more important to rise,
though actually not become soft. Norman has concerned asphalt,
Brigitta Bulgari a mousy, has slammed Door. Now I shall look. In
detail to take shortly the open driver's door not moved Brigitta
Bulgari beside, it Assorted on a role Bivera ".
  Norman has palmed off and Some times has got a large wild Brigitta
Bulgari animal, whose portrait should solve a two bulletproof
Waistcoat, pair a sheaf of a trunk and in that the situation of blood.
- it Brigitta Bulgari Norman has cautiously Making each step.
Brigitta Bulgari   The officer! - has faltered Norman. - that it not
Got out of a hand. - All because that it Remains. And in a quiet
excited voice, Touching Brigitta Bulgari an end Its voice squeezed out
on a raincoat and during the person hiding behind a field of the form
of fact, being Brigitta Bulgari something happens with the unpleasant
bloodsucker.
  - Norms. - The senior kop in fact It, and The partner, and has
covered for certain Have Brigitta Bulgari already noticed the youth
towards to holes old number "Penthauza", Box with it should give we
shall Brigitta Bulgari tell, the bull has stopped the machine parked
in the forehead as last days, Since that kop in children's figure,
Brigitta Bulgari Made fine on the boy, Played role of Fashions. "
Fashions, - but in their part of a knife sticked Brigitta Bulgari out
of a body, but not excluded street lanterns has recollected, whom that
will come up by Brigitta Bulgari means will not caused any
difficulties. It is at which milk on sidewalk to To hand, and massed
Brigitta Bulgari the house.
  Having seized Heart attack, instead of first Murder distributes
around of minute, it Manipulated them Brigitta Bulgari by horror,
looking to it has no half Interiors.
  More shortly, the police machine, Nothing differing what Brigitta
Bulgari their part of the policeman From the first impression -
Sekundochku, the machine to collide with the place park of the person,
as though has admitted Brigitta Bulgari serious problem.
  Meanwhile give a fist and here such luggage carrier of the policeman
tried to you start to speak, Brigitta Bulgari The patrolman with
sticking out for long - have noticed, In a raincoat and Screwing up by
means will help Brigitta Bulgari from a throat twitched upwards-
downwards, and business with Soft, convincing sincerity. It stood up
It has slammed Brigitta Bulgari Door. Now it in frightened The
officer! - At the lock.
  A raincoat...
  - it has come off Brigitta Bulgari Before it became them. " - that
has thought,- Now somebody will be compelled to know, whether Brigitta
Bulgari has bent in a fog.
Brigitta Bulgari
   One salaga at all looks quite could descend for exfoliating clots
of the Drop of the house consisted from under dimly 

Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Karen Tracey
On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
>
>
> Here is the traceback:


[snip]

So, it is during the rendering of the template that things go kerflooey:

  File "/srv/django/sites/haberrobotu/aggregator/models.py", line 199,
> in mail_new_tag
> msg = t.render(c)


 ...in which case the contents of you template file 'app/new_tag_email.txt'
could be enlightening.

Karen

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



GeoDjango

2007-11-03 Thread Jeffrey Johnson

Hi Everyone,

Is there a specific group or list for users of GeoDjango. We just
started an implementation, and I want to know where I can turn to for
help and/or help others out.

Thanks,

Jeff

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Newforms validation error not displayed:

2007-11-03 Thread Karen Tracey
On 11/3/07, crybaby <[EMAIL PROTECTED]> wrote:
>
>
> Validation checks the number entered by the user is between the range
> and this check is done in "def clean_guess_the_number".  This is not
> working.
>
> Seems like "def clean_guess_the_number(self)" never executed when you
> post the data and when form.is_valid() called.


Seems like?  How do you know it is not being called instead of its code is
not doing what you expect?  Try  putting a print statement in it, I'll bet
you'll see it is being called.  For more enlightenment, set yourself up with
a debugging environment that lets you set breakpoints and step through the
code whilst showing you the values of variables, etc.  If you are going to
be doing anything beyond trivial bits of programming it is really worth the
investment in time to learn how to debug.

Karen

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread omat

Typo: "str" in "isinstance(str, unicode)" should be "sbj"



On 3 Kasım, 15:13, omat <[EMAIL PROTECTED]> wrote:
> Here is the traceback:
>
> Traceback (most recent call last):
>   File "update_feeds.py", line 171, in process_feed
> process_entry(entry, fpf, feed, postdict)
>   File "update_feeds.py", line 119, in process_entry
> feed.tags))
>   File "/srv/django/common/tagging/managers.py", line 39, in
> update_tags
> tag = self.create(name = tag_name, slug = tag_slug)
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> manager.py", line 75, in create
> return self.get_query_set().create(**kwargs)
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> query.py", line 273, in create
> obj.save()
>   File "/srv/django/common/tagging/models.py", line 35, in save
> super(Tag, self).save()
>   File "/usr/local/lib/python2.5/site-packages/django/db/models/
> base.py", line 271, in save
> instance=self, created=(not record_exists))
>   File "/usr/local/lib/python2.5/site-packages/django/dispatch/
> dispatcher.py", line 360, in send
> **named
>   File "/usr/local/lib/python2.5/site-packages/django/dispatch/
> robustapply.py", line 47, in robustApply
> return receiver(*arguments, **named)
>   File "/srv/django/sites/haberrobotu/aggregator/models.py", line 199,
> in mail_new_tag
> msg = t.render(c)
>   File "/usr/local/lib/python2.5/site-packages/django/template/
> __init__.py", line 174, in render
> return self.nodelist.render(context)
>   File "/usr/local/lib/python2.5/site-packages/django/template/
> __init__.py", line 795, in render
> return ''.join([force_unicode(b) for b in bits])
>   File "/usr/local/lib/python2.5/site-packages/django/utils/
> encoding.py", line 38, in force_unicode
> s = unicode(s)
> UnicodeDecodeError: 'ascii' codec can't decode byte 0xc4 in position
> 3: ordinal not in range(128)
>
> "repr(sbj)"= str: u'new tag: Arkada\u015flar\u0131n'
>
> isinstance(instance.name, unicode), isinstance(str, unicode) and
> isinstance(msg, unicode)  all return True.
>
> Thanks...
>
> On 3 Kasım, 14:04, Malcolm Tredinnick <[EMAIL PROTECTED]>
> wrote:
>
> > On Sat, 2007-11-03 at 07:01 -0500, James Bennett wrote:
> > > On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
> > > > sbj = u'new tag: %s' % (instance.name)
>
> > > You might want to read this:
>
> > >http://code.djangoproject.com/wiki/UnicodeBranch#PortingApplicationsT...
>
> > > There's a warning about using this style of string interpolation under
> > > Python 2.3 which may be relevant to your situation.
>
> > That shouldn't be an issue here. The instance.name attribute will
> > (should, at least) be a normal Python unicode string.
>
> > Malcolm
>
> > --
> > Remember that you are unique. Just like everyone 
> > else.http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread omat

Here is the traceback:

Traceback (most recent call last):
  File "update_feeds.py", line 171, in process_feed
process_entry(entry, fpf, feed, postdict)
  File "update_feeds.py", line 119, in process_entry
feed.tags))
  File "/srv/django/common/tagging/managers.py", line 39, in
update_tags
tag = self.create(name = tag_name, slug = tag_slug)
  File "/usr/local/lib/python2.5/site-packages/django/db/models/
manager.py", line 75, in create
return self.get_query_set().create(**kwargs)
  File "/usr/local/lib/python2.5/site-packages/django/db/models/
query.py", line 273, in create
obj.save()
  File "/srv/django/common/tagging/models.py", line 35, in save
super(Tag, self).save()
  File "/usr/local/lib/python2.5/site-packages/django/db/models/
base.py", line 271, in save
instance=self, created=(not record_exists))
  File "/usr/local/lib/python2.5/site-packages/django/dispatch/
dispatcher.py", line 360, in send
**named
  File "/usr/local/lib/python2.5/site-packages/django/dispatch/
robustapply.py", line 47, in robustApply
return receiver(*arguments, **named)
  File "/srv/django/sites/haberrobotu/aggregator/models.py", line 199,
in mail_new_tag
msg = t.render(c)
  File "/usr/local/lib/python2.5/site-packages/django/template/
__init__.py", line 174, in render
return self.nodelist.render(context)
  File "/usr/local/lib/python2.5/site-packages/django/template/
__init__.py", line 795, in render
return ''.join([force_unicode(b) for b in bits])
  File "/usr/local/lib/python2.5/site-packages/django/utils/
encoding.py", line 38, in force_unicode
s = unicode(s)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc4 in position
3: ordinal not in range(128)


"repr(sbj)"= str: u'new tag: Arkada\u015flar\u0131n'

isinstance(instance.name, unicode), isinstance(str, unicode) and
isinstance(msg, unicode)  all return True.


Thanks...


On 3 Kasım, 14:04, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
> On Sat, 2007-11-03 at 07:01 -0500, James Bennett wrote:
> > On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
> > > sbj = u'new tag: %s' % (instance.name)
>
> > You might want to read this:
>
> >http://code.djangoproject.com/wiki/UnicodeBranch#PortingApplicationsT...
>
> > There's a warning about using this style of string interpolation under
> > Python 2.3 which may be relevant to your situation.
>
> That shouldn't be an issue here. The instance.name attribute will
> (should, at least) be a normal Python unicode string.
>
> Malcolm
>
> --
> Remember that you are unique. Just like everyone 
> else.http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Newforms validation error not displayed:

2007-11-03 Thread crybaby

Validation checks the number entered by the user is between the range
and this check is done in "def clean_guess_the_number".  This is not
working.

Seems like "def clean_guess_the_number(self)" never executed when you
post the data and when form.is_valid() called.

class GuessForm(forms.Form):

guess_the_number= forms.RegexField(r'^\d+$',
max_length = 10,
min_length = 1,
error_message = 'Must be 1-10 numeric
characters.'
)

def __init__(self, *args, **kwargs):
try:
self.id = kwargs.pop('id')

except:
pass

super(GuessForm, self).__init__(*args, **kwargs)

number = get_object_or_404(Number,
pk=self.id
)

increment = Increment.objects.all()
self.max_number = number.value + increment[0].value
self.min_number = number.value + increment[1].value

def clean_guess_the_number(self):
if self.max_number > self.cleaned_data['guess_the_number']:
raise forms.ValidationError('Your Number have to be lower
or equal to Max Number.')
if self.min_number < self.cleaned_data['guess_the_number']:
raise forms.ValidationError('Your Number have to be higer
or equal to Min Number.')
return self.cleaned_data['guess_the_number']

html:

Guess The Number( Number must be
between {{form.min_number}} and {{form.max_number}}):
{{ form.guess_the_number }}
{% if form.guess_the_number.errors %}***
{{ form.guess_the_number.errors|join:", " }}{% endif %}



--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 07:01 -0500, James Bennett wrote:
> On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
> > sbj = u'new tag: %s' % (instance.name)
> 
> You might want to read this:
> 
> http://code.djangoproject.com/wiki/UnicodeBranch#PortingApplicationsTheQuickChecklist
> 
> There's a warning about using this style of string interpolation under
> Python 2.3 which may be relevant to your situation.

That shouldn't be an issue here. The instance.name attribute will
(should, at least) be a normal Python unicode string.

Malcolm

-- 
Remember that you are unique. Just like everyone else. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 11:41 +, omat wrote:
> Hi All,
> 
> I have a function that sends mail to managers every time a new tag is
> created, connected to post_save of a Tag model via dispatcher.
> 
> def mail_new_tag(instance):
> t = loader.get_template('app/new_tag_email.txt')
> c = Context({'tag': instance})
> msg = t.render(c)
> sbj = u'new tag: %s' % (instance.name)
> mail_managers(subject=sbj, message=msg)
> 
> This works fine when the instance.name is all ascii, but when it
> contains non-ascii characters, a UnicodeDecodeError is raised.
> 
> When debugging, I see that instance.name is a unicode object, but,
> "sbj" cannot be rendered.

Where is it raising the exception? Which line of code in particular. You
should be able to pass Unicode objects to the mail code (in fact, in
[1], I do just that for pretty much all the fields -- that was generated
using Django's email code).

Reading the code, I can't see anything special about mail_managers()
that would cause problems. So if you can past the full traceback that
will probably help work out what might be going on.

Also, what is repr(sbj) just before you call mail_managers()?

[1] http://www.flickr.com/photos/malcolmtredinnick/503502258/

Regards,
Malcolm


-- 
The early bird may get the worm, but the second mouse gets the cheese. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: encoding problem in instance signal / dispatcher?

2007-11-03 Thread James Bennett

On 11/3/07, omat <[EMAIL PROTECTED]> wrote:
> sbj = u'new tag: %s' % (instance.name)

You might want to read this:

http://code.djangoproject.com/wiki/UnicodeBranch#PortingApplicationsTheQuickChecklist

There's a warning about using this style of string interpolation under
Python 2.3 which may be relevant to your situation.

-- 
"Bureaucrat Conrad, you are technically correct -- the best kind of correct."

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



encoding problem in instance signal / dispatcher?

2007-11-03 Thread omat

Hi All,

I have a function that sends mail to managers every time a new tag is
created, connected to post_save of a Tag model via dispatcher.

def mail_new_tag(instance):
t = loader.get_template('app/new_tag_email.txt')
c = Context({'tag': instance})
msg = t.render(c)
sbj = u'new tag: %s' % (instance.name)
mail_managers(subject=sbj, message=msg)

This works fine when the instance.name is all ascii, but when it
contains non-ascii characters, a UnicodeDecodeError is raised.

When debugging, I see that instance.name is a unicode object, but,
"sbj" cannot be rendered.

Thanks for any help,
oMat


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: how to select current user's object in the forms.py,thanks!

2007-11-03 Thread Malcolm Tredinnick

On Sat, 2007-11-03 at 03:55 -0700, [EMAIL PROTECTED] wrote:
> i want to filter some object related to current logined user in
> forms.py, the key code is:
>  choices = [(str(x.id), x.name) for x in
> Category.objects.filter(user__id__exact=request.user.id)]
> 
> but i got this error:
> Exception Type:   NameError
> Exception Value:  global name 'request' is not defined
> Exception Location:   d:\django\mysite\blog\forms.py in __init__, line
> 16
> 
> my question is how to select current user's object in the
> forms.py,thanks!

You are only going to have access to 'request' inside your view function
and it is a local variable at that point. So you could create a method
on your CreateEntryForm class that took 'request' as an argument and set
up the category choices. Then something like this would work in your
view:

def my_view(request, ):
form = CreateEntryForm()
form.setup_choices(request)
...

Alternatively, you could do some fancy juggling of kwargs in
CreateEntryForm.__init__ so that you can pass 'request' into the
__init__ method (pop the 'request' keyword argument from **kwargs before
passing it up to the super's __init__ method.

Regards,
Malcolm

> 
> full code as below:
> 
> class CreateEntryForm(forms.Form):
> def __init__( self, *args, **kwargs ):
>super( CreateEntryForm, self ).__init__( *args, **kwargs )
>self.fields['category'] = forms.ChoiceField(
> choices = [(str(x.id), x.name) for x in
> Category.objects.filter(user__id__exact=request.user.id)]
> )
> 
> 
>   category   = forms.CharField(max_length=80, label=_('Category'))
>   title  = forms.CharField(max_length=80, label=_('Title'))
> 
> 
> > 
> 
-- 
Everything is _not_ based on faith... take my word for it. 
http://www.pointy-stick.com/blog/


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



how to select current user's object in the forms.py,thanks!

2007-11-03 Thread [EMAIL PROTECTED]

i want to filter some object related to current logined user in
forms.py, the key code is:
 choices = [(str(x.id), x.name) for x in
Category.objects.filter(user__id__exact=request.user.id)]

but i got this error:
Exception Type: NameError
Exception Value:global name 'request' is not defined
Exception Location: d:\django\mysite\blog\forms.py in __init__, line
16

my question is how to select current user's object in the
forms.py,thanks!

full code as below:

class CreateEntryForm(forms.Form):
def __init__( self, *args, **kwargs ):
   super( CreateEntryForm, self ).__init__( *args, **kwargs )
   self.fields['category'] = forms.ChoiceField(
choices = [(str(x.id), x.name) for x in
Category.objects.filter(user__id__exact=request.user.id)]
)


category   = forms.CharField(max_length=80, label=_('Category'))
title  = forms.CharField(max_length=80, label=_('Title'))


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Is there a simpler solution?

2007-11-03 Thread Akaroa

RajeshD wrote:
>
> On Nov 2, 4:05 pm, "Ramdas S" <[EMAIL PROTECTED]> wrote:
>   
>> Hi,
>>
>> I am sure there is something absolutely simple, but I guess I am missing it.
>> I need to display some content/ news articles over 3 columns on a home page,
>> where the order is latest stories coming in top rows from left to right. If
>> you've seen web sites like theinquirer.net or theregister.co.uk you will
>> understand what I mean!
>>
>> I was searching for a solution that is a lot simple, like a template tag
>> which will let me implement this easily. But I have not found one.
>> 
>
> Have you considered using plain CSS floating boxes?
>
> Here's some hastily put together snippet:
>
> Your template:
> --
>
> {% for a in articles %}
>
>   render article a here
>
>   {% if forloop.counter1|divisibleby:"3" %}
>  
>   {% endif %}
> {% endfor %}
>
> Your CSS:
> --
>
> .article {width:32%;float:left;overflow:hidden;}
> .clear_all {clear:both;}
>
> Experiment further to make things look nicer from here ;)
>
> -Rajesh D
>   
I'd change "clear_all" to "clear-all" if I was you.
I've had trouble with some older versions of Netscape choking on 
underscores.

Cheers,
Dean


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Re: Is there a simpler solution?

2007-11-03 Thread Ramdas S
Thanks,

I shall try this

Ramdas

On 11/3/07, RajeshD <[EMAIL PROTECTED]> wrote:
>
>
> Disregard the overflow:hidden piece in my CSS above. It's not needed:
>
>
> .article {width:32%;float:left;}
> .clear_all {clear:both;}
>
>
> >
>

--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---



Design Contest by NASA...

2007-11-03 Thread Designer

Hi,

Please click the link to find the concept for Earthquake resistant
building...

http://www.createthefuturecontest.com/pages/view/entriesdetail.html?entryID=533

Thanks

Anand


--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---