class Seller(object):
type = ...
name = ...
cars = models.ManyToManyField(Car)
class PotentialBuyer(object):
name = ...
cars = models.ManyToManyField(Car)
class Car(object):
extra_field = ...
extra_field2 = ...
Suppose I have a relationship like this. I would like to use extra queryset modifier to get the list of cars that are already been picked out by PotentialBuyers when I fetch a seller object. I suppose the query queryset will something like this.
def markPending(self)
return self.extra(select={'pending': 'select images from PotentialBuyer as t ...'})
How can I accomplish this? Is there a better way? I could fetch the seller object and the potential object and do sets, but I'd think it would be cleaner to make it handled by the database. I am using PostgreSQL 9.5.
I think the Exists subquery expression will do what you want. Or at least it'll get you started on the right path. Docs Or you might want to use an aggregate to count the number of them.
Edit: If you need to select the full objects rather than the count, existence or a single entity, then use a Prefetch instance in prefetch_related. https://docs.djangoproject.com/en/2.0/ref/models/querysets/#django.db.models.Prefetch
Not quite the answer, but this is the solution I ended up with and I am satisfied by the performance. Perhaps someone can answer the question later :
from api.models import PotentialBuyer
potentials = PotentialBuyer.objects.filter(owner=user_id, default=True).first().cars.all()
Car.objects.filter(....).annotate(pending=Case(When(id__in=potentials, then=Value(True)), default=Value(False), output_field=BooleanField()))
Related
So, i would like to filter my query by exact match in result of multiset. Any ideas how to do it in JOOQ?
Example:
val result = dsl.select(
PLANT_PROTECTION_REGISTRATION.ID,
PLANT_PROTECTION_REGISTRATION.REGISTRATION_NUMBER,
PLANT_PROTECTION_REGISTRATION.PLANT_PROTECTION_ID,
multiset(
select(
PLANT_PROTECTION_APPLICATION.ORGANISM_ID,
PLANT_PROTECTION_APPLICATION.ORGANISM_TEXT
).from(PLANT_PROTECTION_APPLICATION)
.where(PLANT_PROTECTION_APPLICATION.REGISTRATION_ID.eq(PLANT_PROTECTION_REGISTRATION.ID))
).`as`("organisms")
).from(PLANT_PROTECTION_REGISTRATION)
// here i would like to filter my result only for records that their organisms contain specific
// organism id
.where("organisms.organism_id".contains(organismId))
I've explained the following answer more in depth in this blog post
About the MULTISET value constructor
The MULTISET value constructor operator is so powerful, we'd like to use it everywhere :) But the way it works is that it creates a correlated subquery, which produces a nested data structure, which is hard to further process in the same SQL statement. It's not impossible. You could create a derived table and then unnest the MULTISET again, but that would probably be quite unwieldy. I've shown an example using native PostgreSQL in that blog post
Alternative using MULTISET_AGG
If you're not nesting things much more deeply, how about using the lesser known and lesser hyped MULTISET_AGG alternative, instead? In your particular case, you could do:
// Using aliases to make things a bit more readable
val ppa = PLANT_PROTECTION_APPLICATION.as("ppa");
// Also, implicit join helps keep things more simple
val ppr = ppa.plantProtectionRegistration().as("ppr");
dsl.select(
ppr.ID,
ppr.REGISTRATION_NUMBER,
ppr.PLANT_PROTECTION_ID,
multisetAgg(ppa.ORGANISM_ID, ppa.ORGANISM_TEXT).`as`("organisms"))
.from(ppa)
.groupBy(
ppr.ID,
ppr.REGISTRATION_NUMBER,
ppr.PLANT_PROTECTION_ID)
// Retain only those groups which contain the desired ORGANISM_ID
.having(
boolOr(trueCondition())
.filterWhere(ppa.ORGANISM_ID.eq(organismId)))
.fetch()
I'm having an issue getting a query to work.
I'm essentially trying to write something like the following SQL, with the literal 5 replaced with a variable:
SELECT *
FROM "my_table"
WHERE 5 BETWEEN "my_table"."minimum" AND "my_table"."maximum"
This is what I have at the moment:
MyModel.where(
Arel::Nodes::Between.new(
my_variable, (MyModel.arel_table[:minimum]).and(MyModel.arel_table[:maximum])
)
)
Please ignore the way I am using arel_table, the actual query has multiple joins and is more complex, but this is the most minimum reproducible example I have to demonstrate the problem.
The error, as in the subject of the question is as follows:
undefined method `and' for #<Arel::Attributes::Attribute:0x00007f55e15514f8>
and method is for Arel::Nodes::Node i.e. MyModel.arel_attribute[:name].eq(Arel::Nodes::Quoted.new('engineersmnky')) This is an Arel::Nodes::Equality and you can chain with and.
That being said you can construct an Arel::Nodes::And for yourself via
Arel::Nodes::And.new([left,right])
Then we can pass this to the Between class like so
Arel::Nodes::Between.new(
Arel::Nodes::Quoted.new(my_variable),
Arel::Nodes::And.new([
MyModel.arel_table[:minimum],
MyModel.arel_table[:maximum]])
)
The Arel::Nodes::Quoted (also: Arel::Nodes.build_quoted(arg)) is not needed in your case since your my_variable is an Integer which can be visited and will be treated as an Arel::Nodes::SqlLiteral but I find it best to let arel decide how to handle the quoting in case your my_variable ends up being some other un-visitable Object
There are other ways to create a Between and other ways to create an And depending on what objects you are dealing with.
between is a Arel::Predication and these predications are available to Arel::Nodes::Attribute objects e.g.
MyModel.arel_table[:minimum].between([1,6])
and as mentioned is available to Arel::Nodes::Node and instances of this class provides a convenience method (create_and) for creating an And so we could do the following:
Arel::Nodes::Node.new.create_and([
MyModel.arel_table[:minimum],
MyModel.arel_table[:maximum]])
There are a number of other ways to hack this functionality together by using other Arel classes but this should get you headed in the right direction.
I am using STI and have a table Widget that a bunch of other subclasses inherit from using STI. I want a query that gets the last created of each object with a uniq type.
I have a predefined array of types I want so lets say:
types = ["type1", "type2", "type3"]
So my query would be something like: (this doesnt work)
Widget.where(type: types).uniq.reverse
My goal is to get the last of each object that matches 1 of those types..
Not 100% sure, but something like this might work (untested):
ids = Thing.where(type: types).group(:type).maximum(:id).values
last_per_type = Thing.find(ids)
Thing.select("distinct type")
By the way, type is a special variable in rails and can't be used as a column name.
Given two models:
class Post(models.Model):
...
class Comment(models.Model):
post = models.ForeignKey(Post)
...
Is there a good way to get the last 3 Comments on a set of Posts (i.e. in a single roundtrip to the DB instead of once per post)? A naive implementation to show what I mean:
for post in Post.objects.filter(id__in=some_post_ids):
post.latest_comments = list(Comment.objects.filter(post=post).order_by('-id')[:3])
Given some_post_ids == [1, 2], the above will result in 3 queries:
[{'sql': 'SELECT "myapp_post"."id" FROM "myapp_post" WHERE "myapp_post"."id" IN (1, 2)', 'time': '0.001'},
{'sql': 'SELECT "myapp_comment"."id", "myapp_comment"."post_id" FROM "myapp_comment" WHERE "myapp_comment"."post_id" = 1 LIMIT 3', 'time': '0.001'},
{'sql': 'SELECT "myapp_comment"."id", "myapp_comment"."post_id" FROM "myapp_comment" WHERE "myapp_comment"."post_id" = 2 LIMIT 3', 'time': '0.001'}]
From Django's docs:
Slicing. As explained in Limiting QuerySets, a QuerySet can be sliced, using Python’s array-slicing syntax. Slicing an unevaluated QuerySet usually returns another unevaluated QuerySet, but Django will execute the database query if you use the “step” parameter of slice syntax, and will return a list. Slicing a QuerySet that has been evaluated (partially or fully) also returns a list.
Your naive implementation is correct and should only make one DB query. However, don't call list on it, I believe that will cause the DB to be hit immediately (although it should still only be a single query). The queryset it already iterable and there really shouldn't be any need to call list. More on calling list from the same doc page:
list(). Force evaluation of a QuerySet by calling list() on it. For example:
entry_list = list(Entry.objects.all())
Be warned, though, that this could have a large memory overhead, because Django will load each element of the list into memory. In contrast, iterating over a QuerySet will take advantage of your database to load data and instantiate objects only as you need them.
UPDATE:
With your added explanation I believe the following should work (however, it's untested so report back!):
post.latest_comments = Comment.objects.filter(post__in=some_post_ids).order_by('-id')
Admittedly it doesn't do the limit of 3 comments per post, I'm sure that's possible but can't think of the syntax off the top of my head. Also, remember you can always do a manual query on any Model to get better optimisation, so you can run Comment.query("Select ...;")
Given the information here on the "select top N from group" problem, if you're IN clause will be a small number of posts, it may just be cheaper to either a) do the multiple queries or b) select all comments for the posts then filter in Python. I'd suggest using a if it's a small number of posts with lots of comments and b if there will be relatively few comments per post.
Here is the single line from one of my functions to test if any objects in my array have a given property with a matching value
Return ((From tag In DataCache.Tags Where (tag.FldTag = strtagname) Select tag).Count = 1)
WHERE....
DataCache.Tags is an array of custom objects
strtagname = "brazil"
and brazil is definitely a tag name stored within one of the custom objects in the array.
However the function continually returns false.
Can someone confirm to me that the above should or should not work.
and if it wont work can someone tell me the best way to test if any of the objects in the array contain a property with a specific value.
I suppose in summary I am looking for the equivalent of a SQL EXISTS statement.
Many thanks in hope.
Your code is currently checking whether the count is exactly one.
The equivalent of EXISTS in LINQ is Any. You want something like:
Return DataCache.Tags.Any(Function(tag) tag.FldTag = strtagname)
(Miraculously it looks like that syntax may be about right... it looks like the docs examples...)
Many Thanks for the response.
Your code did not work. Then I realised that I was comparing to an array value so it would be case sensitive.
However glad I asked the question, as I found a better way than mine.
Many thanks again !