distinct doesn't work after sorting by reverse foreign key

[ryan]

For instance, I have two objects:
1) Blog
2) Entry with a ForeignKey to Blog, and a Date field titled DateAdded

Now, I want to list all my blogs, ordering by the most recent added
entry.

I tried:
blog.objects.all().sort_by('entry__dateAdded').distinct()

But the result was that is a given blog had more than 1 entry, then it
shows up in the resulting queryset more than once. It appears distinct
() is doing nothing in this case.

are more knowledgeable minds willing to share their insight on this
problem?

Thanks!

--~--~-~--~~~---~--~~
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: distinct doesn't work after sorting by reverse foreign key

[ryan]

thanks for the thorough reply! I now understand how to proceed.

However, I'm not convinced that distinct shouldn't do this for me...
but I understand this is a philosophical point, and I put my trust in
the django developers to get it right (just like I do with the
american government... )

On Dec 4, 9:24 pm, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
> On Thu, 2008-12-04 at 19:09 -0800, ryan wrote:
> > For instance, I have two objects:
> > 1) Blog
> > 2) Entry with a ForeignKey to Blog, and a Date field titled DateAdded
>
> > Now, I want to list all my blogs, ordering by the most recent added
> > entry.
>
> > I tried:
> > blog.objects.all().sort_by('entry__dateAdded').distinct()
>
> > But the result was that is a given blog had more than 1 entry, then it
> > shows up in the resulting queryset more than once. It appears distinct
> > () is doing nothing in this case.
>
> The distinct() call is working correctly (you're right that it is
> effectively doing nothing, wrong to say it "doesn't work"). The problem
> is that you are ordering by something that has multiple values and each
> value contributes one output row. There is nothing in your ordering
> statement to say that the database should only use the *maximum* value
> of dateAdded for each entry. The only reasonable way to interpret
> ordering by multi-valued fields is to treat each one as generating a
> different output row, since otherwise we might as well just pick one at
> random (the SQL specification agrees with that interpretation by the way
> -- which is why Django does it. Only MySQL allows effective ordering by
> columns not in the output list in this case and it's very non-standard).
>
> Refer to the massive callout box in the documentation for distinct() for
> a more SQL-specific 
> explanation:http://docs.djangoproject.com/en/dev/ref/models/querysets/#distinct
>
> You could achieve the effect you're after by using extra(select=...,
> order_by=...) to select the maximum date added value as an extra
> column.
>
> Alternatively (and this would require some fairly deep poking into Query
> internals, so probably not recommended) you could add some extra stuff
> to the where-clause to say "where dateAdded = (Select max(dateAdded)
> from ...)". Possible some extra(where="") usage could get you most
> of the way there on that one.
>
> Regards,
> Malcolm
--~--~-~--~~~---~--~~
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: distinct doesn't work after sorting by reverse foreign key

[ryan]

thanks for the thorough reply! I now understand how to proceed.

However, I'm not convinced that distinct shouldn't do this for me...
but I understand this is a philosophical point, and I put my trust in
the django developers to get it right (just like I do with the
american government... )

On Dec 4, 9:24 pm, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
> On Thu, 2008-12-04 at 19:09 -0800, ryan wrote:
> > For instance, I have two objects:
> > 1) Blog
> > 2) Entry with a ForeignKey to Blog, and a Date field titled DateAdded
>
> > Now, I want to list all my blogs, ordering by the most recent added
> > entry.
>
> > I tried:
> > blog.objects.all().sort_by('entry__dateAdded').distinct()
>
> > But the result was that is a given blog had more than 1 entry, then it
> > shows up in the resulting queryset more than once. It appears distinct
> > () is doing nothing in this case.
>
> The distinct() call is working correctly (you're right that it is
> effectively doing nothing, wrong to say it "doesn't work"). The problem
> is that you are ordering by something that has multiple values and each
> value contributes one output row. There is nothing in your ordering
> statement to say that the database should only use the *maximum* value
> of dateAdded for each entry. The only reasonable way to interpret
> ordering by multi-valued fields is to treat each one as generating a
> different output row, since otherwise we might as well just pick one at
> random (the SQL specification agrees with that interpretation by the way
> -- which is why Django does it. Only MySQL allows effective ordering by
> columns not in the output list in this case and it's very non-standard).
>
> Refer to the massive callout box in the documentation for distinct() for
> a more SQL-specific 
> explanation:http://docs.djangoproject.com/en/dev/ref/models/querysets/#distinct
>
> You could achieve the effect you're after by using extra(select=...,
> order_by=...) to select the maximum date added value as an extra
> column.
>
> Alternatively (and this would require some fairly deep poking into Query
> internals, so probably not recommended) you could add some extra stuff
> to the where-clause to say "where dateAdded = (Select max(dateAdded)
> from ...)". Possible some extra(where="") usage could get you most
> of the way there on that one.
>
> Regards,
> Malcolm
--~--~-~--~~~---~--~~
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: distinct doesn't work after sorting by reverse foreign key

[ryan]

There's no mind reading required. I have a queryset of Blog objects,
not entry objects (or SQL rows for that matter). When I call distinct,
I would expect a "distinct set of blog objects". Call it a strong
technical point if you prefer, but I still feel this is what the
common person would expect.

On Dec 4, 10:17 pm, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
> On Thu, 2008-12-04 at 20:15 -0800, ryan wrote:
> > thanks for the thorough reply! I now understand how to proceed.
>
> > However, I'm not convinced that distinct shouldn't do this for me...
>
> It cannot read your mind. How is distinct() meant to know which of the
> many multiple values it should use for ordering? Should it pick one at
> random? That will only lead to disappointment, I'm sure.
>
> > but I understand this is a philosophical point,
>
> No, it's a strong technical point. To pick one value, that distinguished
> value has to be well-defined (that is, there has to be a natural,
> unambiguous meaning of which value to pick, for all possible
> circumstances). This isn't the case here.
>
> Your trust is welcome (and, I'm sure, well-placed). But realise that
> this isn't simply an arbitrary opinion. It's a technical constraint.
>
> Regards,
> Malcolm
--~--~-~--~~~---~--~~
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: distinct doesn't work after sorting by reverse foreign key

[ryan]

I guess what I'm imagining is that distinct() would take the current
queryset, go from the top, and keep the first occurrence of each
distinct object in that set (users don't care about SQL rows, that's
why they're using django in the first place). Do you think this is NOT
what most users would expect from a distinct() function?

On Dec 4, 10:22 pm, ryan <[EMAIL PROTECTED]> wrote:
> There's no mind reading required. I have a queryset of Blog objects,
> not entry objects (or SQL rows for that matter). When I call distinct,
> I would expect a "distinct set of blog objects". Call it a strong
> technical point if you prefer, but I still feel this is what the
> common person would expect.
>
> On Dec 4, 10:17 pm, Malcolm Tredinnick <[EMAIL PROTECTED]>
> wrote:
>
> > On Thu, 2008-12-04 at 20:15 -0800, ryan wrote:
> > > thanks for the thorough reply! I now understand how to proceed.
>
> > > However, I'm not convinced that distinct shouldn't do this for me...
>
> > It cannot read your mind. How is distinct() meant to know which of the
> > many multiple values it should use for ordering? Should it pick one at
> > random? That will only lead to disappointment, I'm sure.
>
> > > but I understand this is a philosophical point,
>
> > No, it's a strong technical point. To pick one value, that distinguished
> > value has to be well-defined (that is, there has to be a natural,
> > unambiguous meaning of which value to pick, for all possible
> > circumstances). This isn't the case here.
>
> > Your trust is welcome (and, I'm sure, well-placed). But realise that
> > this isn't simply an arbitrary opinion. It's a technical constraint.
>
> > Regards,
> > Malcolm
--~--~-~--~~~---~--~~
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: AlreadyRegistered Exception after newforms-admin merge

[ryan]

Good catch. Thank you

On Aug 3, 10:14 pm, "Pedro Valente" <[EMAIL PROTECTED]> wrote:
> I'm not sure if it's your case, but I got the AlreadyRegistered errors
> because before the merge I used the NFA branch and had an admin import
> inside __init__.py (not needed anymore).
>
> I had forgotten about that, and after I got rid of if my app worked fine.
>
> Pedro Valente
>
> On Sun, Aug 3, 2008 at 14:19, Brian Morton <[EMAIL PROTECTED]> wrote:
>
> > As my original post states, my admin registration calls are all in
> > admin.py for each app.  No admin declarations or registrations are
> > going on in my models.
>
> > On Aug 3, 5:55 am, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
> > wrote:
> > > The correct way to handle it is to put all your admin stuff inside
> > > admin.py for each application and than follow these directions:
> >http://www.djangoproject.com/documentation/admin/#hooking-adminsite-i...
>
> > > On Aug 3, 3:32 am, Ludwig <[EMAIL PROTECTED]> wrote:
>
> > > > I encountered the same exception -- it is triggered when, assuming your
> > > > admin registration takes place in models.py, is imported by something
> > else,
> > > > e.g. template tags, inclusion into another model.
>
> > > > A try/except block around the admin registration will suppress this.
>
> > > > While I consider this an inelegant hack, I am not sufficiently steeped
> > in
> > > > the logic of the admin registration to say that such multiple
> > registration
> > > > should or should not trigger an exception. Maybe the developers could
> > > > suggest the right pattern for this.
>
> > > > HTH
>
> > > > Ludwig
>
> > > > 2008/8/2 Brian Morton <[EMAIL PROTECTED]>
>
> > > > > Using revision 8194 from TRUNK (the latest as of writing this post),
> > I
> > > > > am encountering AlreadyRegistered while starting my application.  I
> > > > > have all ModelAdmin definitions in admin.py for my two applications.
> > > > > App2's models file references the app1's model with an import (like
> > in
> > > > >http://code.djangoproject.com/ticket/6776).  However, I am still
> > > > > getting exceptions about each one of my models from app1 being
> > already
> > > > > registered.  If I comment out the admin.site.register lines in
> > > > > app1.admin, only about 1/3 of my models appear in the admin.
>
> > > > > Any ideas what might be causing 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
-~--~~~~--~~--~--~---

GEOIP_LIBRARY_PATH in settings.py

[ryan]

Is this correct?:

GEOIP_LIBRARY_PATH = '/home/USERNAME/geoip/lib/libGeoIP.a'
--~--~-~--~~~---~--~~
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: ViewDoesNotExist No module. where is it being called, I don't want it to.

[ryan]

"I do have A feild in my times
model named user"

Is it a ForeignKey to Django's User model?
If so you need to import that at the top of your models.py file

-ryan


On Oct 17, 3:42 pm, KillaBee <[EMAIL PROTECTED]>
wrote:
> I keep getting this error that says ViewDoesNotExist:  Could not
> import intranet.timesheets.views. Error was: No module named User.  No
> i do not have a User module, I don't want nor need one.  How do I find
> out Where it is being call at?  I got this error today, and I have
> been trying to get the users logged in.  I do have A feild in my times
> model named user, and Have been getting errors about not know what
> username is.  not sure where it's coming from.
> View:  Because It is an error in the view I was thinking that it was
> in the view.py or a template.  no matter want the url or view is bing
> used it still comes up.  Does that sound like I am looking in the
> right spot?
>
> error                      http://dpaste.com/85152/
> url                          http://dpaste.com/85153/
> views                      http://dpaste.com/85154/
> models                    http://dpaste.com/85158/
>
> Hope you can understand 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: ViewDoesNotExist No module. where is it being called, I don't want it to.

[ryan]

no, i just checked yr code.  that's not it.

On Oct 17, 5:25 pm, ryan <[EMAIL PROTECTED]> wrote:
> "I do have A feild in my times
> model named user"
>
> Is it a ForeignKey to Django's User model?
> If so you need to import that at the top of your models.py file
>
> -ryan
>
> On Oct 17, 3:42 pm, KillaBee <[EMAIL PROTECTED]>
> wrote:
>
> > I keep getting this error that says ViewDoesNotExist:  Could not
> > import intranet.timesheets.views. Error was: No module named User.  No
> > i do not have a User module, I don't want nor need one.  How do I find
> > out Where it is being call at?  I got this error today, and I have
> > been trying to get the users logged in.  I do have A feild in my times
> > model named user, and Have been getting errors about not know what
> > username is.  not sure where it's coming from.
> > View:  Because It is an error in the view I was thinking that it was
> > in the view.py or a template.  no matter want the url or view is bing
> > used it still comes up.  Does that sound like I am looking in the
> > right spot?
>
> > error                      http://dpaste.com/85152/
> > url                          http://dpaste.com/85153/
> > views                      http://dpaste.com/85154/
> > models                    http://dpaste.com/85158/
>
> > Hope you can understand 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
-~--~~~~--~~--~--~---

upload_to ignored outside of admin?

[ryan]

class Creative(models.Model):
creative = models.FileField(upload_to='creative')
element = models.ForeignKey(Element)

#views.py(1)
attachment = request.FILES[attachment_num]
element.creative_set.create(creative=attachment)

#views.py(2)
attachment = request.FILES[attachment_num]
Creative.objects.create(creative=attachment,element=new_element)

Both of these make an entry in the database 'filename.txt'
But doing so from the admin results in 'creative/filename.txt' which
is what i want

Can anyone spot an error?
Do I have to pass upload_to to create function in views.py?
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

model subclasses

[ryan]

Let's say I have a Polygon model

class Polygon(Models.model):
side_length = models.IntegerField()

Depending on the view I'm in, I want to calculate the perimeter of the
polygons as a square, or as a triangle (or anything other polygon).

What I'd like to do is somehow create a subclass of polygon (that
doesn't create a new database table), such as a Square or Triangle,
and then after I pull a Polygon queryset out of the database, I'd like
to cast all the Polygons to either Squares or Triangles, and then I
can call the perimeter() method for that object:

class Square(Polygon)
def perimeter(self): return 3*self.side_length
class Meta: abstract=True

But what I've ended up doing is added a traingle_perimeter() and a
square_perimeter() method to the Polygon object, and it just doesn't
seem right. Is there a preferred pythonic / djangoistic way to do
this?

Thanks!
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

differences in seek method for TemporaryUploadedFile & InMemoryUploadedFile

[ryan]

In the following code, the second loop of "for row in csv_reader:" is
empty when dealing with an InMemoryUploadedFile object.  Setting
FILE_UPLOAD_MAX_MEMORY_SIZE equal to zero forces the use of a
TemporaryUploadedFile object and solves the problem.  It seems that
seek(0) doesn't work with an InMemoryUploadedFile and csv.reader.
Anyone know why?

if request.method == 'POST':
dedupe_csv_form = DedupeCSVForm(request.POST,request.FILES)
if dedupe_csv_form.is_valid():
relevant_column = dedupe_csv_form.cleaned_data
['duplicates_column']
csv_file = request.FILES['csv_file']
csv_reader = csv.reader(csv_file)

items_in_relevant_column = []
duplicates_in_relevant_column = []
col = int(relevant_column)-1

for row in csv_reader:
items_in_relevant_column.append(row[col].strip())
items_in_relevant_column.sort()

for (i, item) in enumerate(items_in_relevant_column):
if i == 0:
continue
else:
last_item = items_in_relevant_column[i-1]
if item == last_item:
duplicates_in_relevant_column.append(item)
csv_file.seek(0)
response = HttpResponse(mimetype='text/csv')
response['Content-Disposition'] = 'attachment;
filename=output.csv'
csv_writer = csv.writer(response)

for row in csv_reader:
item = row[col].strip()
if item in duplicates_in_relevant_column:
continue
else:
csv_writer.writerow(row)
return response
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: differences in seek method for TemporaryUploadedFile & InMemoryUploadedFile

[ryan]

answering my own question, i think this has something to do with an
exhausted iterator

-ryan

On Jul 20, 12:39 pm, ryan  wrote:
> In the following code, the second loop of "for row in csv_reader:" is
> empty when dealing with an InMemoryUploadedFile object.  Setting
> FILE_UPLOAD_MAX_MEMORY_SIZE equal to zero forces the use of a
> TemporaryUploadedFile object and solves the problem.  It seems that
> seek(0) doesn't work with an InMemoryUploadedFile and csv.reader.
> Anyone know why?
>
>     if request.method == 'POST':
>         dedupe_csv_form = DedupeCSVForm(request.POST,request.FILES)
>         if dedupe_csv_form.is_valid():
>             relevant_column = dedupe_csv_form.cleaned_data
> ['duplicates_column']
>             csv_file = request.FILES['csv_file']
>             csv_reader = csv.reader(csv_file)
>
>             items_in_relevant_column = []
>             duplicates_in_relevant_column = []
>             col = int(relevant_column)-1
>
>             for row in csv_reader:
>                 items_in_relevant_column.append(row[col].strip())
>             items_in_relevant_column.sort()
>
>             for (i, item) in enumerate(items_in_relevant_column):
>                 if i == 0:
>                     continue
>                 else:
>                     last_item = items_in_relevant_column[i-1]
>                 if item == last_item:
>                     duplicates_in_relevant_column.append(item)
>             csv_file.seek(0)
>             response = HttpResponse(mimetype='text/csv')
>             response['Content-Disposition'] = 'attachment;
> filename=output.csv'
>             csv_writer = csv.writer(response)
>
>             for row in csv_reader:
>                 item = row[col].strip()
>                 if item in duplicates_in_relevant_column:
>                     continue
>                 else:
>                     csv_writer.writerow(row)
>             return response
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

django-facebookconnect

[Ryan]

Hey folks,

I posted code for a facebook connect django app on google code. I
refactored code from the News Mixer application to use for another
project, and figured it was useful enough that others might want it.

http://code.google.com/p/django-facebookconnect/
http://code.google.com/p/newsmixer/

The app is pretty simple, its just built on top of django.contrib.auth
and pyfacebook. It works along side django-registration and regular
django user accounts. New facebook users can link an existing django
user account when they log in or just use a user account that gets
automatically generated.

News Mixer used the facebook API extensively. So this app handles a
lot of the BS headache problems that we ran into working with facebook
connect. It uses caching extensively to minimize API calls to
facebook, handles session timeouts and a few other things.

Check it out, tell me what you think.

--
Ryan Mark
http://ryan-mark.com
847 691 8271

--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

inline model's choices field empty in admin

[ryan]

When I edit a User in the admin, the sales_team and user_class
dropdowns are empty.

If anyone can point out my error or point me to the django core code
that ignores the choices, I would greatly appreciate it.

#models.py
SALES_TEAM_CHOICES = enumerate(('CLS','CCS','TPS'))
USER_CLASS_CHOICES = enumerate(('DOM','DSK','DFA','DTM'))

class UserProfile(models.Model):
user = models.ForeignKey(User, unique=True)
sales_team = models.IntegerField(choices=SALES_TEAM_CHOICES)
user_class = models.IntegerField(choices=USER_CLASS_CHOICES)

#admin.py
class UserProfileInline(admin.StackedInline):
model = UserProfile
max_num = 1

class CustomUserAdmin(UserAdmin):
inlines = [UserProfileInline,]

admin.site.unregister(User)
admin.site.register(User, CustomUserAdmin)



--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: inline model's choices field empty in admin

[ryan]

This additional model, which uses the same choices is emptying the
choices dropdown of both User with inline UserProfile and UserProfile
itself.  Add it prior to UserProfile in models.py of your test app:

class Person(models.Model):
sales_team = models.IntegerField(choices=SALES_TEAM_CHOICES)
user_class = models.IntegerField(choices=USER_CLASS_CHOICES)
last_name = models.CharField(max_length=40)
first_name = models.CharField(max_length=30)

This behavior is not mentioned as a caveat in
http://docs.djangoproject.com/en/dev/ref/models/fields/#choices

ryan

On Jun 2, 11:22 am, Karen Tracey  wrote:
> On Tue, Jun 2, 2009 at 10:59 AM, ryan  wrote:
>
> > When I edit a User in the admin, the sales_team and user_class
> > dropdowns are empty.
>
> > If anyone can point out my error or point me to the django core code
> > that ignores the choices, I would greatly appreciate it.
>
> > [snip code]
>
> I cannot recreate this with either Django 1.0.2 or current trunk.  When I
> cut and paste the models/admin defs you show into a test app and then edit a
> user in admin, the two profile dropdowns are populated with an empty choice
> (all dashes) plus the choices you have specified.  There must be something
> beyond what you have shown here that is coming into play.
>
> 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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: inline model's choices field empty in admin

[ryan]

Thank you sir.  I got this from "Python Web Dev. w/ Django".
An unforseen side effect.

ryan

On Jun 2, 12:05 pm, Daniel Roseman 
wrote:
> On Jun 2, 4:53 pm, ryan  wrote:
>
> > This additional model, which uses the same choices is emptying the
> > choices dropdown of both User with inline UserProfile and UserProfile
> > itself.  Add it prior to UserProfile in models.py of your test app:
>
> > class Person(models.Model):
> >     sales_team = models.IntegerField(choices=SALES_TEAM_CHOICES)
> >     user_class = models.IntegerField(choices=USER_CLASS_CHOICES)
> >     last_name = models.CharField(max_length=40)
> >     first_name = models.CharField(max_length=30)
>
> > This behavior is not mentioned as a caveat 
> > inhttp://docs.djangoproject.com/en/dev/ref/models/fields/#choices
>
> > ryan
>
> Because enumerate() is an iterator. So it is consumed the first time
> it is run.
>
> You could call list() on it to flatten it:
> SALES_TEAM_CHOICES = list(enumerate(('CLS','CCS','TPS')))
>
> I've never heard of anyone doing this, which is presumably why it's
> not mentioned in the documentation...
> --
> DR.
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

__all__in instead of __in

[ryan]

#models.py
class Station(models.Model):
station_name = models.CharField(max_length=20)

class Order(models.Model):
station = models.ManyToManyField(Station, blank=True, null=True)

class UserProfile(models.Model):
user = models.ForeignKey(User, unique=True)
station = models.ManyToManyField(Station, blank=True, null=True)

>>> from myapp.models import Station, Order
>>> from django.contrib.auth.models import User
>>> u = User.objects.get(username__exact='ryan')
>>> user_stations = u.get_profile().station.all().values_list('pk', flat=True)
>>> user_stations
[1L, 2L, 3L]
>>> o = Order.objects.filter(station__in=user_stations)
>>> o
[Too many orders!]

Order.objects.filter(station__in=user_stations) returns any order
where any of its stations are in user_stations.  What I need is only
those orders where ALL of its stations are in user_stations.  Is there
a way to specify __all__in: Order.objects.filter
(station__all__in=user_stations)

-ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: __all__in instead of __in

[ryan]

answering my own question, seems like:

Order.objects.filter(Q(station=1), Q(station=2), Q(station=3))

-ryan

On Jun 4, 2:29 pm, ryan  wrote:
> #models.py
> class Station(models.Model):
>     station_name = models.CharField(max_length=20)
>
> class Order(models.Model):
>     station = models.ManyToManyField(Station, blank=True, null=True)
>
> class UserProfile(models.Model):
>     user = models.ForeignKey(User, unique=True)
>     station = models.ManyToManyField(Station, blank=True, null=True)
>
> >>> from myapp.models import Station, Order
> >>> from django.contrib.auth.models import User
> >>> u = User.objects.get(username__exact='ryan')
> >>> user_stations = u.get_profile().station.all().values_list('pk', flat=True)
> >>> user_stations
> [1L, 2L, 3L]
> >>> o = Order.objects.filter(station__in=user_stations)
> >>> o
>
> [Too many orders!]
>
> Order.objects.filter(station__in=user_stations) returns any order
> where any of its stations are in user_stations.  What I need is only
> those orders where ALL of its stations are in user_stations.  Is there
> a way to specify __all__in: Order.objects.filter
> (station__all__in=user_stations)
>
> -ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: __all__in instead of __in

[ryan]

actually, you need to do something like this:

user_stations = u.get_profile().station.all().values_list('pk',
flat=True)
excluded_stations = Station.objects.exclude(id__in=user_station_pks)
all_terrestrial_active_orders = Order.objects.filter(order_type=0,
in_market=True, end_date__gte=today)
terrestrial_active_orders = all_terrestrial_active_orders.exclude
(station__in=excluded_terrestrial_stations)

On Jun 4, 3:05 pm, ryan  wrote:
> answering my own question, seems like:
>
> Order.objects.filter(Q(station=1), Q(station=2), Q(station=3))
>
> -ryan
>
> On Jun 4, 2:29 pm, ryan  wrote:
>
> > #models.py
> > class Station(models.Model):
> >     station_name = models.CharField(max_length=20)
>
> > class Order(models.Model):
> >     station = models.ManyToManyField(Station, blank=True, null=True)
>
> > class UserProfile(models.Model):
> >     user = models.ForeignKey(User, unique=True)
> >     station = models.ManyToManyField(Station, blank=True, null=True)
>
> > >>> from myapp.models import Station, Order
> > >>> from django.contrib.auth.models import User
> > >>> u = User.objects.get(username__exact='ryan')
> > >>> user_stations = u.get_profile().station.all().values_list('pk', 
> > >>> flat=True)
> > >>> user_stations
> > [1L, 2L, 3L]
> > >>> o = Order.objects.filter(station__in=user_stations)
> > >>> o
>
> > [Too many orders!]
>
> > Order.objects.filter(station__in=user_stations) returns any order
> > where any of its stations are in user_stations.  What I need is only
> > those orders where ALL of its stations are in user_stations.  Is there
> > a way to specify __all__in: Order.objects.filter
> > (station__all__in=user_stations)
>
> > -ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Handling IntegrityError on transaction middleware commit

[Ryan]

I had an IntegrityError come up on transaction commit (using postgres)
today on a production site -- trying to insert an invalid foreign key
value (an admin deleted something she shouldn't have).  The problem
was that instead of emailing me the exception like usual, django
logged the exception to my apache (using mod_wsgi) error log and then
that apache process was semi-broken for about half and hour, refusing
to recognize session cookies or log anyone in, but also not throwing
up any exceptions either.

I realize that I should be validating my data before saving it, but
I'll admit I don't guard against every eventuality.  My question is,
should I be doing something in my code to prevent this kind of poor
exception handling, or should I file a bug against django?  I would
certainly prefer the whole server not to break when the transaction
commit fails for whatever reason -- it should rollback the transaction
and give a 500 error page.

Here's the middleware tuple from my settings.py:

MIDDLEWARE_CLASSES = (
'django.middleware.common.CommonMiddleware',
'django.contrib.sessions.middleware.SessionMiddleware',
'django.contrib.auth.middleware.AuthenticationMiddleware',
'django.middleware.transaction.TransactionMiddleware',
)

Any thoughts?  Thanks.

--

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

override TabularInline get_formset - need to call super method?

[ryan]

Hi

I'm trying to change the extra value based on a related object's
attribute.
The relevant bit of code starts with:  ## begin alterations

This way of doing it works, but should I be calling super() at the
end?  if so, why?
I've been reading up on super a lot.


from admin.py

class HDChannelInline(admin.TabularInline):
model = HDChannel
extra = 0
template = 'admin/edit_inline/tabular.html'

def get_formset(self, request, obj=None, **kwargs):
"""Returns a BaseInlineFormSet class for use in admin add/
change views."""
if self.declared_fieldsets:
fields = flatten_fieldsets(self.declared_fieldsets)
else:
fields = None
if self.exclude is None:
exclude = []
else:
exclude = list(self.exclude)
defaults = {
"form": self.form,
"formset": self.formset,
"fk_name": self.fk_name,
"fields": fields,
"exclude": exclude + kwargs.get("exclude", []),
"formfield_callback": self.formfield_for_dbfield,
"extra": self.extra,
"max_num": self.max_num,
}
## begin alterations
if obj is not None:
if obj.asset_type == 'broadcast_radio':
defaults['extra'] = 2
## end alterations
defaults.update(kwargs)
return inlineformset_factory(self.parent_model, self.model,
**defaults)

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

Re: override TabularInline get_formset - need to call super method?

[ryan]

would this be preferable?

from admin.py

class HDChannelInline(admin.TabularInline):
model = HDChannel
extra = 0
template = 'admin/edit_inline/tabular.html'

def get_formset(self, request, obj=None, **kwargs):
if obj is not None:
if obj.asset_type == 'broadcast_radio':
kwargs['extra'] = 2
return super(HDChannelInline, self).get_formset(request, obj,
**kwargs)

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

Re: Ajax header not sent in all browsers

[Ryan]

If you simplify things down to something like the snippets below, does
the alert display 'Said it was NOT ajax' for you? It shouldn't (and
doesn't for me).

Also, do you have the Tamper Data Firefox add-on installed to validate
the headers being sent in? That could help narrow things down.

urls.py:

(r'^isajax/?$', 'some_site.views.is_ajax_worky'),

views.py:

def is_ajax_worky(request):
ajax_header = request.META.get('HTTP_X_REQUESTED_WITH', 'None')
if request.is_ajax():
return HttpResponse('Said it WAS ajax. header=' + ajax_header)
return HttpResponse('Said it was NOT ajax. header=' + ajax_header)

test.html:

This test requires javascript!
https://ajax.googleapis.com/ajax/
libs/jquery/1.4.2/jquery.min.js">

$(document).ready(function() {
$.get('/isajax', function(data) {
  alert(data);
});
});


On Jun 9, 2:14 am, David Escobar  wrote:
> Hi everyone,
> I'm using Django 1.1 with jQuery 1.4.2 and currently testing with the
> Django development server. Whenever I send an AJAX request with
> $.get(), the HTTP_X_REQUESTED_WITH header only gets sent with Chrome
> and Safari. It does not get sent with IE or Firefox. I've verified
> this by outputting the request.META keys to a text file for each one.
> Needless to say, my AJAX only works correctly with Chrome and Safari.
> Also, I'm using request.is_ajax() on the Django side.
>
> Does anyone know why this might be occurring? Could it be because I'm
> testing using the Django development server? It seems to me like this
> might be a jQuery issue, but after doing many online searches, I
> haven't seen anyone else run into this problem. Every documentation
> says that jQuery sends HTTP_X_REQUESTED WITH.
>
> Thanks in advance.

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

select_related removes broken Foreign Keys, but How?

[ryan]

I noticed this first in the Django Admin.  If you have 100 rows in
Table A, and 5 of those have a foreign key entry pointing to a
nonexistent item in Table B, rather than throwing an error, the admin
displays 95. This happens if you set
list_select_related = True, or if one of the list_display fields is a
ForeignKey. (http://docs.djangoproject.com/en/dev/ref/contrib/admin/
#django.contrib.admin.ModelAdmin.list_select_related)

So you can reproduce this by calling select_related.  But my question
is "How does this happen?".  What are the lines of code that remove
the 5 rows with broken foreign keys from the queryset?  I've been
digging through the Django codebase and can't find it.

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

Strange cookie error, only for some IE users?

[Ryan]

At a client site, I have occasionally had her customers complain that
they get cookie errors when they try to log in, even though they have
cookies enabled and they are able to log in to other sites.  Clearing
out their cookies doesn't seem to help.  Most (95-99%) users can log
in fine.  The affected ones seem to be all internet explorer users,
but those are the majority of the site's users, so that may just be a
coincidence.

I had basically dismissed this all as a few funny user browsers
malfunctioning, but tonight my client herself got the same error, and
was also unable to log in to the django admin, also with a cookie
error, so I figure there's actually a real problem somewhere, and it's
not just in my code.

A few days ago I added the SESSION_COOKIE_DOMAIN setting to
"..com" so users would be able to use the same cookie
for both the www..com and .com URLs,
thinking that might have been part of the problem, but that doesn't
seem to have helped.  Just now I cleared all the session rows out of
the database, to no avail.

Any ideas?  I'm hosting on webfaction, will post in their forums too.

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

Re: Strange cookie error, only for some IE users?

[Ryan]

Solved my own problem.  User's clocks were set ahead so the Internet
Explorer (pinnacle of stupid design) expired the cookie instantly.

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

How to set individual cache items so they don't timeout?

[Ryan]

I found this thread about how a timeout of 0 (= infinite) isn't passed
to memcached properly:

http://groups.google.com/group/django-developers/browse_thread/thread/a934ec625d35e942

The thread is from July, but the issue still exists in svn, and I
couldn't find a ticket for this issue.

How do I go about caching an item so it doesn't timeout if I don't
want to set the default timeout to zero?

Thanks,
Ryan

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

Re: How to set individual cache items so they don't timeout?

[Ryan]

I'll answer my own question:  The way to do this is probably to write
your own cache backend that is a wrapper around the django versions so
that timeout values of 0 are handled the way you'd like.  Johnny Cache
does this:

http://packages.python.org/johnny-cache/

"The custom backend setting enables a thin wrapper around Django’s
memcached (or locmem) cache class that allows cache times of “0”,
which memcached interprets as “forever” and locmem is patched to see
as forever."

Ryan

On Nov 24, 2:02 am, Ryan  wrote:
> I found this thread about how a timeout of 0 (= infinite) isn't passed
> to memcached properly:
>
> http://groups.google.com/group/django-developers/browse_thread/thread...
>
> The thread is from July, but the issue still exists in svn, and I
> couldn't find a ticket for this issue.
>
> How do I go about caching an item so it doesn't timeout if I don't
> want to set the default timeout to zero?
>
> Thanks,
> Ryan

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

Proxy model ContentType question

[Ryan]

In django/contrib/contenttypes/models.py (get_for_model), I noticed
that for a proxy model object, the contenttype being returned is of
the base concrete class since it sets opts to the _meta of
proxy_for_model .  If one uses a generic relation in the admin app,
you can set the generic contenttype to that of a proxy model class.  I
was just wondering if the contenttype get_for_model call is returning
the correct type (concrete base class)?

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

Re: Proxy model ContentType question

[Ryan]

As an update:

Since you can actually assign a generic relation in the admin app to a
proxy model object, when using a generic relation, the proxy objects
aren't returned because the lookup is using get_for_model which
returns an instance of the base class.  (I only figured this out after
having wrapped the ContentType get_by_natural_key call).  I'm
wondering what other implications there are if I continue to use the
get_by_natural_key as a way to get the proxy content type.


On Feb 16, 10:05 pm, Ryan  wrote:
> In django/contrib/contenttypes/models.py (get_for_model), I noticed
> that for a proxy model object, thecontenttypebeing returned is of
> the base concrete class since it sets opts to the _meta of
> proxy_for_model .  If one uses a generic relation in the admin app,
> you can set the genericcontenttypeto that of a proxy model class.  I
> was just wondering if thecontenttypeget_for_model call is returning
> the correct type (concrete base class)?

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

Re: Is there any way to custom group models in the admin panel?

[Ryan]

I am interested in this as well. I've searched for a way to organize
models in the admin panel but I haven't found anything. It seems
illogical to create multiple apps simply for the fact to separate them
in the admin panel.


--~--~-~--~~~---~--~~
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: Best Practices to Make your Apps Portable

[Ryan]

I like this setup. I will use this as a basis for my own upcoming
project.

Thanks.
RG

On Jul 25, 11:12 am, Sebastian Macias <[EMAIL PROTECTED]>
wrote:
> Thanks a lot for the feedback everyone.
>
> I have come up a perfect setup  and folder structure  (at least
> perfect for my needs) that will allow me to work on generic apps and
> project specific apps efficiently and just wanted to share it with
> everyone in case it can save a anyone a headache.
>
> *Folder structure for a django project*
>
> /var/django_root/my_project_name/
> urls.py
> settings.py
> apps/
> my_project_specific_app_1/
> my_project_specific_app_2/
> my_project_specific_app_3/
>
> *Folder structure for generic/portable apps*
>
> /var/django_root/shared/
> my_generic_portable_app_1/
> my_generic_portable_app_2/
> my_generic_portable_registration_app/
>
> *Development Setup*
>
> I added the following to the top  of my_project_name/settings.py so it
> appends the portable/generic apps folder to the python path.
>
> DEVELOPMENT = True
>
> if DEVELOPMENT:
> import sys
> sys.path.append('/var/django_root/shared')
>
> For extended convenience I symlinked my portable/generic apps folder
> to my django project so I can quickly make changes to my generic apps
> without having to go outside my django project folder structure
>
> ln -s `pwd`/var/django_root/shared /var/django_root/my_project_name/
> shared
>
> *Production Setup*
>
> My Apache conf file:
>
> 
>   ServerName championsound.local
>   ServerAlias *.championsound.local
>   SetHandler python-program
>   PythonPath "['/var/django_root', '/var/django_root/shared'] +
> sys.path"
>   PythonHandler django.core.handlers.modpython
>   SetEnv DJANGO_SETTINGS_MODULE championsound.settings
>   PythonDebug On
> 
>
> Note how '/var/django_root' and '/var/django_root/shared' are added to
> the PythonPath
>
> Enjoy it!
>
> Sebastian Macias
>
> On Jul 25, 6:19 am, "Adrian Holovaty" <[EMAIL PROTECTED]> wrote:
>
> > On 7/24/07, Sebastian Macias <[EMAIL PROTECTED]> wrote:
>
> > > My dilemma is... what is the point of having projects and apps if the
> > > applications created for my project won't be portable in other
> > > projects (becase the namespaces will always start with the project
> > > name). I can't just copy and app from one project to another.
>
> > Personally I never use "projects" -- they were just a quick thing we
> > made up just before we open-sourced Django, with the thinking being
> > "projects" would make it quicker and easier for people to get started.
> > They are *not* a good method to use if you want to distribute your
> > application, however.
>
> > We need better documentation about best practices to make apps portable.
>
> > Adrian
>
> > --
> > Adrian Holovaty
> > holovaty.com | djangoproject.com


--~--~-~--~~~---~--~~
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: choices/ and getting rid of the dashes?

[Ryan]

Use initial when calling your form class.

formClass = forms.form_for_model(Person)
form = formClass(initial={'gender': 'm'})

On Sep 6, 4:31 pm, Mark Green <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> This is my model:
>
> class Person(models.Model):
> GENDER_CHOICES = (
> ( 'm', 'Male' ),
> ( 'f', 'Female' ),
> )
> gender = models.CharField( blank=False, "gender", maxlength=1, 
> choices=GENDER_CHOICES, default='m' )
>
> Using form_for_model() on the above model results in HTML like this:
>
> 
> -
> Male
> Female
> 
>
> How do I get rid of the first option (the dashes)?
> I would prefer to have the -widget default to my default-value.
>
> Any help appreciated!
>
> -mark


--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---

Canberra developers

[Ryan]

I live in Canberra, Australia and I'm putting together a web start-up,
hopefully using the Django framework. Does anyone know of any
competent Django developers, or at least Python coders, in my neck of
the woods?
--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---

Negative numbers comparison bug(?)

[Ryan]

Hi,
   I am currently making a mashup of the Weather Channel's RSS feeds
(temperature and wind speed data) and Google maps with Django. Briefly
it puts weather data on a map, e.g. sunny in London, rain in
Edinburgh. I am having problems with comparing longitudes and
latitudes in the database when they are negative. As an example, If I
have in the database: locationID=1, lon = 3, lat = -1

If I try to find locations that are within one degree of longitude I
believe I would do something like this:
Location.objects.filter(lon__range=(2, 4)), this returns the correct
locationIDs within 2 and 4 degrees longitudes. However if I do the
same for latitude: Location.objects.filter(lat__range=(-3, -1)), it
fails.

Essentially my problem is I can't filter out database entries when
comparing them to negative numbers (always works when positive). Does
anyone know it it's a problem with SQL's between syntax or if it's
Django? If it helps, the values are floats (decimal 5,2).

 Could anyone please help me?

   Thanks in advance,
  Ryan


--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---

Location of non-app-specific static files?

[Ryan]

Hi,

I have been wondering where people put their non-app-specific static files 
in their django projects?  For example, the base css file that applies to 
all pages across the project or perhaps the jquery file?

Currently I have the following structure:

.
├── app
│   ├── __init__.py
│   ├── models.py
│   ├── static
│   │   └── app
│   │   ├── css
│   │   ├── img
│   │   └── js
│   ├── templates
│   │   └── app
│   ├── tests.py
│   └── views.py
├── manage.py
└── mysite
├── __init__.py
├── settings.py
├── static
├── templates
├── urls.py
└── wsgi.py

My STATIC_ROOT setting points to the ./mysite/static directory as this is 
where I would like collectstatic to dump all the static files for 
deployment, however this is also the obvious place (for me anyway) to place 
non-app-specific static files as they apply to the project as a whole like 
the templates directory at this level.

Does any one have any input on this?

Thanks,

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/4H39KqmnTugJ.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Location of non-app-specific static files?

[Ryan]

That is not a bad idea as everything that is project specific, but not app 
specific, would live in one place.

Thanks,

Ryan

On Sunday, 4 November 2012 15:14:43 UTC, Xavier Ordoquy wrote:
>
> Hi,
>
> One thing I've seen - and adopted - is to have one application that 
> contains the static files.
> It looks like:
>
> └── mysite
> ├── __init__.py
> ├── settings.py
>
> ├── theme
>
> ├── __init__.py
>
> ├── models.py (empty file)
>
> └── static
>
> ├── static
> ├── templates
> ├── urls.py
> └── wsgi.py
>
>
> Then you add the application in your INSTALLED_APPS and you're done.
>
> Regards,
> Xavier Ordoquy,
> Linovia.
>
> Le 4 nov. 2012 à 00:23, Ryan > a écrit 
> :
>
> Hi,
>
> I have been wondering where people put their non-app-specific static files 
> in their django projects?  For example, the base css file that applies to 
> all pages across the project or perhaps the jquery file?
>
> Currently I have the following structure:
>
> .
> ├── app
> │   ├── __init__.py
> │   ├── models.py
> │   ├── static
> │   │   └── app
> │   │   ├── css
> │   │   ├── img
> │   │   └── js
> │   ├── templates
> │   │   └── app
> │   ├── tests.py
> │   └── views.py
> ├── manage.py
> └── mysite
> ├── __init__.py
> ├── settings.py
> ├── static
> ├── templates
> ├── urls.py
> └── wsgi.py
>
> My STATIC_ROOT setting points to the ./mysite/static directory as this is 
> where I would like collectstatic to dump all the static files for 
> deployment, however this is also the obvious place (for me anyway) to place 
> non-app-specific static files as they apply to the project as a whole like 
> the templates directory at this level.
>
> Does any one have any input on this?
>
> Thanks,
>
> Ryan
>
> -- 
> You received this message because you are subscribed to the Google Groups 
> "Django users" group.
> To view this discussion on the web visit 
> https://groups.google.com/d/msg/django-users/-/4H39KqmnTugJ.
> To post to this group, send email to django...@googlegroups.com
> .
> To unsubscribe from this group, send email to 
> django-users...@googlegroups.com .
> For more options, visit this group at 
> http://groups.google.com/group/django-users?hl=en.
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/v0gcHJaIbLcJ.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Editing .po files for RTL languages

[Ryan]

I'm needing to translate several Django sites/apps into arabic (a
Right-To-Left language) and I'm having difficulty editing the .po
files when the message to translate contains Left-To-Right variables
or HTML markup.

Whether I'm using TextMate, TextWrangler or poedit, if I'm editing a
translation it correctly has the text RTL, but it also turns any
variables/markup RTL when it needs to stay LTR.

For example, I get this:

verbose_name)s)% تم تحديثه بنجاح.

When it needs to be this:

%(verbose_name)s تم تحديثه بنجاح.

Similar problem if I need to do something like have a  or 
tag around a piece of text in the translation file. As I type the LTR
HTML tags in the RTL text it tries to make the tags RTL too.

It looks like Django's arabic translation file [1] has variables LTR,
so I'm curious what editor the Django arabic translators are using,
what fixes I need for poedit or what I'm doing wrong when I edit.

>From Googling around, I did see a poedit issue logged a few years ago
[2], but they said it's a native control issue, which might explain
why I see this same behavior in every Mac OS X editor I've tried thus
far.


[1] 
http://code.djangoproject.com/svn/django/trunk/django/conf/locale/ar/LC_MESSAGES/django.po
[2] http://www.poedit.net/trac/ticket/296

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

South African Django Developers

[Ryan]

Are there any Django developers in Johannesburg, South Africa? If so,
who can I get in touch with? Thanks.

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Django admin and jQuery select filtering

[Ryan]

Hi,

First of all I will start by saying that I am fairly new to django, jQuery 
and javascript, but have been using python for a number of years now.  In 
order to get to know django, I thought I would try to develop a ladder 
application to display player rankings for my local squash club.  As part of 
this I wanted to add some ajax to some of my forms in the django admin in 
order to dynamically filter some of the options in a select box depending on 
the users selection in the previous box.  For example, when adding a new 
player to a ladder, it makes little sense to allow the user to select a 
player who is already a member of that ladder.  In order to do this, I have 
created a new model form as shown here: http://pastebin.com/y2uZxmnF.  I 
then wrote the following javascript (selectfilter.js): 
http://pastebin.com/vyHXzMH2.  Finally I wrote a view to select the required 
objects: http://pastebin.com/D2JVNHxS.

This works great, however, I need to use the same kind of functionality on 
other forms, but there is a lot hardcoded in to the javascript file and view 
so would require a new javascript file for each new form.  So I was 
wondering if there was a good way to make this more dynamic so the same code 
could be used for most (if not all) forms?

Any help would be greatly appreciated.

Many thanks,

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: choice_set.all()

[Ryan]

Can you post your model code please?

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Form and ForeignKey Limiting

[Ryan]

Could you not change the queryset of the ModelChoiceField in the view? That 
way you will have access to the current user from request.user.

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Auditing Record Changes in All Admin Tables

[Ryan]

This will show you how to achieve this in the django admin: 
https://docs.djangoproject.com/en/dev/ref/contrib/admin/#django.contrib.admin.ModelAdmin.save_model

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Django dynamic form simple Example

[Ryan]

I do not know the details of implementing this, however I do know that the 
approach you are trying to take is incorrect.  It is not generally a good 
idea to dynamically change the definition of a form like that.  What I 
suggest is that you use 
formsets<https://docs.djangoproject.com/en/1.3/topics/forms/formsets/>to 
display multiple instances of the same form.  So for example, if you 
changed your form definition to:

class testform(forms.Form):
Q = forms.CharField()

You could then create a formset like so:

from django.forms.formsets import formset_factory
formset = formset_factory(testform, extra=2)

Once displayed in the template, this would show two instances of the form 
giving you the two fields you require.  You could then either change the 
extra attribute in the view if you know how many fields you require, or you 
could use the 
django-dynamic-formset<http://code.google.com/p/django-dynamic-formset/>jQuery 
plugin in order to allow the user to add or remove more forms 
dynamically.

Hope this helps,

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Form and ForeignKey Limiting

[Ryan]

Yes it is possible, but I think I may have misunderstood you at first.  All 
you want to do is to force the user who created the task to become its 
manager? If so you can do this very simply in the django admin using this 
code: 
https://docs.djangoproject.com/en/1.3/ref/contrib/admin/#django.contrib.admin.ModelAdmin.save_model.
  
Just make sure you set editable=False in your model definition for manager 
in the Task model.

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: custom User class

[Ryan]

If all you want to do is store additional information about your users, the 
recommended way to do this is to use a user profile: 
https://docs.djangoproject.com/en/dev/topics/auth/#storing-additional-information-about-users

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Form and ForeignKey Limiting

[Ryan]

You could always try this:

class TaskForm(models.ModelForm):
class Meta:
model = Task

def __init__(self, queryset, *args, **kwargs):
super(TaskForm, self).__init__(*args, **kwargs)
self.fields['manager'].queryset = queryset

and then just pass the required queryset when you create the form instance 
in your view.

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Django - UserProfile m2m field in admin - error

[Ryan]

I can confirm that I get the same error.  I wonder if it is anything to do 
with the two different forms auth uses for user creation and  change?

On a side note, how did you get your code so nicely formatted?

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Auditing Record Changes in All Admin Tables

[Ryan]

That code is supposed to go in the models admin definition like so:

admin.py:
--
class AuditAdmin(admin.ModelAdmin):
def save_model(self, request, obj, form, change):
obj.user = request.user
obj.save()

Ryan

-- 
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 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Auditing Record Changes in All Admin Tables

[Ryan]

The model code still needs to stay in the model, it is just the save_model 
method had to be moved to the admin.py.  I think the following is what you 
are looking for:

models.py
---
from django.db import models
from django.contrib.auth.models import User

class AuditedTable(models.Model):
created = models.DateTimeField(auto_now_add=True)
created_by = models.ForeignKey(User, editable=False)
updated = models.DateTimeField(auto_now=True)
updated_by = models.ForeignKey(User, editable=False)

class Meta:
abstract = True

class Entity1(AuditedTable):
title = models.CharField(max_length=20)

def __unicode__(self):
return self.title

admin.py
-
from django.contrib import admin
from app.models import Entity1

class AuditAdmin(admin.ModelAdmin):
def save_model(self, request, obj, form, change):
if change:
obj.updated_by = request.user
else:
obj.created_by = request.user
obj.save()

class Entity1Admin(AuditAdmin):
pass

admin.site.register(Entity1, Entity1Admin)

Hope that helps,

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/NV9uTUJxUVRkTW9K.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Auditing Record Changes in All Admin Tables

[Ryan]

That bit is the easy bit :p

Just change request.user to request.user.username

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/RXpUd1BTX2E3RVVK.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

column cannot be null

[Ryan]

blank = True means that a django form will not require a value, but the 
database still does so it's expected that you provide a value elsewhere. In 
order to allow a null entry into the database, you will need to add null = True 
to that definition aswell

Hope this helps,

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/WFcxX2R4ZFB1VW9K.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: Auditing Record Changes in All Admin Tables

[Ryan]

No problem.

One final piece of advice would be to move the readonly_fields out of 
Entity1Admin and into AuditAdmin. That way you only have to define it once and 
inherit from it as with all the other options rather than re-defining it for 
each new model. 

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/RDk4VnBBdU1mck1K.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: confused over use of XYZ.objects.get() method

[Ryan]

Right, if what you have typed in your post is the code you are actually 
trying to use, there are two problems that jump out at me.

The first is that *question = models.CaaQuestion(pk=210) *will actually 
create a new question with the primary key 210, not fetch one from the 
database, so when you try to use this to find answers in the subsequent 
lines it will not find anything as it is not currently in any of those 
tables.

The second if the FieldError you are receiving.  This is because the 
question_id and assessment_id arguments should have double underscores to 
tell django that you are trying to span a relationship.  So these two should 
actually be question__id and assessment__id.  However the line above that 
with question = question and assessment = assessment should work fine when 
you have a question that is actually in the database.

Hope that helps you,

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/VHF3WmMyZVVCSVlK.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: ManyToManyField limit_choices_to that instance via Django admin.

[Ryan]

This is possible, but I'm not sure if you could do it via limit_choices_to.  
The admin docs shows how to accomplish this here: 
https://docs.djangoproject.com/en/dev/ref/contrib/admin/#django.contrib.admin.ModelAdmin.formfield_for_manytomany

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/ZlhkM0JHY3dkbTRK.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

Re: VERY cheap django hosting?

[Ryan]

Do you know how their python is provided? mod_python? mod_wsgi? fgci?

Thanks,

Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/django-users/-/6KIjlspcjIwJ.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to 
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en.

migrate command fails with foreign key to user model

[Ryan]

I have used django in the past, but not for some time.  So the changes 
surrounding the user models are new to me.  I downloaded v1.7b2 and 
proceeded to start a new development I am planning.  However I came across 
a problem when trying to run migrate after generating the migrations on an 
app that has a model that contains a foreign key to the user model.  So I 
created a simple test app to isolate the problem, with a model defined as 
below:

from django.db import models
from django.conf import settings

class TestModel(models.Model):
user = models.ForeignKey(settings.AUTH_USER_MODEL)

When I run "makemigrations test_app" the migrations seem to be created fine 
as there are no errors produced.  However when I run "migrate" I get the 
folllowing errors:

C:\Users\osborn_r\django\test_project>python manage.py migrate
Operations to perform:
  Synchronize unmigrated apps: admin, contenttypes, auth, sessions
  Apply all migrations: test_app, auth
Synchronizing apps without migrations:
  Creating tables...
  Installing custom SQL...
  Installing indexes...
Running migrations:
  Applying test_app.0001_initial...Traceback (most recent call last):
  File "manage.py", line 10, in 
execute_from_command_line(sys.argv)
  File "C:\Python27\lib\site-packages\django\core\management\__init__.py", 
line
427, in execute_from_command_line
utility.execute()
  File "C:\Python27\lib\site-packages\django\core\management\__init__.py", 
line
419, in execute
self.fetch_command(subcommand).run_from_argv(self.argv)
  File "C:\Python27\lib\site-packages\django\core\management\base.py", line 
288,
 in run_from_argv
self.execute(*args, **options.__dict__)
  File "C:\Python27\lib\site-packages\django\core\management\base.py", line 
337,
 in execute
output = self.handle(*args, **options)
  File 
"C:\Python27\lib\site-packages\django\core\management\commands\migrate.py
", line 145, in handle
executor.migrate(targets, plan, fake=options.get("fake", False))
  File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
line 60
, in migrate
self.apply_migration(migration, fake=fake)
  File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
line 88
, in apply_migration
if self.detect_soft_applied(migration):
  File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
line 13
2, in detect_soft_applied
apps = project_state.render()
  File "C:\Python27\lib\site-packages\django\db\migrations\state.py", line 
63, i
n render
model=dangling_lookup[0]))
ValueError: Lookup failed for model referenced by field 
auth.Permission.content_
type: contenttypes.ContentType

Can anyone spot what I have done wrong?

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to django-users+unsubscr...@googlegroups.com.
To post to this group, send email to django-users@googlegroups.com.
Visit this group at http://groups.google.com/group/django-users.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/django-users/f38bf94b-da37-4b2e-8722-a3263b12442d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: migrate command fails with foreign key to user model

[Ryan]

Thanks for the info.  At least I now know it wasn't anything I was doing 
wrong!

On Monday, 28 April 2014 17:54:29 UTC+1, Simon Charette wrote:
>
> This is a release blocker for Django 1.7 which is being tracked in 
> #22485<https://code.djangoproject.com/ticket/22485>
> .
>
> Le lundi 28 avril 2014 04:54:32 UTC-4, Ryan a écrit :
>>
>> I have used django in the past, but not for some time.  So the changes 
>> surrounding the user models are new to me.  I downloaded v1.7b2 and 
>> proceeded to start a new development I am planning.  However I came across 
>> a problem when trying to run migrate after generating the migrations on an 
>> app that has a model that contains a foreign key to the user model.  So I 
>> created a simple test app to isolate the problem, with a model defined as 
>> below:
>>
>> from django.db import models
>> from django.conf import settings
>>
>> class TestModel(models.Model):
>> user = models.ForeignKey(settings.AUTH_USER_MODEL)
>>
>> When I run "makemigrations test_app" the migrations seem to be created 
>> fine as there are no errors produced.  However when I run "migrate" I get 
>> the folllowing errors:
>>
>> C:\Users\osborn_r\django\test_project>python manage.py migrate
>> Operations to perform:
>>   Synchronize unmigrated apps: admin, contenttypes, auth, sessions
>>   Apply all migrations: test_app, auth
>> Synchronizing apps without migrations:
>>   Creating tables...
>>   Installing custom SQL...
>>   Installing indexes...
>> Running migrations:
>>   Applying test_app.0001_initial...Traceback (most recent call last):
>>   File "manage.py", line 10, in 
>> execute_from_command_line(sys.argv)
>>   File 
>> "C:\Python27\lib\site-packages\django\core\management\__init__.py", line
>> 427, in execute_from_command_line
>> utility.execute()
>>   File 
>> "C:\Python27\lib\site-packages\django\core\management\__init__.py", line
>> 419, in execute
>> self.fetch_command(subcommand).run_from_argv(self.argv)
>>   File "C:\Python27\lib\site-packages\django\core\management\base.py", 
>> line 288,
>>  in run_from_argv
>> self.execute(*args, **options.__dict__)
>>   File "C:\Python27\lib\site-packages\django\core\management\base.py", 
>> line 337,
>>  in execute
>> output = self.handle(*args, **options)
>>   File 
>> "C:\Python27\lib\site-packages\django\core\management\commands\migrate.py
>> ", line 145, in handle
>> executor.migrate(targets, plan, fake=options.get("fake", False))
>>   File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
>> line 60
>> , in migrate
>> self.apply_migration(migration, fake=fake)
>>   File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
>> line 88
>> , in apply_migration
>> if self.detect_soft_applied(migration):
>>   File "C:\Python27\lib\site-packages\django\db\migrations\executor.py", 
>> line 13
>> 2, in detect_soft_applied
>> apps = project_state.render()
>>   File "C:\Python27\lib\site-packages\django\db\migrations\state.py", 
>> line 63, i
>> n render
>> model=dangling_lookup[0]))
>> ValueError: Lookup failed for model referenced by field 
>> auth.Permission.content_
>> type: contenttypes.ContentType
>>
>> Can anyone spot what I have done wrong?
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to django-users+unsubscr...@googlegroups.com.
To post to this group, send email to django-users@googlegroups.com.
Visit this group at http://groups.google.com/group/django-users.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/django-users/d3b85387-1c06-4f48-876c-0f6a75bbb4d6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: Syncing two django applications via cURL

[Ryan Nowakowski]

You could create a view that accepts a post that includes the following:
  * admin username (for auth)
  * admin password (for auth)
  * username
  * user password
  * other attrs...

Then from your admin hooks, you can do an http post to the other sites
using httplib[1] keeping stuff in sync.

1. http://www.python.org/doc/2.5.2/lib/httplib-examples.html

- Ryan

On Thu, Nov 13, 2008 at 10:13:25AM -0800, Kurczak wrote:
> 
> Hello everyone,
> I've got this unusual problem - I've got few django sites, and I need
> to synchronize Users and profiles between them. (it's a closed site
> where only admins create accounts so OpenID or similar won't help) I
> thought that probably the easiest way is to hook somewhere in admin
> interface (add, edit, delete actions) and run a cURL script that would
> log in to the rest of the sites as another user and execute same
> action. (I hope my description is understandable)
> Is it by any chance a good solution? If so, where is the best place to
> hook this?
> 
> I'd be profoundly grateful for any insights on this.
> 
> Thanks,
> Kurczak
> 
> 
> > 

--~--~-~--~~~---~--~~
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: FastCGI and Django as a continuously running server

[Ryan Nowakowski]

On Wed, Nov 05, 2008 at 11:21:14AM -0800, russellneufeld wrote:
>   The one thing left that I'd like to do is set up Django to run
> continuously, even when there are no http requests.  My application
> runs a bunch of periodic background tasks in addition to serving up
> web content, and it seems that after a few minutes with no web
> activity, the app is terminated.  I'm coming from the java world where
> you typically have one long-running jvm which handles all requests and
> does not start and stop with new Apache instances.  I'd like to set up
> something similar with Django.

With Django all state is in the database(unless you've setup cookies).
There's no guaranteed persistent process.  You'll need to use whatever
periodic scheduler Dreamhost gives you (cron, other?) to run your
background tasks as separate python scripts.  Here's an example:

#!/usr/bin/python
import sys
# This is my django model that I want to do something with
from mfg.bundle.models import Bundle

try:
bundles = Bundle.objects.all()
for bundle in bundles:
  bundle.do_something()
except:
print "Something went crazy wrong!"
sys.exit(1)

sys.exit(0)

--~--~-~--~~~---~--~~
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 create a form with dynamic number of fields?

[Ryan Witt]

I've recently had a situation where I needed to create dynamic forms
and corresponding code to handle them.

One thing that you should know about python is that there's nothing
special about classes, you can take a collection of attributes and
methods and use them to create a new class on-the-fly using python's
type() function. You can use this to dynamically build a form class
with a new field.

My roommate wrote a great blog post on how to do build classes
dynamically (http://askawizard.blogspot.com/2008/09/metaclasses-python-
saga-part-4_30.html), which you should read before attempting this.

Once you understand what that is talking about (or not), you end up
writing some code like this:

If you want this form to be dynamic:

class MyDynamicForm(Form):
foo = CharField(max_len = 100)
bar = IntegerField()
def clean_foo(self):
value = self.cleaned_data['foo']
if value != 'foo':
raise ValidationError('foo is not foo!')
def clean_bar(self):
value = self.cleaned_data['bar']
if value != 42:
raise ValidationError('wrong answer')

You could build it programatically with:

# Our base classes (which you could build programatically if you
wanted)
bases  = (Form,)

# Our class attributes
attys  = {
'foo' : CharField(max_len = 100),
'bar' : IntegerField(),
}

# This returns a clean method built from some parameters
def general_clean(name, correct_value, error_text):
def inner(self):
value = self.cleaned_data[name]
if value != correct_value:
raise ValidationError(error_text)
return inner

# Now add the clean methods. You could just define methods and assign,
but this illustrates generating them programatically.
for name, correct_value, error_text in (
('foo', 'foo', 'foo is not foo!'),
('bar', 42, 'wrong answer'),
):
attys['clean_' + name] = general_clean(name, correct_value,
error_text)

# Now, to create the dynamic class, you need the following
incantation:
DynamicForm = getattr(Form, '__metaclass__', type)('DynamicForm',
bases, attys)

# That only creates the *class*, we still have to create an instance
form = DynamicForm({'foo': 'foo', 'bar': 42})

Does this help? It would let you dynamically add elements and handlers
to the form that django is aware of, thus you get all the benefits of
using the form framework. Let me know if this is a bit thick and I'll
be happy to explain it and perhaps blog a more coherent example.

--Ryan



--~--~-~--~~~---~--~~
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 create a form with dynamic number of fields?

[Ryan Witt]

I've recently had a situation where I needed to create dynamic forms
and corresponding code to handle them.

One thing that you should know about python is that there's nothing
special about classes, you can take a collection of attributes and
methods and use them to create a new class on-the-fly using python's
type() function. You can use this to dynamically build a form class
with a new field.

My roommate wrote a great blog post on how to do build classes
dynamically (http://askawizard.blogspot.com/2008/09/metaclasses-python-
saga-part-4_30.html), which you should read before attempting this.

Once you understand what that is talking about (or not), you end up
writing some code like this:

If you want this form to be dynamic:

class MyDynamicForm(Form):
foo = CharField(max_len = 100)
bar = IntegerField()
def clean_foo(self):
value = self.cleaned_data['foo']
if value != 'foo':
raise ValidationError('foo is not foo!')
def clean_bar(self):
value = self.cleaned_data['bar']
if value != 42:
raise ValidationError('wrong answer')

You could build it programatically with:

# Our base classes (which you could build programatically if you
wanted)
bases  = (Form,)

# Our class attributes
attys  = {
'foo' : CharField(max_len = 100),
'bar' : IntegerField(),
}

# This returns a clean method built from some parameters
def general_clean(name, correct_value, error_text):
def inner(self):
value = self.cleaned_data[name]
if value != correct_value:
raise ValidationError(error_text)
return inner

# Now add the clean methods. You could just define methods and assign,
but this illustrates generating them programatically.
for name, correct_value, error_text in (
('foo', 'foo', 'foo is not foo!'),
('bar', 42, 'wrong answer'),
):
attys['clean_' + name] = general_clean(name, correct_value,
error_text)

# Now, to create the dynamic class, you need the following
incantation:
DynamicForm = getattr(Form, '__metaclass__', type)('DynamicForm',
bases, attys)

# That only creates the *class*, we still have to create an instance
form = DynamicForm({'foo': 'foo', 'bar': 42})

Does this help? It would let you dynamically add elements and handlers
to the form that django is aware of, thus you get all the benefits of
using the form framework. Let me know if this is a bit thick and I'll
be happy to explain it and perhaps blog a more coherent example.

--Ryan




--~--~-~--~~~---~--~~
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: Django processes under Apache

[Ryan Kelly]

> Based on my experiments, it does seem like each Apache process
> maintains a live Django object which it reuses request after request
> and which maintains state such as global variables in modules between
> requests. Does that mean that a Django instance has some kind of loop
> in it that waits for requests and then creates responses?

The wait loop is inside Apache - Django just exports a funtion which
Apache calls for every request, e.g. the "handler" function in
core/handlers/modpython.py

> The slave script that gets called with subprocess.Popen uses
> fcntl.flock to lock onto an arbitrary file while it's working. A
> middleware looks for tasks that need to be completed on every response
> and if work needs to be done, quickly tries to get a lock on the lock
> file in a non-blocking way, and if it can, it releases the lock and
> uses Popen to start the slave. Since the task files are numbered, they
> form a sort of queue. The point of all this is to generate thumbnails
> and not wait around for it.

You might get some mileage out of the multiprocessing module, which has
proper lock and queue objects.  Although not available natively in
python 2.5, it can be installed as a stand-alone package.

This way you could spawn a single worker process (or pool of processes)
for each django instance, and have your middleware simply push tasks
onto a queue for them to handle.  It also lets you specify your worker
processes as "daemonic" so they're cleaned up automatically when Django
exits.


  Cheers,

 Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

model inheritance without a new database table

[Ryan Kelly]

Hi All,


  I'm currently working on an authentication backend that adds some
functionality to the standard one, and as part of this I want to modify
the behaviour of the User class.  I've been subclassing auth.models.User
using standard model inheritance and it seems to be working fine.

  However, I don't need to add any fields to the User class, just
add/modify some methods.  It seems like a waste to have an additional
database table created for my model subclass when it won't contain any
useful information.  I briefly toyed with the idea of just
monkey-patching the User class, but figured there must be a better
way :-)

  So, is there a way to have a model subclass avoid the creation of a
new database table,and just take its data straight out of the table for
its superclass?  


  Thanks,

 Ryan


-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: model inheritance without a new database table

[Ryan Kelly]

> There's a ticket open in Trac (my internet connection is terrible at the
> moment, so I'm not going to hunt it out) that is to add an option to the
> Meta class saying "this model is not to be managed by Django".

Thanks Malcolm, I'll definitely take a look and see if I can help out on
the ticket at all.

  Cheers,

 Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: model inheritance without a new database table

[Ryan Kelly]

> > is there a way to have a model subclass avoid the creation of a
> > new database table,and just take its data straight out of the table for
> > its superclass?  
> 
> Right now, I suspect you would also have to manually set Meta.db_table
> and some things like that. However, I strongly suspect each of those are
> fairly straightforward to sort out. We can set db_table automatically
> for such models, if it's not set. We have to make sure the pk property
> talks to the first ancestor model, too. Basically a bunch of fiddly
> stuff, but nothing that's brain surgery levels of difficulty.

Following Malcolm's tips, I have managed to get this working.  After
applying the patch from ticket #3163, I've formulated this class
decorator to make the necessary tweaks to the model's meta info:


def virtual_inheritance(cls):
base = cls.__bases__[0]
pk = base._meta.pk
#  don't create a new db table
cls._meta.managed = False
#  use the superclass table name
cls._meta.db_table = base._meta.db_table
#  remove the auto-created parent reference
cls._meta.local_fields[:] = []
#  use the superclass primary key
cls._meta.pk = pk
cls._meta.parents[base] = pk


Now I can do the following:


class MyUser(auth.models.User):
def an_extra_method(self):
   print "hooray!"
virtual_inheritance(MyUser)


The resulting model behaves exactly as I wanted, adding functionality to
the default User model but taking its data straight out of the auth_user
table.


If I'm feeling inspired tomorrow, I might try to formulate similar logic
as a patch to the ModelBase metaclass, so that subclasses that don't add
any fields will get pure-python inheritance by default.  In the
meantime, I hope some other people might find this trick useful.


  Cheers,

 Ryan


-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: model inheritance without a new database table

[Ryan Kelly]

> > If I'm feeling inspired tomorrow, I might try to formulate similar logic
> > as a patch to the ModelBase metaclass, so that subclasses that don't add
> > any fields will get pure-python inheritance by default.
> 
> That (changing the default behaviour) would be a bad idea. The
> implicit primary key field shouldn't just magically not appear because
> something else isn't there.


Indeed - and on further reflection it seems that what I want is a little
different from normal subclassing anyway.  Standard subclasses require
only that all instances of the subclass are also instances of the base
class.  In this case, I'm trying to make all instances of the base class
automatically be available as instances of the subclass.

I wonder if there's a standard terminology for this - I'm calling them
"virtual subclasses" but maybe they should be called "facade subclasses"
or something like that?

> The good news, though, is that I suspect a patch to ModelBase will be
> a bit simpler than your decorator, since we can just conditionally
> check _meta.managed to determine whether to add things like the
> auto-fields and the links back to the parent.

Yep, getting a patch working for this was pretty straightforward.
However, I think it would be better to use a distinct option instead of
overloading "managed", since it would be quite legitimate for someone to
want an unmanaged subclass that still lived in its own (manually
created) database table.

My current patch uses "virtual", so I can do the following:

def MyUser(auth.models.User):
class Meta:
virtual = True
def an_extra_method(self):
print "hooray!"


What do you think?  I intend to pretty this up and submit a patch
sometime this evening.


  Cheers,

Ryan


-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: model inheritance without a new database table

[Ryan Kelly]

> Oh, okay. Yeah, that's quite different. I'm not convinced that should go
> into Django. Again, there's a consistency thing: we present the object /
> class type that is asked for, not some transparently descended version.

I think my explanation may have been a bit off...

> The use-case for the pure-Python inheritance that has been floating
> around is the case when you want to, say, use a different User manager
> in your own views. You don't want to change the auth app's behaviour in,
> say, admin, but you'd prefer a different default when you're using it.
> So you subclass User, using Python-only inheritance, add a new default
> manager and then actually use the subclass, rather than User in your
> code.

because this sounds like exactly how I want to use this feature.  I
subclass auth.User and add my own managers/methods/etc, and use this
subclass in my own code.  But if someone creates a new instance of
auth.User via the standard class, it should also show up in queries
posed through my subclass.

Likewise, if I add a new user through my special subclass, it will show
up as an ordinary user to e.g. the admin interface.

So I'm not trying to add anything magical - the "virtual subclass" is
simply a "facade" or "wrapper" around the data in the base class table
that presents a modified interface.  Yikes, there's that naming problem
again...

> Still, I wouldn't let that hold things up. I'll think about it and come
> up with a name later on, unless a better option appears beforehand.

Originally I named it "use_superclass_table" but decided that was just
too ugly to deal with.  "interface_only"?  Glad not to be the one stuck
making that decision :-)


Thanks for all your feedback, I'll get some code into trac tonight.


   Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: model inheritance without a new database table

[Ryan Kelly]

Thanks again for your feedback on this Malcolm, I've created the
following ticket in Trac:

   Proxy models: subclass a model without creating a new table
   http://code.djangoproject.com/ticket/10356



  Cheers,

 Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Re: Browser Testing - Selenium or Windmill

[Ryan Kelly]

> What's the best tool for doing automated browser testing with Django
> apps? I have some personal experience with Selenium (although not when
> testing Django), which seems fairly mature, and has a great Firefox
> extension, but has some serious problems dealing with frames and
> popups. I recently discovered Windmill, but haven't heard too much
> about it.

I don't know about "best", but I've had good experiences using Windmill.
It ships with a "WindmillDjangoUnitTest" class for easy integration with
the Django testrunner, and it's straightforward to 
write your testcases in Python using a style quite similar to the native
Django test client. Whether it handles frames and popups in the way you
want, I have no idea.

On the other hand, I think the docs and other community resources
available for Selenium are more comprehensive than for Windmill -
probably because the latter is a younger project.

In the end I went with Windmill because it seemed to fit my
development/thinking style better than Selenium - so of course your
mileage may vary.

  Cheers,

Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part

Possible bug with DateTimeFields and admin site

[Ryan Duffield]

Hello,

I thought I would check here first to see if what I am experiencing
is, in fact, a bug (using Django 1.0.2).

The admin site JavaScript, when creating a new instance of a model
with a DateTimeField, attempts to focus on an element that does not
exist.  Take this example:

class Reservation(models.Model):
start_date = models.DateTimeField(blank=False, null=False)

This results in the use of the AdminSplitDateTime widget being used.
The two respective HTML fields have IDs of "id_start_date_0" and
"id_start_date_1".  However, the JavaScript that is generated is:

document.getElementById("id_start_date").focus();

This fails, due to there not being an element on the page with an ID
of "id_start_date".

I tracked this down to the _auto_id method within the BoundField
class.  Creating the "auto_id" depends on the field's "html_name" --
in the case outlined above, both fields have html_name's of
"start_date".

Is this a bug?

Cheers,
Ryan

--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Flatpages-like application and caching

[Ryan K]

Greetings,

I am designing an application that is very simple like flatpages but I
need the ability to associate a menu with the content which will be
done through the URL (i.e. example.com/products/widget2009 would
invoke the products menu and the widget2009 static page).

I expect these pages to change rarely but I would still like my client
to have the ability to modify menus and and page content.

My question is in the Menu model I plan on writing. I want to possibly
use signals to create an XHTML document that is the menu code instead.
This is basically my simple version of caching instead of dynamically
creating the menu upon each access.

I'm not really sure if that is true because I don't really know how
the cache system works in Django. Is the manager smart enough to where
if I created a MenuModel.generate_xhtml() menu which plucked links
associated with the menu from the database to create the XHTML, what
is the best method to cache this resulting XHTML.

Are there any applications that already exist that extend flatpages to
include one level deep sub navigation?

any help is greatly appreciated!

Cheers,
Ryan Kaskel
http://consulting.ryankaskel.com/
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Checking if a ForeignKey field is NULL

[Ryan K]

Lets say I retrieve from my database an object foo and foo's model has
a ForeignKey field, baz, with null and blank set to True. Will a
simple test like if foo.baz ... test whether the field is set to NULL
or not?

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Django and its caching system help

[Ryan K]

Greetings,

I am designing an application that includes a menu with a many to many
relationship with links. These menus will hardly ever change. What is
the best way to cache these pages? Should I use the messaging system
to create an html file to use when the menu is changed? Is database
access quicker than disk access? Do I even need to do this considering
Django has a cache system? Any help would be greatly appreciated.

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Circular imports problem

[Ryan K]

I am creating an application that uses the dispatcher in Django. In
the apps models.py file I connect the post_save of two models to
functions in a signals.py file. The problem is that I need to use
several models in models.py and that leads to a circular import. What
is the best way to solve this problem?

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Circular imports problem

[Ryan K]

Well all of the models are part of the same application. So here is
the code:

models.py:

from django.db import models
from django.db.models import signlas
from django.dispatch import disaptacher

from asqcom.apps.staticpages import signals
from asqcom.abstract_models import CommonAbstractModel

class Link(CommonAbstractModel):
url = models.CharField(max_length=1024)
title = models.CharField(max_length=64)
text = models.CharField(max_length=64)

class Meta:
verbose_name = 'static page'
verbose_name_plural = 'static pages'
ordering = ('url',)

def __unicode__(self):
return u"%s - %s" (self.text, self.url,)

class Menu(CommonAbstractModel):
name = models.CharField(max_length=64)
title = models.CharField(max_length=64) # used in the 
links = models.ManyToManyField(Link)

class Meta:
verbose_name = 'menu'
verbose_name_plural = 'menus'
ordering = ('name',)

def __unicode__(self):
return u"%s" % (self.name)

def to_xhtml(self):
pass

class StaticPage(CommonAbstractModel):
url = models.CharField(max_length=128, db_index=True)
title = models.CharField(max_length=256)
content = models.TextField(blank=True)
menu = models.ForeignKey(Menu, null=True, blank=True)
template_name = models.CharField(max_length=70, blank=True)
registration_required = models.BooleanField()

class Meta:
verbose_name = 'static page'
verbose_name_plural = 'static pages'
ordering = ('url',)

def __unicode__(self):
return "%s - %s (menu: %s)" % (self.title, self.url, self.menu
or 'no menu',)

class CachedMenuXhtml(CommonAbstractModel):
menu = models.ForeignKey(Menu)
xhtml = models.TextField()

def __unicode__(self):
return self.xhtml

dispatcher.connect(signals.build_menu_from_link_mod,
   signal=signals.signals.post_save, sender=Link)
dispatcher.connect(signals.build_menu_from_menu_mod,
   signal=signals.signals.post_save, sender=Menu)


signals.py:

import threading

from asqcom.apps.staticpages.models import Link, Menu, StaticPage
from asqcom.apps.staticpages.models import CachedMenuXhtml

class GenerateMenuXhtmlByLink(threading.Thread):
def __init__(self, instance):
self.instance = instance
threading.Thread.__init__(self)

def run(self):
pass

class GenerateMenuXhtmlByMenu(threading.Thread):
def __init__(self, instance):
self.instance = instance
threading.Thread.__init__(self)

def run(self):
pass

def build_menu_from_link_mod(...

def build_menu_from_menu_mod(..


I hope this better specifies the issue. Should I import these in the
run method for each Thread subclass?

On Jul 8, 4:47 pm, Alex Gaynor  wrote:
> On Wed, Jul 8, 2009 at 3:45 PM, Ryan K  wrote:
>
> > I am creating an application that uses the dispatcher in Django. In
> > the apps models.py file I connect the post_save of two models to
> > functions in a signals.py file. The problem is that I need to use
> > several models in models.py and that leads to a circular import. What
> > is the best way to solve this problem?
>
> > Cheers,
> > Ryan
>
> Depending on how you're using these other models you can a) do imports
> inside of functions, or 
> b)http://docs.djangoproject.com/en/dev/ref/models/fields/#lazy-relation
>  Or both (probably this)
>
> Alex
>
> --
> "I disapprove of what you say, but I will defend to the death your right to
> say it." --Voltaire
> "The people's good is the highest law."--Cicero
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Middleware help -- screwing up the admin

[Ryan K]

I'm getting a very strange error:

TypeError at /admin/

object.__new__() takes no parameters

I've isolated the problematic code to my middleware for an application
I built called staticpages. It uses the same mechanism and so the same
middleware as flatpages. This is the code:

from django.http import Http404
from django.conf import settings

from asqcom.apps.staticpages.views import staticpage

class StaticpageFallbackMiddleware(object):
def process_response(self, request, response):
if response.status_code != 404:
return response
try:
return staticpage(request, request.path_info)
# Return the original response if any errors happened. Because
this
# is a middleware, we can't assume the errors will be caught
elsewhere.
except Http404:
return response
except:
if settings.DEBUG:
raise 'staticpages.middleware error'
return response

When I comment out this middleware, everything is okay.

Below is the traceback:

#  /usr/local/lib/python2.6/dist-packages/django/core/handlers/base.py
in get_response

  85. # Apply view middleware
  86. for middleware_method in self._view_middleware:
  87. response = middleware_method(request, callback, callback_args,
callback_kwargs)
  88. if response:
  89. return response
  90.
  91. try:

  92. response = callback(request, *callback_args,
**callback_kwargs) ...

  93. except Exception, e:
  94. # If the view raised an exception, run it through exception
  95. # middleware, and if the exception middleware returns a
  96. # response, use that. Otherwise, reraise the exception.
  97. for middleware_method in self._exception_middleware:
  98. response = middleware_method(request, e)

# /usr/local/lib/python2.6/dist-packages/django/contrib/admin/sites.py
in wrapper

 185. return update_wrapper(inner, view)
 186.
 187. def get_urls(self):
 188. from django.conf.urls.defaults import patterns, url, include
 189.
 190. def wrap(view):
 191. def wrapper(*args, **kwargs):

 192. return self.admin_view(view)(*args, **kwargs) ...

 193. return update_wrapper(wrapper, view)
 194.
 195. # Admin-site-wide views.
 196. urlpatterns = patterns('',
 197. url(r'^$',
 198. wrap(self.index),

# /usr/local/lib/python2.6/dist-packages/django/contrib/admin/sites.py
in inner

 177. url(r'^my_view/$', self.admin_view(some_view))
 178. )
 179. return urls
 180. """
 181. def inner(request, *args, **kwargs):
 182. if not self.has_permission(request):
 183. return self.login(request)

 184. return view(request, *args, **kwargs) ...

 185. return update_wrapper(inner, view)
 186.
 187. def get_urls(self):
 188. from django.conf.urls.defaults import patterns, url, include
 189.
 190. def wrap(view):

# /usr/local/lib/python2.6/dist-packages/django/views/decorators/
cache.py in _wrapped_view_func

  37.
  38. def never_cache(view_func):
  39. """
  40. Decorator that adds headers to a response so that it will
  41. never be cached.
  42. """
  43. def _wrapped_view_func(request, *args, **kwargs):

  44. response = view_func(request, *args, **kwargs) ...

  45. add_never_cache_headers(response)
  46. return response
  47. return wraps(view_func)(_wrapped_view_func)

# /usr/local/lib/python2.6/dist-packages/django/contrib/admin/sites.py
in index

 359. context = {
 360. 'title': _('Site administration'),
 361. 'app_list': app_list,
 362. 'root_path': self.root_path,
 363. }
 364. context.update(extra_context or {})
 365. return render_to_response(self.index_template or 'admin/
index.html', context,

 366. context_instance=template.RequestContext(request) ...

 367. )
 368. index = never_cache(index)
 369.
 370. def display_login_form(self, request, error_message='',
extra_context=None):
 371. request.session.set_test_cookie()
 372. context = {


# /usr/local/lib/python2.6/dist-packages/django/template/context.py in
__init__

  99. def __init__(self, request, dict=None, processors=None):
 100. Context.__init__(self, dict)
 101. if processors is None:
 102. processors = ()
 103. else:
 104. processors = tuple(processors)
 105. for processor in get_standard_processors() + processors:

 106. self.update(processor(request)) ...


--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Middleware help -- screwing up the admin

[Ryan K]

Please ignore this. I'm an idiot and placed the reference to the
middleware in the template_context_processros settings. And so strange
errors resulted. Please delete if possible.
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Some post_related_objects_save signal or workaround?

[Ryan K]

Is there any reason you can't create your own signal and put it in the
rounds save() method and then just call the parents save?

http://docs.djangoproject.com/en/dev/topics/signals/#defining-and-sending-signals

Cheers,
Ryan


On Jul 9, 6:12 am, Eugene Mirotin  wrote:
> Hello!
>
> I have a tricky (as seems to me :) problem.
>
> Consider I have 2 simply related models - a Game and a Round (with FK
> to Game). Each has some generic fields like name and description.
> Also, the Round has the ordering field (called number) (oh, I can
> imagine your sighs =)
> They are edited on the single admin page for Game with TabularInline
> for game rounds.
>
> I know about the way to use jQuery for inlines reordering, but this is
> not the question. Currently I am exactly interested in the manual
> number field filling.
>
> So what happens when I save the game? First, the game itself is saved
> because of the need of the saved FK object for rounds saving, and I
> understand it. Then the rounds are saved in order.
> At this point due to some reason the rounds' numbers might be any
> numbers. But my applications requires them to be exactly the
> consequent positive integers starting from 1.
>
> This could be easily fixed by a simple loop, but it should be done at
> the appropriate point - on the per-game basis and _after_ all the
> related objects are saved (otherwise nasty bugs with incorrect
> ordering are possible).
>
> So I have to know the moment when all the current game's related
> objects are saved. And there is no appropriate standard signal for
> this.
> Any thoughts?
>
> Gene
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

post_save signal bug

[Ryan K]

Hello,

I am trying to implement a cache system using the post_save signal.
The problem is the code below, a Thread subclass that ends up calling
run(), the cache database never runs. It's driving me crazy!!! Is
there something going on with the way Django cache's queries which is
what I expect? Can I turn this off? This is the Thread subclass code
is below.

What I am trying to do is implement a staticpages app which is much
like flatpages but it adds a menu foreign key to what is basically a
flatpage. I am trying to make a cache system whereby when a user saves
a link or a menu, the below code rebuilds the XHTML to use in the
staticpage Context.

Here is a sample run:

1) I am in the admin and change a menu to consist of a title called
"Cool Sites 222" and three links
2) post_save runs and here is what is supposed to be saved in the
database (a print is called in my code:
Menu 2

  Cool Sites 222
  
   http://www.google.com/";>Google Inc

http://www.microsoft.com";>Microsoft

http://www.ryankaskel.com";>Ryan
Kaskel

http://www.yahoo.com";>Yahoo Homepage


  


If I don't change anything in the menu, eventually it gets the right
XHTML and prints it

3) The cached XHTML in the database is never changed and still
contains the information from its creation.

I think this is a bug because the following this code:

def _cache_menu(self, menu, xhtml):
"""
Stores xhtml into the CachedMenuXhtml table where the row's
menu matches
the menu_obj.
"""
if menu and xhtml:
cm = self.CachedMenuXhtml.objects.filter(menu__exact=menu)
if cm[0]:
print cm[0].xhtml
cm[0].xhtml = xhtml
cm[0].save()
print cm[0].xhtml
print xhtml
else:
new_cm = self.CachedMenuXhtml(menu=menu,xhtml=xhtml)
new_cm.save()

produces...


  Cool Sites
  
   http://www.microsoft.com";>Microsoft

http://www.ryankaskel.com";>Ryan
Kaskel

http://www.yahoo.com";>Yahoo Homepage


  



[11/Jul/2009 05:03:47] "POST /admin/staticpages/menu/2/ HTTP/1.1" 302
0

  Cool Sites
  
   http://www.microsoft.com";>Microsoft

http://www.ryankaskel.com";>Ryan
Kaskel

http://www.yahoo.com";>Yahoo Homepage


  




  Cool Sites 222
  
   http://www.google.com/";>Google Inc

http://www.microsoft.com";>Microsoft

http://www.ryankaskel.com";>Ryan
Kaskel

http://www.yahoo.com";>Yahoo Homepage


  


CODE for Thread subclass that post_save signal calls

# Signlas for caching of menu XHTML

import threading

from django.template.loader import get_template
from django.template import Context
from django.utils.safestring import mark_safe
from django.core.exceptions import ObjectDoesNotExist

from asqcom.apps.staticpages.settings import STATICPAGE_TEMPLATE,
MENU_TEMPLATE
from asqcom.apps.staticpages.settings import LINK_TEMPLATE,
MENU_CSS_ID

class GenerateMenuXhtml(threading.Thread):
"""
Subclasses a threading.Thread class to generate a menu's XHTML in
a separate
thread. All Link objects that have this menu associated with it
are gathered
and combined in an XHTML unordered list.

If the sender is of type Link, then all menus associated with that
link are
iterated through and rebuilt.
"""
def __init__(self, instance):
from asqcom.apps.staticpages.models import Menu, Link,
CachedMenuXhtml
self.Link = Link
self.Menu = Menu
self.CachedMenuXhtml = CachedMenuXhtml
self.instance = instance
self.menus_to_build = []
self.xhtml_for_menus = []
self.menus_for_cache = []
threading.Thread.__init__(self)

def _cache_menu(self, menu, xhtml):
"""
Stores xhtml into the CachedMenuXhtml table where the row's
menu matches
the menu_obj.
"""
if menu and xhtml:
print menu
print xhtml
cm = self.CachedMenuXhtml.objects.filter(menu__exact=menu)
if cm[0]:
cm[0].xhtml = xhtml
cm[0].save()
else:
new_cm = self.CachedMenuXhtml(menu=menu,xhtml=xhtml)
new_cm.save()

def cache_menus(self):
"""
Unpacks the menu objects and the XHTML for that menu and
iterates
through the list calling _cache_menu which does the real work.
"""
for pair in self.menus_for_cache:
self._cache_menu(pair[0],pair[1])

def _generate_xhtml(self, menu):
link_tmpl = get_template(LINK_TEMPLATE or 'staticpages/
link_default.html')
menu_tmpl = get_template(MENU_TEMPLATE or 'staticpages/
menu

Re: post_save signal bug

[Ryan K]

I modified the above _cached_menu method to just create a new object
every time (shown below). It only seems to get the links right, a
ManyToMay field, after I save it _twice_? Hmmm...how I get the updated
many2many table?

def _cache_menu(self, menu, xhtml):
"""
Stores xhtml into the CachedMenuXhtml table where the row's
menu matches
the menu_obj.
"""
if menu and xhtml:
cm = self.CachedMenuXhtml(menu=menu,xhtml=xhtml)
cm.save()


--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Does post_save signal need to be in models.py?

[Ryan K]

I am trying to avoid circular dependency issues  between signals.py
and models.py. Is there any reason why I can't connect the post_save
signal in signals.py itself (the function is connects with would be in
the same file? Should I import it in the apps/__init__.py file to
ensure the signal is installed?

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Question regarding post_save signal and ManyToMany field objects

[Ryan K]

Recently I've been trying to get answers about the post_save signal
behavior and perhaps have been asking too much of readers without
narrowing down the problem enough myself. So here is my effort to ask
a rather straightforward question:

The code that connects to the post_save signal, according to the
signal docs, is passed three objects and I am only concerned about
one: the instance that was saved. Now this instance has a
ManyToManyField and what I am finding out according to be debugging is
that the instance's ManyToMany field is NOT being updated after save
is called.

So, for example, when the signal calls the handler (this is being done
via the admin interface but it shouldn't matter) and I check
instance.manytomanyfield, it does not show the most recent changes.
What is the best way to "force" the update of the table so I can get
accurate data from the ManyToMany field? Even if I explicitly do
something like this:

def __init__(self, sender, instance):
self.sender = sender
self.instance = sender.objects.select_related().get
(id=instance.id)

What seems like cached data is being when I test:

self.instance.manytomanyfield.all()

Can anyone help with this behavior?

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: .views._CheckLogin.__call__ didn't return an HttpResponse object

[Ryan K]

Are you returning render_to_response in the view?

On Jul 14, 6:12 am, alecs  wrote:
> Hi all :) Can u help me with .views._CheckLogin.__call__ didn't return
> an HttpResponse object . I'm trying from my view 'jump' to another
> view by calling     mail_to_user(request,username,dummy)       The
> mail_to_user view contains render_to_response and if I simply copy the
> contents of mail_to_user into main view everything works OK. But I
> think it's better to call another view than to copy a whole conent.
> What can be wrong ?
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: .views._CheckLogin.__call__ didn't return an HttpResponse object

[Ryan K]

To be a little more explicit, perhaps your view code contains:

render_to_response('template.html",{},RequestContextInstance)

as opposed to

return render_to_response(...)

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Question regarding post_save signal and ManyToMany field objects

[Ryan K]

Yea...it would be so elegant too! Here is a "bug" filing:
http://code.djangoproject.com/ticket/5390. It's been open for two
years

I understand that your really getting what you asked for, just the
instance of the model you just saved but to me (and it seems you and a
lot of other people), intuition leads people to write code that drives
them nuts because its not really "post_transaction."

On Jul 14, 4:08 am, Eugene Mirotin  wrote:
> If I understand right, the problem is that you want the signal after
> all related fields are already saved. If so, this is exactly the
> problem I have too.
> You see, when you have such a relation (and, for example, edit the
> parent object with inlined related objects from the admin page, Django
> _has to_ save the parent object first, and only then it saves the
> related objects (you can verify it by adding print statements to your
> save methods or handlers and running the dev server from terminal).
> The reason is that when you initially create the object, related
> objects have to link to the existing object, so the parent object is
> saved before them.
>
> On Jul 13, 9:27 pm, Ryan K  wrote:
>
>
>
> > Recently I've been trying to get answers about the post_save signal
> > behavior and perhaps have been asking too much of readers without
> > narrowing down the problem enough myself. So here is my effort to ask
> > a rather straightforward question:
>
> > The code that connects to the post_save signal, according to the
> > signal docs, is passed three objects and I am only concerned about
> > one: the instance that was saved. Now this instance has a
> > ManyToManyField and what I am finding out according to be debugging is
> > that the instance's ManyToMany field is NOT being updated after save
> > is called.
>
> > So, for example, when the signal calls the handler (this is being done
> > via the admin interface but it shouldn't matter) and I check
> > instance.manytomanyfield, it does not show the most recent changes.
> > What is the best way to "force" the update of the table so I can get
> > accurate data from the ManyToMany field? Even if I explicitly do
> > something like this:
>
> >     def __init__(self, sender, instance):
> >         self.sender = sender
> >         self.instance = sender.objects.select_related().get
> > (id=instance.id)
>
> > What seems like cached data is being when I test:
>
> > self.instance.manytomanyfield.all()
>
> > Can anyone help with this behavior?
>
> > Cheers,
> > Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: .views._CheckLogin.__call__ didn't return an HttpResponse object

[Ryan K]

I'm not really sure I understand your question but you don't need to
jump to another view...you just need to pass the relevant objects to
the mailer function and then "return render_to_response(...)." If the
mailer function is already in another view, refactor the code and
create a non view function that handles mail.

Cheers,
Ryan

On Jul 14, 6:33 am, alecs  wrote:
> Ok, a simple question:
> Are there any possibilities to call from current view another view.
> For instance you have a large view and in the middle of it you need to
> perform THE SAME actions which are already defined in def foo
> (request,username).
> Of course, you can copy the contents of foo(request,username) view
> into your big view, but I don't think it's a correct solution.
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Object attributes in change_form.html

[Ryan K]

I'm not sure but try {{ original }} ...here is the code:

http://code.djangoproject.com/browser/django/trunk/django/contrib/admin/options.py

context = {
860 'title': _('Change %s') % force_unicode
(opts.verbose_name),
861 'adminform': adminForm,
862 'object_id': object_id,
863 'original': obj,
864 'is_popup': request.REQUEST.has_key('_popup'),
865 'media': mark_safe(media),
866 'inline_admin_formsets': inline_admin_formsets,
867 'errors': helpers.AdminErrorList(form, formsets),
868 'root_path': self.admin_site.root_path,
869 'app_label': opts.app_label,
870 }

On Jul 14, 5:37 am, janedenone  wrote:
> On Jul 13, 10:05 pm, janedenone  wrote:
>
>
>
>
>
> > On Jul 13, 6:07 pm, Daniel Roseman  wrote:
>
> > > On Jul 13, 4:02 pm, janedenone  wrote:
>
> > > > Hi,
>
> > > > I read the documentation on customizing admin templates, but I still
> > > > wonder how to access the object attributes in the change_form.html. I
> > > > am capable of getting the id using {{ object_id }}:
>
> > > > {% if change %}{% if not is_popup %}
> > > >   
> > > >      > > > class="historylink">{% trans "View Calculation" %}
> > > > ...
>
> > > > but I need to access a couple of other attributes. {{ object_name }}
> > > > does not work, nor does {{ object_cottage_name}} for getting a related
> > > > object's name.
>
> > > > Thankful for any hint.
>
> > > > - Jan
>
> > > The syntax for looking up attributes - in templates just like in
> > > Python - is a dot, not an underscore. So you should use object.name,
> > > etc.
>
> > Many thanks, but that does not work either.
>
> > - Jan
>
> PS. I was using {{ object_id }} because this worked in the
> object_tools block. "object" does not seem to be the object modified
> via change_form.html, though.
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Question regarding post_save signal and ManyToMany field objects

[Ryan K]

Alex,

So I have my Menus that contain links and a Link can be in multiple
Menus. Are you saying make an explicit model like:

class MenuLinks(models.Model):
menu = models.ForeignKey(Menu)
link = models. ForeignKey(Link)

?

Basically what I am doing is making an enhanced flatpages app called
staticpages. I want my client to have full control over the page's
content, menus associated with a page as well as links in the menu..
Here is my signals.py:

http://pastebin.com/m3c333dd

If I modify it to use the above table and connect to its post_save
signal, is my solution worth it? I use signals to generate an XHTML
menu and cache it in my database. I do this in a thread. Someone said
don't use threads. What is the consensus on that?

Cheers,
Ryan

On Jul 14, 10:43 am, Alex Gaynor  wrote:
> On Tue, Jul 14, 2009 at 9:35 AM, Russell Keith-Magee 
>
>
>
>
> > wrote:
>
> > On Tue, Jul 14, 2009 at 6:27 PM, Ryan K wrote:
>
> > > Yea...it would be so elegant too! Here is a "bug" filing:
> > >http://code.djangoproject.com/ticket/5390. It's been open for two
> > > years
>
> > A little history may help explain why this has been open for so long.
>
> > Historically (i.e., when dinosaurs roamed the Django source code)
> > signals were _very_ expensive - even signals that had no listeners. As
> > a result, there was a lot of resistance to adding new signals. Given
> > the performance cost, the decision was made to live without m2m
> > signals.
>
> > Just prior to v1.0, the signal framework was updated, which made
> > signals much faster. However, in an attempt to actually hit the
> > deadline for getting 1.0 out the door, adding new signals wasn't on
> > the plan.
>
> > Adding new signals was discussed as a possible feature for v1.1 (see
> > the features page [1]); however, nobody took charge of getting it
> > done, and other development priorities took over. The feature deadline
> > arrived before anyone got around to implementing (and integrating) the
> > change.
>
> > When the v1.2 development cycle starts, I'm sure this will be back on
> > the schedule again.
>
> > [1]http://code.djangoproject.com/wiki/Version1.1Features
>
> > Yours,
> > Russ Magee %-)
>
> The best way now to get this feature is to use an intermediary model which
> obviously get's its own signals.  Additionally I've been working on
> refactoring m2ms to always use an intermediary model, thus signals would
> always be sent.
>
> Alex
>
> --
> "I disapprove of what you say, but I will defend to the death your right to
> say it." -- Voltaire
> "The people's good is the highest law." -- Cicero
> "Code can always be simpler than you think, but never as simple as you want"
> -- Me
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: post_save signal bug

[Ryan K]

What I tried to build is stopped by the fact that I am incorrectly
assuming that a ManyToMany field is updated at the time Menu.save() is
called (or post_save signal handler called in this case). These two
methods of customizing what happens after you save an object DO NOT
apply to ManyToMany fields because they are not updated yet. The
transaction is not complete. I think Django should abstract
transactions and allow signals like pre_transaction_finished and
post_transaction_finished in addition to the ideas found at the page
for this "bug:" http://code.djangoproject.com/ticket/5390. (this post
if for anyone searching the topic)

On Jul 11, 1:57 am, Ryan K  wrote:
> I modified the above _cached_menu method to just create a new object
> every time (shown below). It only seems to get the links right, a
> ManyToMay field, after I save it _twice_? Hmmm...how I get the updated
> many2many table?
>
> def _cache_menu(self, menu, xhtml):
>         """
>         Stores xhtml into the CachedMenuXhtml table where the row's
> menu matches
>         the menu_obj.
>         """
>         if menu and xhtml:
>             cm = self.CachedMenuXhtml(menu=menu,xhtml=xhtml)
>             cm.save()
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: problems with large data

[Ryan K]

I'm starting to see that I too need to participate in Django sprints
to help incorperate the features I want in Django and I think we all
should and perhaps you do which is great. This is great software there
is still a lot of work to be done.

Ryan Kaskel
http://consulting.ryankaskel.com/

On Jul 16, 1:29 pm, Petry  wrote:
> Hi Ian
>
> I had been looking on django source and I changed some attributes on
> base.py file
>
> see this[1] ticket, it is works fine for me!
>
> [1]http://code.djangoproject.com/ticket/11487
>
> On 16 jul, 13:15, Ian Kelly  wrote:
>
> > On Jul 16, 7:51 am, Petry  wrote:
>
> > > Hello,
>
> > > I am working on a project where I have to store a large amount of
> > > content, (html) in a record from a table, I am using Oracle as the
> > > database
>
> > > What is very strange ... is put a text reasonably small, (around 2000
> > > characters) and it works correctly, save without problems, but
> > > doubling this size, this content is saved completely changed.
>
> > > Initially used with standard and type of field in Oracle Textfield
> > > that it generates a type NCLOB, then when i saved it adds various
> > > ideograms that believe they are Chinese. changed the type of field to
> > > CLOB and now when saved, several questions are stored in text ("??")
>
> > > About my work environment: Django 1.1 (svn), debian, and Oracle 10g
> > > with cx_oracle 4.4.1
>
> > Hi Petry,
>
> > I've heard of similar problems a couple of times 
> > (seehttp://code.djangoproject.com/ticket/9152), but so far nobody seems to
> > want to stick around long enough to figure out what's going on.  What
> > are the database character set and national character set of your
> > database?  Is the text you're trying to store str or unicode?  If str,
> > what encoding is it?
>
> > You mentioned that it fails around 4000 characters.  Does it fail just
> > in the 2001-4000 character range, just in the 4001+ character range,
> > or both?
>
> > Thanks,
> > Ian
>
>
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Checking if user is logged in and log in form

[Ryan K]

 > urlpatterns = patterns('',
> (r'^someurl/', login_required(direct_to_template), { 'template':
> 'template.html', }),
> )

http://groups.google.com/group/django-users/browse_thread/thread/e9f85ce0666da8c/

Cheers,
Ryan
http://consulting.ryankaskel.com

On Jul 19, 5:31 pm, AKK  wrote:
> Hi, i have a login form using the generic view like so:
>
> (r'^accounts/login/$', 'django.contrib.auth.views.login',
> {'template_name': 'blogSite/login.html'}),
>
> but i only want the form to show up if your not logged in, but since
> its is a generic view how can i use
>
> if not request.user.is_authenticated():
>
> as well as the view and also pass the template name.
>
> Thanks
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Checking a User in another Table ...

[Ryan K]

What exactly are your models? If you are using a ManyToMany fields you
could do it Rating.objects.get(user=request.user) and see if that
returns anything.

Cheers,
Ryan

On Jul 19, 10:06 am, The Danny Bos  wrote:
> Hey there, this should be easy, but my code just won't play along.
>
> I've let a logged in user Rate a music record. I'm saving it like so
> 'rating | user | record'. (Rating model)
> What I need to do is only display the Rating form if the logged in
> user hasn't rated the record yet, know what I mean.
>
> What code would I need to check that?
> I'm still new, albeit I feel I've read every book on the subject
> twice.
>
> d
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: ModelForm and ForeignKeys

[Ryan Bales]

I have experienced this same issue, and that link didn't really help.
I am trying to hook into a legacy database using "inspectdb", and I
can't index the tables.

class Character(models.Model):
 guid = models.IntegerField(primary_key=True)
 name = models.CharField()
 ...
 ...
class CharacterHome(models.Model):
 character = models.IntegerField(primary_key=True)
 ...
 ...

Even though django accurately assesses the datatypes of these tables,
I want to be able to set CharacterHome.character as a ForeignKey
(Character) and have Character return the "guid" integer value.


On Aug 22, 5:29 am, Léon Dignòn  wrote:
> http://docs.djangoproject.com/en/dev/topics/forms/modelforms/#using-a...
>
> Good luck!
>
> -ld
>
> On Aug 21, 3:49 pm, goobee  wrote:
>
> > hi there
>
> > I'm in dire need of a good idea. django drives me crazy with its
> > foreignkey-resolver.
>
> > class person(models.Model):
> >     name ...
> >     firstname 
> >     
> >     
>
> > class participant(models.Model):
> >     group ..
> >     person(foreignkey(person))
> >     funk ...
>
> > I want to show 'participant' using the ModelForm-feature. From model
> > 'person' I need name and firstname only (or __unicode__) of the
> > particular participant, but django delivers the entire table 'person'
> > which is an unnecessary overkill (especially with several FKs in
> > 'participant'). There must be an option to avoid this behaviour!?
>
> > thanks for any ideas

--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Strange error while trying to access admin: rendering fails (latest SVN)

[Ryan K]

I am getting a very strange error and I can't tell if it is coming
from my code or Django's code. I have not modified the templates of
the admin interface at all. Below is the error. I am trying to access
http://localhost:8080/. This error still occurrs if I remove all apps
from the install apps list and only leave contrib apps.



TemplateSyntaxError at /admin/

Caught an exception while rendering: unsupported operand type(s) for
+=: 'function' and 'list'

Request Method: GET
Request URL:http://localhost:8000/admin/
Exception Type: TemplateSyntaxError
Exception Value:

Caught an exception while rendering: unsupported operand type(s) for
+=: 'function' and 'list'

Exception Location: /usr/local/lib/python2.6/dist-packages/django/
template/debug.py in render_node, line 81
Python Executable:  /usr/bin/python
Python Version: 2.6.2

-

Cheers,
Ryan
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Strange error while trying to access admin: rendering fails (latest SVN)

[Ryan K]

This is where it says the problem is in the template:

Template error

In template /usr/local/lib/python2.6/dist-packages/django/contrib/
admin/templates/admin/base.html, error at line 30
Caught an exception while rendering: unsupported operand type(s) for
+=: 'function' and 'list'
20  
21  
22  
23  {% block branding %}{% endblock %}
24  
25  {% if user.is_authenticated and user.is_staff %}
26  
27  {% trans 'Welcome,' %}
28  {% firstof user.first_name user.username %}.
29  {% block userlinks %}
30  {% url django-admindocs-docroot as docsroot %}   <- ERROR HERE
--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Strange error while trying to access admin: rendering fails (latest SVN)

[Ryan K]

State in admin rendering process when error occurred:

/usr/local/lib/python2.6/dist-packages/django/template/debug.py in
render_node

  74. e.source = node.source
  75. raise
  76. except Exception, e:
  77. from sys import exc_info
  78. wrapped = TemplateSyntaxError(u'Caught an exception while
rendering: %s' % force_unicode(e, errors='replace'))
  79. wrapped.source = node.source
  80. wrapped.exc_info = exc_info()

  81. raise wrapped ...

  82. return result
  83.
  84. class DebugVariableNode(VariableNode):
  85. def render(self, context):
  86. try:
  87. output = force_unicode(self.filter_expression.resolve(context))

▼ Local vars
VariableValue
context
[{'block': ,
, , , ,
, , , ,
,  / '>, , , , ,
,  '>]>}, {'root_path': None, 'app_list': [{'app_url':
'auth/', 'models': [{'perms': {'add': True, 'change': True, 'delete':
True}, 'admin_url': 'auth/group/', 'name':
}, {'perms':
{'add': True, 'change': True, 'delete': True}, 'admin_url': 'auth/
user/', 'name': }], 'has_module_perms': True, 'name': 'Auth'}, {'app_url':
'sites/', 'models': [{'perms': {'add': True, 'change': True, 'delete':
True}, 'admin_url': 'sites/site/', 'name':
}],
'has_module_perms': True, 'name': 'Sites'}, {'app_url':
'staticpages/', 'models': [{'perms': {'add': True, 'change': True,
'delete': True}, 'admin_url': 'staticpages/link/', 'name': u'Links'},
{'perms': {'add': True, 'change': True, 'delete': True}, 'admin_url':
'staticpages/menu/', 'name': u'Menus'}, {'perms': {'add': True,
'change': True, 'delete': True}, 'admin_url': 'staticpages/
staticpage/', 'name': u'Static pages'}], 'has_module_perms': True,
'name': 'Staticpages'}], 'title': u'Site administration'},
{'MEDIA_URL': 'http://localhost/static/sccom'}, {'perms':
,
'messages': [], 'user': }, {}]
e
TypeError("unsupported operand type(s) for +=: 'function' and
'list'",)
exc_info

node

self
[, , , , ,
, , , ,
,  / '>, , , , ,
,  '>]
wrapped
TemplateSyntaxError(u"Caught an exception while rendering: unsupported
operand type(s) for +=: 'function' and 'list'",)

--~--~-~--~~~---~--~~
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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: Strange error while trying to access admin: rendering fails (latest SVN)

[Ryan K]

Thanks Karen you've been helping me out recently with debugging a few
things and have been a really great asset.

Cheers,
Ryan Kaskel

On Aug 26, 7:19 pm, Karen Tracey  wrote:
> On Wed, Aug 26, 2009 at 3:33 PM, Ryan K  wrote:
>
> > This is where it says the problem is in the template:
>
> > Template error
>
> > In template /usr/local/lib/python2.6/dist-packages/django/contrib/
> > admin/templates/admin/base.html, error at line 30
> > Caught an exception while rendering: unsupported operand type(s) for
> > +=: 'function' and 'list'
> > 20      
> > 21      
> > 22      
> > 23      {% block branding %}{% endblock %}
> > 24      
> > 25      {% if user.is_authenticated and user.is_staff %}
> > 26      
> > 27      {% trans 'Welcome,' %}
> > 28      {% firstof user.first_name user.username %}.
> > 29      {% block userlinks %}
> > 30      {% url django-admindocs-docroot as docsroot %}   <- ERROR HERE
>
> Double check your url patterns.  The {% url %} tag causes your entire url
> configuration to get processed, and any errors will get reported as being
> due to the {% url %} tag, even though that specific tag is blameless.  The
> problem is likely in something you recently added/changed in your url
> configuration.
>
> 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 
django-users+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/django-users?hl=en
-~--~~~~--~~--~--~---

Re: interaction of django with paypal

[Ryan Kelly]

On Sun, 2009-08-30 at 19:54 -0700, Rafael Ferreira wrote:
> you could also use this (I haven't used it myself but looks
> promising):
>
> http://github.com/johnboxall/django-paypal/tree/master


I can second this recommendation, I've been doing some basic paypal
integration and django-paypal has saved me a *heap* of mucking around.


  Ryan

-- 
Ryan Kelly
http://www.rfk.id.au  |  This message is digitally signed. Please visit
r...@rfk.id.au|  http://www.rfk.id.au/ramblings/gpg/ for details



signature.asc
Description: This is a digitally signed message part