LGTM
On Thu, Nov 7, 2013 at 8:40 AM, Elyézer Rezende wrote:
> +1
>
> This gives an overview that what behaviours users could find when dealing
> with some use cases of the ORM.
>
>
>
> On Wed, Nov 6, 2013 at 6:14 PM, Anssi Kääriäinen
> wrote:
>>
>>
>>
>> On Tuesday, November 5, 2013 1:02:41 PM U
+1
This gives an overview that what behaviours users could find when dealing
with some use cases of the ORM.
On Wed, Nov 6, 2013 at 6:14 PM, Anssi Kääriäinen wrote:
>
>
> On Tuesday, November 5, 2013 1:02:41 PM UTC+2, Anssi Kääriäinen wrote:
>>
>> Adding something about this to release notes s
On Tuesday, November 5, 2013 1:02:41 PM UTC+2, Anssi Kääriäinen wrote:
>
> Adding something about this to release notes shouldn't hurt anybody.
>
As usual, I am a bit late. Luckily release notes can be changed after
release.
Proposed changes here: https://github.com/django/django/pull/1887
Adding something about this to release notes shouldn't hurt anybody.
I will aim for a more generic wording about changes in the ORM. For
example, there are likely cases where the generated join aliases of the
query aren't the same as they were in 1.5. This could affect .extra()
users for examp
I wouldn't ask for a bug that has long existed to be considered a
release blocker. However a behavior change with no documentation is
not a nice thing to do to your users, it seems to me Elyézer makes a
good suggestion, informing users of the known bug in the release notes
seems quite reasonable.
Maybe it could be add to a "Known Bugs" or something like that section in
the release notes?
On Tue, Nov 5, 2013 at 3:42 AM, Anssi Kääriäinen wrote:
> On Monday, November 4, 2013 8:16:12 PM UTC+2, jga...@gmail.com wrote:
>>
>> I was going to file a ticket in trac about this and found this one(
>
On Monday, November 4, 2013 8:16:12 PM UTC+2, jga...@gmail.com wrote:
>
> I was going to file a ticket in trac about this and found this one(
> https://code.djangoproject.com/ticket/21192) which seems related. The
> thing is that one was supposedly resolved 5 weeks ago, which would mean
> that fi
I was going to file a ticket in trac about this and found this one(
https://code.djangoproject.com/ticket/21192) which seems related. The thing
is that one was supposedly resolved 5 weeks ago, which would mean that fix
would be in 1.6rc1...
Should I reopen that ticket or file a new one?
On Monda
On Monday, November 4, 2013 7:24:47 PM UTC+2, jga...@gmail.com wrote:
>
> Anssi,
>
> Thanks for helping.
> I'm sorry to say that your answer went somewhat over my head, my
> proficiency with SQL is lacking.
>
> What I understood from your explanation:
> - A filter/exclude that traverses a 1:N r
I managed to get the desired behavior by doing the following ugly query:
q_obj = (Q(a__confirmation=True) & Q(a__state=1)) | (Q(a__state__gt=1) &
Q(a__state__lt=1))
bees = B.objects.filter(q_obj)
This is obviously not an ideal solution but is working for me so far...
On Monday, November 4, 2013
Anssi,
Thanks for helping.
I'm sorry to say that your answer went somewhat over my head, my
proficiency with SQL is lacking.
What I understood from your explanation:
- A filter/exclude that traverses a 1:N relationship(such as foreign key)
should target the same row with all of its criteria(kw
On Monday, November 4, 2013 6:06:55 PM UTC+2, Anssi Kääriäinen wrote:
>
> I'll look into this.
>
The situation is that this query didn't work properly in 1.5.x, but this
doesn't work properly in 1.6.x either.
The basic problem here is that in 1.5.x .exclude(Q(anything)) didn't work
correctly.
jango 1.6RC1 exclude behavior change
I've found what looks like a serious behavior change in the exclude queryset
method from Django 1.5.5 to Django 1.6 rc1.
It seems that on 1.5.5 exclude when traversing relationships only excluded
items if all criteria on the kwargs were matched on the sa
I've found what looks like a serious behavior change in the exclude
queryset method from Django 1.5.5 to Django 1.6 rc1.
It seems that on 1.5.5 exclude when traversing relationships only excluded
items if all criteria on the kwargs were matched on the same related item.
On 1.6rc1 it excludes it
14 matches
Mail list logo
