On Jan 27, 2013 5:58 PM, "Sayth Renshaw" wrote:
>
> Hi
>
> Just seeking a little clarification.
>
> In the Polls tutorial we get this section.
>
> >>> p = Poll.objects.get(pk=1)
>
> # Display any choices from the related object set -- none so far.
> >>> p.choice_set.all()
> []
>
> # Create three c
The regex you are using will only match a SINGLE letter or number, followed
by a dash, and then the nid.
I'm betting it will work if you have a url like a-2355.html.
Check out this SO post, it's pretty much identical to what you are looking
for:
http://stackoverflow.com/questions/2525327/regex-fo
This is probably a STATIC_URL issue. Check to make sure it is set correctly
and that you've run manage.py collectatatic and that apache, etc. is set to
serve those files directly. The dev server does some funny magic behind the
scenes to get these files served up.
-James
On Mar 25, 2013 2:12 PM, "
ested it with Django 1.6.9, syncdb does not add fields to existing
> models
>
> On Mon, Jan 12, 2015 at 5:20 PM, James Schneider
> wrote:
>
>> Adding a column to a model is supported by syncdb, no need for South
>> migrations. Any other operation (ie changing/delet
gt;
>> Thanks heaps for your help. your help is very much appreciated . I manage
>> to fix my issue. I wouldn't have done it without your help. :)
>>
>> Cheers.
>>
>> On Tue, Jan 13, 2015 at 10:24 AM, sum abiut wrote:
>>
>>> Hi James,
>>
You should be running 'python manage.py runserver' from inside the
directory of your project root , not using the django-admin.py command. The
manage.py file should have been automatically generated when you created
the project using django-admin.py. I hadn't even realized that
django-admin.py supp
Change poll_id to poll.id.
poll_id is probably not defined anywhere, so your URL tag is getting an
argument of '' (empty string).
-James
On Jan 16, 2015 11:23 AM, "Adil Erpat" wrote:
> {% if latest_poll_list %}
>
> {% for poll in latest_poll_list %}
> {{ poll.question
> }}
>
I'm guessing that if you look at the generated source code in the browser,
all of the names for the fields in the pform form's are the same, which
likely collide on the Django side (if the browser actually posts all the
copies with the same field names, not sure what the typical browser
behavior is
You can reroll your original example like this:
def __str__(self):
pk = self.pk
customer_name = None
customer_phone = None
try:
customer_name = self.lead.customer_name
customer_phone = self.lead.customer_phone
except AttributeError:
pass
return _(
Yes, it's a bit more verbose, but much easier to follow. Hope it helps.
-James
On Jan 17, 2015 2:58 PM, "Micky Hulse" wrote:
> Hi James! Thank you for the reply and for the example code!
>
> On Sat, Jan 17, 2015 at 2:52 PM, James Schneider
> wrote:
> > You can
You can definitely do it that way if all you need to do is check if values
exist and provide a default if they don't. I do this all the time for my
__str__() methods as well.
You also mentioned similar situations that may require more logic checks
(variables depending on each other or printing com
If I understand you right, you want to set MAIN_CATEGORY as a "global"
variable/setting containing a Category object with an ID of 1. Is that
right? If so...
Rather than populating a "global" variable, I would instead create a custom
model manager for the Category model:
https://docs.djangoprojec
question is:
> How do I now access the CategoryManager.get_main_category() in my
> template.html? I guess I have to pass it over to context within the view?
>
>
> Am Sonntag, 18. Januar 2015 12:28:30 UTC+1 schrieb James Schneider:
>>
>> If I understand you right, you want to set MAIN_CATEG
After a quick search, this bug looks similar, but it's for an ancient
version of Django. Might still give you some hints as to where to look.
https://code.djangoproject.com/ticket/8965
If you can replicate the bug with an empty project and your template paths
set, I would recommend filing a bug,
I second moving some (or most) of the functionality to models.py. For
instance, calculating the SHA value should be done as part of the model's
save() function, not done in the view.
Convention dictates that the only code that is placed in the view itself
should be logic related to prepping the te
I wouldn't decorate them as class methods. You would want to call them from
the objects themselves. For the save_to_disk() method, I was actually
referring to the Django save() method (
https://docs.djangoproject.com/en/1.7/topics/db/models/#overriding-predefined-model-methods
).
As you have it no
I guess I'm not clear on what you are trying to achieve. There are a couple
of scenarios to consider.
As it stands with the default contrib.auth authentication backend, sending
both the username and email address entered by the user will only work IF
the user registered/was created using their ema
Quick guess, you are visiting the site in your browser using
http://localhost:8000. I'm betting if you instead visited it using
http://127.0.0.1:8000, the AJAX query would work fine.
You have http://127.0.0.1:8000 hard coded (or being generated) everywhere.
Either change your JavaScript to use loc
That's an interesting (but understandable) requirement, which means you are
probably in for an interesting time.
Undoubtedly you'll need to roll your own authentication backend (
https://docs.djangoproject.com/en/1.7/topics/auth/customizing/#authentication-backends)
and a custom user that contains
Hmm, yes, Shibboleth will require some extra trickery in multiple views
with redirects to the respective campus portal, etc. I've never done it
myself, but I believe the API is pretty well documented.
-James
On Jan 19, 2015 1:48 PM, "James Schneider" wrote:
> That
For a pure authentication scenario where permission checks never go beyond
user.is_authenticated(), that's probably true. If all the OP is doing is
displaying data, they may be able to get away with manually associating the
campus and user within the session after, and displaying data based on
thos
r model and avoid creating separate Django
> users that shadow the LegacyUser, as I do a lot of synchronization of
> LegacyUsers with the legacy system. I'll see how it goes and post a
> solution if I get that far.
>
> Erik
>
> > Den 19/01/2015 kl. 23.24 skre
d.
-James
On Jan 19, 2015 11:32 PM, "James Schneider" wrote:
> "A pallet of authentication systems..."
>
> Yep, you work for an educational entity, as do I. :-D
>
> You can pursue the LegacyUser model as your custom user model. That would
> make your Legac
Wow, nice. You are probably right in not inheriting from PermissionMixin
and AbstractBaseUser and just re-implementing the needed functionality. I
ended up doing the same thing for my custom user and auth backend as well.
Not as convenient, but hey, it works right? ;-)
Bravo.
-James
On Jan 20, 20
Not sure what you mean.
Are you looking to build the model definitions in your models.py files, or
are you trying to take the results of a query against MSSQL and pass them
to a constructor for a model that has already been defined?
Either way, eventually you'll end up with a copy of the 'model'
The FBV's included in contrib.auth.views don't use RequestContext(), they
jump straight to TemplateResponse():
https://github.com/django/django/blob/master/django/contrib/auth/views.py#L192
IIRC, rendering a context via a RequestContext() is what triggers the
TEMPLATE_CONTEXT_PROCESSORS to run.
Hmm, I stand corrected:
https://github.com/django/django/blob/1.7.3/django/template/response.py#L156
It does render the context via RequestContext().
However, I can also point out why I thought I was right (in addition to my
apparently flawed/rushed test):
https://github.com/django/django/blob/
How many results do you typically get back from that query?
It sounds like the DB is highly optimized if you can gather results from
tables that large that quickly running the query outside of Django. If you
are returning a large dataset, then it is more likely that Django is
furiously coercing ev
th the same Django model call
several times to see if the timing drops down into the sub-second range (it
should).
-James
On Fri, Jan 23, 2015 at 12:05 AM, Erik Cederstrand <
erik+li...@cederstrand.dk> wrote:
>
> > Den 23/01/2015 kl. 08.19 skrev James Schneider >:
> >
>
If you want any sort of enforcement of a unique email address at the
database level, you'll need a custom user model. If you don't care about
that, all you'd need to do is override the user forms and create a custom
clean_email function that checks for duplicates before saving the form
data. Not 10
Instead of looping through the all() set of objects (which works quickly
with small sets of objects, imagine >20K objects, that would be noticeably
slower, probably seconds or more), you should just query the object you
want directly using something like:
obj = newleave.objects.order_by('-pk')[0]
Did you read the comments on that snippet? It appears there may need to be
some extra encoding/decoding needed (example additional code is listed
there), which would make sense given the decoding errors you are seeing.
-James
On Jan 27, 2015 9:54 AM, "Supermario" wrote:
> I try to implement this
Couple things.
I think the return statement for thankyou.html needs to get kicked in a
tab, otherwise the form will show successful every time the form is
submitted, even if it had errors.If there are errors, it will also skip
your email code, but still show as successful.
This line is very confu
That makes perfect sense as to why Google would reject it. You would need
to make your site live, and generate the site map using the real domain
name, otherwise Google has nothing to index against. The address you
provided is either a local proxy or the dev server running on port 8001 on
your loca
You've defined your class-based view as a function:
def PozoList(ListView):
You should be defining it as a class:
class PozoList(ListView):
And you should be calling the to_view() method for the PozoList class from
your urls.py.
Please post your urls.py and any traceback and error messages you
Oh, I see what you are trying to do. It looks like you are running needless
queries though.
Your newleave model has this:
username =models.ForeignKey(User, default =1)
That's an FK to User, which means this line in your view:
to_emails = [u.email for u in
User.objects.filter(username_
I don't see why not. For Apache, you would just place the
relevant WSGIScriptAlias and WSGIPythonPath inside of your VirtualHost
elements. You would need to be cautious about running a single WSGI process
though, since the first site settings read will apply to all of your Django
sites (which is pr
I wouldn't store binary data like an image in a session, especially if you
have a DB-backed session system.
I would instead use a separate model just for the images so that they can
be created on the fly by dropzone via an API call, and then you can save
the ID of the image object in the session (
Do you want code in your Django project to contact an API provided by
another site in order to display information on your site? Or are you
writing an API to provide services for end users?
If the latter, I would recommend the DRF in the link provided in the
previous response by Mike.
-James
On F
You should raise the error in your custom user login form when attempting
to authenticate the user, not directly from the custom user model or
authentication backend.
-James
On Feb 17, 2015 12:20 PM, "Tomáš Sekanina" wrote:
> Hello,
>
> I have a custom user model and I need to extend the user au
Are you displaying the data in a table? If so, I've always used
django-tables2, which includes the ability to have sorting in each of the
columns via a generated link in the header row of any or all columns
(up/down arrow, etc.).
-James
On Feb 17, 2015 2:43 PM, "ThomasTheDjangoFan" <
stefan.eichho
You should be able to handle this pretty easily with Django's built-in
permission system.
Just create a group, add the users to it, and assign the requisite
permissions to the group. Our add the needed permissions individually to
each user.
https://docs.djangoproject.com/en/1.7/topics/auth/defaul
Without more detail, a good place to start would be with the ORM
aggregation documentation:
https://docs.djangoproject.com/en/1.7/topics/db/aggregation/#filter-and-exclude
I would advise reading the whole page, though. Note that the
placement/order of filters are much more significant than with n
I would suggest reading this:
https://docs.djangoproject.com/en/1.7/topics/db/queries/#lookups-that-span-relationships
and see if that helps out. You can span relationships (in this case, using
FK's) using the double-underscore notation.
I would also check out using select_related() if you would
In most cases, including the CSRF token in AJAX requests is trivial, and
processing the token is handled automatically by Django. I would recommend
spending a few minutes to implement the CSRF protection in your code of you
plan to ever move to production. It is easy enough and pays good dividends
For global constants, I would second the strategy that Mike outlined, and
rename your constants in a more specific way, such as TASK_CANCELLED,
TASK_COMPLETE, etc. unless of course CANCELLED can apply to more than just
tasks.
The use of a separate class to store such constants is of course viable,
>>> In my case, a Task can have one or more TaskStatus. I do not wish to
store
>>> the TaskStatus in the database since they are simply constants.
>>>
>>> Does anyone have a better way of how I can approach this?
>>>
I just read this a bit closer. I think what you mean is that you don't want
a sep
Sure, you would just override the get() and post() and possibly dispatch()
methods in your classes as needed.
I always recommend the classy CBV inspector, helps immensely with
understanding the available methods and inheritance in the CBV's.
http://ccbv.co.uk/
Also check out the source code of D
A failure in name resolution indicates a DNS query failure.
>From the machine where the script is failing, can you run 'ping portal'
and/or 'nslookup portal' and see if an IP is returned (even if the ping
fails)? Are the failures occurring in the view on the same machine (and
virtual environment i
What happens when you throw in a print(obj.__dict__[field]) before the val
assignment? You should see on the console what value each iteration of the
loop is using and see if you have a strange attribute that is throwing
things out of kilter.
-James
On Feb 24, 2015 8:33 PM, "Mike Dewhirst" wrote:
I believe a2ensite is a Debian-specific shortcut that allows you to quickly
enable/disable various site configurations through the use of symlinks.
For CentOS you'll need to create a configuration file in your conf.d
directory for your virtual host. I believe it will be included
automatically by h
I assume that you are talking about the select_related() and
prefetch_related() queryset methods?
https://docs.djangoproject.com/en/1.7/ref/models/querysets/#select-related
https://docs.djangoproject.com/en/1.7/ref/models/querysets/#prefetch-related
Both of those sections have excellent examples,
esk` where every desk has an attribute names `favorite_or_nearby_chairs`
> which contains the queryset of chairs that I desrcibed, prefetched.
>
>
> Thanks,
> Ram.
>
> On Wed, Feb 25, 2015 at 11:18 PM, James Schneider > wrote:
>
>> I assume that you are talking
with `Chair.objects`? I'm not looking to get a `Chair`
>> queryset. I'm looking to get a `Desk` queryset, which has a prefetched
>> attribute `favorite_or_nearby_chairs` which contains the `Chair`
>> queryset I wrote down.
>>
>>
>> Thanks,
>> Ram
your friend in these cases, but
obviously a high number of even relatively fast queries can have a
detrimental effect on your load times. Also ensure that the fields you are
using to filter contain indexes, if appropriate/available.
-James
On Thu, Feb 26, 2015 at 10:17 AM, James Schneider
wr
Heh, I just realized that aRkadeFR had replied with a similar idea to use a
property. At least I know I'm not too far off on my thinking. :-D
-James
On Thu, Feb 26, 2015 at 11:02 AM, James Schneider
wrote:
> Whoops, accidentally sent that last one too early, here's the
ed guesses and may not be valid at all given other requirements or
design considerations in the project.
> aRkadeFR
>
>
> On 02/26/2015 08:27 PM, James Schneider wrote:
>>
>> Heh, I just realized that aRkadeFR had replied with a similar idea to
use a property. At least I
The proxy is likely forcing a TLS fallback in a way that Firefox doesn't
like. Ensure you are using the latest version of FF, but even then, unless
you have control off the proxy, there may not be much you can do except
open a support ticket with your IT department.
https://bugzilla.mozilla.org/sh
Generally this happens when you don't apply the initial migration. Have you
run 'python manage.py migrate' (assuming Django >= 1.7 since you mentioned
making migrations) and if you have, what was the output? If you aren't
sure, run it again as it shouldn't hurt anything.
Check your database and ma
all of them).
>
> >
> > Still watching the thread cause I have couple of problems
> > like this one :)
> >
>
> I don't necessarily run in to this specific problem, so I'm not sure how
> much more I can contribute. A lot of the model changes I suggested were
&
Nothing in Django specifically that I'm aware of (although that's not
saying much), but you can probably do something pretty slick with the
operator library and attrgetter:
https://docs.python.org/2/library/operator.html#operator.attrgetter
It would probably look something like this (untested):
Whoops, I meant to swap out 'item' for 'instance' to make it a bit more
readable, but got myself distracted and forgot to finish and hit send.
So, s/item/instance/...
HTH,
-James
On Mon, Mar 2, 2015 at 2:11 PM, James Schneider
wrote:
> Nothing in Django specifi
So, do objects ever get successfully added to the list? I had a bit of
trouble following your workflow and figuring out where your issue actually
lives.
When you say 'send the final list back to an HTML page' do you mean the
Django process is generating the HTML and sending it back as a response t
Is there a possibility of having more than one settings file? If so, ensure
that your manage.py is pointing at the right settings file. The output of
'manage.py runserver' should also show the dotted path of the settings file
that it is using just before that error is displayed (I believe).
You ca
Not technically a bug, although it is a common point of confusion. It
depends on how you installed Django. Both perform the same function on
their respective systems. There's a note with this link on the first page
of the tutorial, although admittedly I believe they should be more explicit
with thi
To me, the question needs more explanation. If a request is made of a
server in the US, how does it make any sense to send a database query from
that server in the US over to Luxembourg if that is in the proximity of
where the user is located? Efficiency? No, that would actually cause double
the tr
Perhaps you can investigate more into the 'why' portion of the requirement
(which is what I was interested in)? You may be able to convince them to
abandon/modify their requirement if you lay out the technical issues and
complication they are introducing, and for what gain?
I suspect the feature y
Does the site load when accessing it from your own browser (or better yet,
from a different machine connected to the Internet)? Are you pasting in an
http://127.0.0.1:8000 address? If so, that won't work with tools like
responsinator.com. Your site will need to be publicly accessible on the
Interne
You also have two different URL's named 'home', which will lead to a bad
time. Make sure that those are unique.
>From what I can tell, none of your views render/reference 'signup.html', so
I would never expect that template to be rendered.
Configure a fields attribute on your form to get rid of t
Can you please paste in a copy of your urls.py file that is referenced in
the error?
-James
On Mar 16, 2015 1:21 AM, "Piyush Sharma"
wrote:
> i was following the django 1.7 documentation and having issue in creating
> an app part 3
> please help
>
> TypeError at /polls/34/results/
>
> 'function'
That sounds more like an exact use case for model inheritance. You would
define a single model that matches the columns in both tables. You can
define that model as abstract, then have the two real models (one for each
table) inherit from that abstract model. In the definition for the child
models,
I would recommend that you use the suds-jurko fork. It should be a drop in
replacement for the suds library available via pip, but is actively
maintained. The original suds library looks to be abandoned and has some
issues parsing current WSDL definition files that suds-jurko can handle.
-James
On
e string replace to switch
> tables. I was hoping that someone had some kind of “switch table” command
> on the queryset or something similar.
>
> On Mar 17, 2015, at 3:08 AM, James Schneider
> wrote:
>
> That sounds more like an exact use case for model inheritance. You wo
I think the end of the traceback is pretty telling:
ValueError: Lookup failed for model referenced by field
quedar.LearningGroup.members: quedar.User
You have a LearningGroup model with an attribute called members which is
still referencing the original User model that was removed.
-James
On Mar
Is it a one-time import or are you looking to have a form that accepts CSV
files?
Either way, I would start with the Python csv module. Works really well and
should be able to convert your CSV file to a standard Python list of lists,
which you should be able to coerce into models, etc.
https://do
Suds is an old (abandoned) library. suds-jurko is a newer fork of it, and
may handle SSL connections better.
You may need to provide more detail as to what you are trying to do, as the
SSL portion is handled by underlying libraries in Python, not by suds-jurko
directly.
-James
On Apr 2, 2015 8:57
Name them appropriately?
You can have as many as you like. They also don't all bed to reside in the
same location. You can keep your code in one folder, your virtual env in
another, and have both of them inside of a patent folder using the name of
the project.
For virtualenvwarpper specifically,
You don't mention running 'manage.py makemigrations' or 'manage.py migrate'
in any of your output. The error you are getting indicates that Django has
access to a database, but the table schema is incorrect. If you haven't
previously run migrate, then you have no tables, which explains the error
yo
No need to switch databases.
Do you have any shell access to the server? Can you manually inspect the
database to see if the table actually exists (outside of any Django
commands)? I'm not sure what ASO offers.
-James
On Apr 19, 2015 9:19 AM, "David F" wrote:
> Or I'm using sqlite should I swit
Have you validated that there is actual data in the govstaff variable
within your template? Nothing is going to show up if it is an empty
variable or doesn't exist in your template context.
While I'm not familiar with the package you're using, I suspect you need to
add the govstaff variable to you
_data() i am getting the
> error non-keyword arg after keyword arg. Is there are other ways to
> generated pdf reports on Django?
>
> Cheers,
>
> On Thu, Apr 23, 2015 at 12:34 PM, James Schneider > wrote:
>
>> Have you validated that there is actual data in the govstaff vari
Have you seen this?
https://docs.djangoproject.com/en/1.8/topics/http/file-uploads/
Has an example showing an example exactly as you asked without any models
involved. Also make sure you recognize the difference between
models.FileField and forms.FileField:
https://docs.djangoproject.com/en/1.8/
You can save the file wherever you like, just change the file path where
you are writing the file to make use of settings.MEDIA_ROOT rather than a
hard-coded URL.
Since you aren't using a model, Django is assuming you want to do something
fancy with the file, so you are responsible for doing the h
What is the view code that you are calling to actually create the user? I
suspect you are simply doing something like:
user.password =
That will store the password in plain text in the DB, but Django is
expecting it to be hashed, so it never matches.
You should be using either the create_user()
Python 2 or Python 3?
What code ate you using to print the object? A simple print statement?
-James
On May 4, 2015 3:20 PM, "Yann Ashuach" wrote:
> In tutorial number one in django documentations, i have tried and tried
> again, and neither the __str__ or the __unicode__ is working! This has be
I'm guessing the issue is actually in your template if you are following
along that page. None of the code in your views would generate that error
from what I can see.
The error indicates that you are trying to reverse('polls:detail') and not
providing any arguments for the URL resolver (or are pa
tion} (with a Q) in views.results and see
> how it goes again.
>
> All the best. :)
>
> Sincerely,
> Muhammad
> On May 8, 2015 10:41 PM, "Luis Zárate" wrote:
>
>> Which urls.py you paste here? The project URLs or the app urls .?
>>
>> It is because
>
> I agree that the typo is also an issue and should be fixed, but that
>> wouldn't result in the OP's error, since reverse() is complaining about a
>> 'detail' URL specifically. The typo would result in a similar error when
>> the result page is displayed, and would show 'guestion' as one of the
There's no requirement that you use views.py, or any of the file names that
are created using 'django-admin.py', and you can create as many others as
you like. You can run an entire Django project from a single file if you
were so inclined. All of the file name references are just for convention
an
>
> Thanks James for the answer, my problem is to update the various import
in my app to refer to the new file with the code I moved into. At the
moment, the only thing that can help is to have all the view tested to show
an error in the import during the Test Harness.
>
> Obviously I have not
If you get enough traffic to trounce a (web server of choice) installation,
you probably are making enough money to hire an expert with that system to
tune it properly or recommend adding additional resources.
Don't get bogged down in Apache vs. Nginx vs. uWSGI, etc. You're nowhere
near that point
> asset_spec =
AssetSpecification.objects.filter(asset_id_id=inventory,utilized_value=0).values_list('asset_id',
flat=True)
> trans = transaction.savepoint() // Here I am trying
not to commit the code and store it in transaction.
>
> if
You should post the code for the view, model, and the entire traceback.
I often get messages like this when using the SuccessMessages view mixin
and I reference an invalid field name in success_messages within the view
(usually due to copy/paste from a similar view).
-James
On May 13, 2015 7:01 A
Did you try changing the position of the save point like I mentioned?
-James
On May 13, 2015 9:33 PM, "Simran Singh" wrote:
> Hi James,
>
> Thanks for your feedback. I basically want to rollback that query where I
> update the db. Basically I want to rollback all the updates that were made
> dur
I'd post a bug report. Based on the behavior you've outlined (haven't
looked at the Django source), there may have been some oversight on the
duck typing that python performs. It sounds like the migration package is
taking advantage of duck typing (since 16 == int('16')), but the validation
functio
The point of using a OneToOne relation to create a 'profile' is that the
profile is meant to contain information that is only accessed on an
individual basis (such as displaying a single users' address, etc.), and
generally not involved on bulk queries as you describe.
If your __str__() method in
My guess is a select_related('follow') call somewhere is causing the issue.
The behavior for select_related() changed in 1.8.
Can you post the entire trace back?
-James
On May 18, 2015 4:49 PM, "Galia Ladiray" wrote:
> Hi,
> I wonder if anyone can help me, or just share the pain ...
>
> I have
Caching is often the culprit in these situations. Disable it and recheck.
Obviously clear cookies and history in the browser as well.
Also, verify that your database shows the correct data state (meaning that
the fields in the database show as checked) once the models are saved,
either through the
; 1. disable memcached -> Nothing changed
> 2. Checked the databased -> Saved correctly
> 3. Checked browser cache -> Used chrome dev mode (with disable cache) ->
> Not working
>
> I'm really banging my head off on this one...:(
>
>
> Op zaterdag 23 mei 2015 11:
ce(selection, int) therefore always gave a False!!! Now I fixed the
> check to also check for long and it works :):):) so happy!
>
> Thank you once more!
>
>
> Op zondag 24 mei 2015 01:13:09 UTC+2 schreef James Schneider:
>>
>> In that case you should post up the re
1 - 100 of 759 matches
Mail list logo
