#7231 closed New feature (wontfix)
New "join" parameter for the "extra" QuerySet method
Reported by: | Davide "Design" Muzzarelli | Owned by: | nobody |
---|---|---|---|
Component: | Database layer (models, ORM) | Version: | dev |
Severity: | Normal | Keywords: | queryset extra left join |
Cc: | elsdoerfer@…, sciyoshi@…, ssadler@…, bendavis78@…, simonotron, real.human@… | Triage Stage: | Design decision needed |
Has patch: | yes | Needs documentation: | yes |
Needs tests: | yes | Patch needs improvement: | yes |
Easy pickings: | no | UI/UX: | no |
Description
This "join" parameter for the "extra" QuerySet method add the possibility to use any join type.
The implementation is simple and resolve just the base needs of custom SQL queries without lose the capabilities of the Django ORM.
This feature could open the possibility of the fast translation of models and many other uses like speed optimizations in views.
An example from the News application:
class News(models.Model): pub_date = models.DateField() class NewsTranslation(model.Model): title = models.CharField(max_length=128) body = models.TextField() language_code = models.CharField(max_length=2, core=True) news = models.ForeignKey(News)
Getting all the news translated in english:
News.objects.extra( select={'title': 'news_newstranslation.title', 'body': 'news_newstranslation.body'}, join=['LEFT JOIN news_newstranslation ON (news_news.id = news_newstranslation.news_id AND news_newstranslation.language_code = \'en\')'] )
The result of the query is a list of News with "title" and "body" attributes. The News without translation are also correctly fetched with the attributes "title" and "body" set at None.
The patch is post queryset-refactor.
Attachments (8)
Change History (33)
by , 17 years ago
Attachment: | extra-join.diff added |
---|
comment:1 by , 16 years ago
milestone: | → post-1.0 |
---|---|
Triage Stage: | Unreviewed → Design decision needed |
comment:2 by , 16 years ago
Needs documentation: | set |
---|---|
Needs tests: | set |
comment:4 by , 16 years ago
Owner: | changed from | to
---|---|
Status: | assigned → new |
by , 16 years ago
Attachment: | extra-join-1.0.diff added |
---|
New parameter "join" for the QuerySet.extra() function. For Django-1.0.
comment:5 by , 16 years ago
Cc: | added |
---|
comment:8 by , 16 years ago
Cc: | added |
---|
by , 15 years ago
Attachment: | extra-join-1.1b.2.diff added |
---|
Modified extra join which will not interfere with django multilingual.
comment:9 by , 15 years ago
Anyone willing to write a test for this? Would be nice to have this in for 1.2...
comment:10 by , 15 years ago
Component: | Core framework → Database layer (models, ORM) |
---|---|
Keywords: | left added |
Updated patch for Django v1.1.
by , 15 years ago
Attachment: | extra-join-1.2-svn-12906.diff added |
---|
Updated for Django version 1.2, svn build number 12906
comment:11 by , 15 years ago
Patch needs improvement: | set |
---|
Latest patch (extra-join-1.2-svn-12906.diff
) is severely broken.
comment:12 by , 15 years ago
Resolution: | → wontfix |
---|---|
Status: | new → closed |
extra() is already a nightmare to manage; I can't see how adding a join argument will make anything easier.
We added raw queries to get around cases where there are complex join requirements but you want to retain the capabilities of the ORM. I'm closing this proposal wontfix in the light of that addition.
comment:13 by , 15 years ago
I have to disagree with russellm about this patch. Sometimes raw queries are not reliable and even harder to maintain.
The extra() covers almost every piece of a query EXCEPT for the join piece and I believe this patch completes it.
In my case I had to replace the auto-generated sub-query SELECT...IN(SELECT...) with a INNER JOIN. By doing this my query improved 100x times on a really small, non-production, database.
Maybe if Django would've optimized the sub-query generation when using the 'in' lookup I wouldn't have needed this.
I looked at creating a custom aggregate, but that's just a pain in the ass and I think there's more work to be done in this area.
Thanks for the patch.
comment:14 by , 15 years ago
Just putting in my 2 cents for jonozz here: I need to do subquery in the join clause of a RESTful application I'm building in with Django + Django Piston. Portability isn't a major (or even minor) issue for this app. I could fall back to raw SQL querying, but this seems like a much cleaner, easier-to-understand-and-maintain solution. It would be nice if the patch supported passing parameters to the custom join statement, as with params and select_params.
comment:15 by , 14 years ago
Cc: | added |
---|
Adding my vote for re-opening this patch. I've been using this patch for nearly two years, and have found it to be useful in several different cases. I agree with jonozzz that it completes the .extra() functionality, and allows for more flexibility than raw() can provide. For example, if I'm dynamically building my query using django, and I need to add custom joins to the existing queryset, it's much cleaner to use .extra() instead of .raw().
comment:16 by , 14 years ago
+1 for us working on support.mozilla.com. We use left joins quite a bit (think optional attrs in a key/value store table), and it would be great to use the ORM for at least part of those queries.
comment:18 by , 13 years ago
Easy pickings: | unset |
---|---|
Severity: | → Normal |
Type: | → Uncategorized |
UI/UX: | unset |
I can't reply to the discussion linked on Google Groups anymore, but wanted to add my comment that I have found many uses for .extra()
and also hit the limitation of being unable to add left outer joins to a query generated by the ORM. I would not like to see it disappear in favour of raw SQL.
My current use case is for joining reverse generic relations on queries against a model that does not have a GenericRelation
field back to the generic model (because I don't know which models it will be used with -- I can't add models to 3rd party code).
To require the whole query in raw SQL in order to join generic relations would mitigate the pluggability of the generic model in question.
comment:19 by , 13 years ago
Just throwing in my vote that this really completes extra, and is a better tradeoff IMHO than switching to raw sql queries in many cases. I would like to see this in Django.
comment:20 by , 13 years ago
Cc: | added |
---|
comment:21 by , 13 years ago
Resolution: | wontfix |
---|---|
Status: | closed → reopened |
comment:22 by , 13 years ago
Resolution: | → wontfix |
---|---|
Status: | reopened → closed |
Type: | Uncategorized → New feature |
Please don't anonymously re-open tickets closed by a core developer. If you want to argue about why this ticket should be re-opened feel free to discuss here. However I think there's already a clear consensus on not adding this parameter.
comment:23 by , 12 years ago
If anyone else wants or needs this functionality, I've updated the patch on a branch at GitHub. I'll try to keep it updated periodically as long as I'm still using it myself.
https://github.com/thirstydigital/django/tree/tickets/7231-extra-join
comment:24 by , 12 years ago
Cc: | added |
---|
comment:25 by , 10 years ago
Vating to have this patch. INNER JOIN is very common in applications and django have to implement better support for it.
New parameter "join" for the QuerySet.extra() function.