I have a list of querysets as follows. Is there a way to merge all these query sets of a list into one query set ? The queryset elements of this list are dynamically determined. As in the example below it is three querysets in the list. In next hypothetical iteration it could be 40 querysets.
capture_list =[<QuerySet [<Fields_Model: Fields_Model object (11)>]>,<QuerySet [<Fields_Model: Fields_Model object (12)>]>,<QuerySet [<Fields_Model: Fields_Model object (13)>]>]
You can union these with:
Fields_Model.objects.none().union(*capture_list)
Here the asterisk will thus "unpack" the querysets as individual parameters for the .union(…) call [Django-doc].
A call with .union(…) will only select distinct objects, so no duplicates. If you want to obtain the duplicates, you can the all=True parameter:
Fields_Model.objects.none().union(*capture_list, all=True)
Related
I have Profile table with a huge number of rows. I was trying to filter out profiles based on super_category and account_id (these are the fields in the model Profile).
Assume I have a list of ids in the form of bulk_account_ids and super_categories
list_of_ids = Profile.objects.filter(account_id__in=bulk_account_ids, super_category__in=super_categories).values_list('id', flat=True))
list_of_ids = list(list_of_ids)
SomeTask.delay(ids=list_of_ids)
This particular query is timing out while it gets evaluated in the second line.
Can I use .iterator() at the end of the query to optimize this?
i.e list(list_of_ids.iterator()), if not what else I can do?
I want to filter multiple collections, to return only documents who have those requirements, the problem is when there is more than one matching value in one collection, the elements shown are repeated.
FOR TurmaA IN TurmaA
FOR TurmaB IN TurmaB
FILTER TurmaA.Disciplinas.Mat >10
FILTER TurmaB.Disciplinas.Mat >10
RETURN {TurmaA,TurmaB}
Screenshot of the problem
What your query does is to iterate over all documents of the first collection, and for each record it iterates over the second collection. The applied filters reduce the number of results, but this is not how you should go about it as it is highly inefficient.
Do you actually want to return the union of the matches from both collections?
(SELECT ... UNION SELECT ... in SQL).
What you get with your current approach are all possible combinations of the documents from both collections. I believe what you want is:
LET a = (FOR t IN TurmaA FILTER t.Disciplinas.Mat > 10 RETURN t)
LET b = (FOR t IN TurmaB FILTER t.Disciplinas.Mat > 10 RETURN t)
FOR doc IN UNION(a, b)
RETURN doc
Both collections are filtered individually in sub-queries, then the results are combined and returned.
Another solution would be to store all documents in one collection Turma and have another attribute e.g. Type with a value of "A" or "B". Then the query would be as simple as:
FOR t IN Turma
FILTER t.Disciplinas.Mat > 10
RETURN t
If you want to return TurmaA documents only, you would do:
FOR t IN Turma
FILTER t.Disciplinas.Mat > 10 AND t.Type == "A"
RETURN t
BTW. I recommend to call variables different from collection names, e.g. t instead of Turma if there is a collection Turma.
Ok
i have this class in my model :
i want to get the agencys value which is a many to many on this class and store them in a list or array . Agency which store agency_id with the id of my class on a seprate table.
Agency has it's own tabel as well
class GPSpecial(BaseModel):
hotel = models.ForeignKey('Hotel')
rooms = models.ManyToManyField('Room')
agencys = models.ManyToManyField('Agency')
You can make it a bit more compact by using the flat=True parameter:
agencys_spe = list(GPSpecial.objects.values_list('agencys', flat=True))
The list(..) part is not necessary: without it, you have a QuerySet that contains the ids, and the query is postponed. By using list(..) we force the data into a list (and the query is executed).
It is possible that multiple GPSpecial objects have a common Agency, in that case it will be repeated. We can use the .distinct() function to prevent that:
agencys_spe = list(GPSpecial.objects.values_list('agencys', flat=True).distinct())
If you are however interested in the Agency objects, for example of GPSpecials that satisfy a certain predicate, you better query the Agency objects directly, like for example:
agencies = Agency.objects.filter(gpspecial__is_active=True).distinct()
will produce all Agency objects for which a GPSpecial object exists where is_active is set to True.
I think i found the answer to my question:
agencys_sp = GPSpecial.objects.filter(agencys=32,is_active=True).values_list('agencys')
agencys_spe = [i[0] for i in agencys_sp]
I have to update in C# code using MongoDB. Here I had implement 2nd level array of update in below (subBranchindex is taken in a generic list object):-
for (var index = 0; index < subBranchindex.Count; index++)
{
if (subBranchindex[index]._id == new ObjectId(subBranchid))
{
IMongoQuery queryEdit = Query.EQ("BranchOffice.SubBranchlist._id", new ObjectId(subBranchid));
UpdateBuilder update = Update.Set("BranchOffice.$.SubBranchlist."+ index +".Name",subBranch.SubName).
SafeModeResult s = dc.Collection.Update(queryEdit, update,
UpdateFlags.None, SafeMode.True);
}
}
Here 2nd level array, I was using (for loop Statement) to taken Index value for array. Next I can use 3rd, 4th and 5th level of array means more than (for loop statement) will be assign. So don't need [for loop Statement] and also don't need to assign hardcore number in index.
For example: ("BranchOffice.$.SubBranchlist.0.Name",subBranch.SubName). Here Don't Hardcore number[index] 0 or 1 or 2. "2nd" level array more than 100 record is there.
Is there any way I can use to array index value? Please explain how to solve this probelm. Please explain me with Example.
Based on your example above, my understanding of your schema is the following:
The top-level document has a BranchOffice field
BranchOffice is an array of objects
Each object within BranchOffice has an _id, SubName and SubBranchlist field
SubBranchlist is an array of objects
Each object within SubBranchlist has a Name field
Your update statement appears to be copying the SubName field to each Name field among objects within SubBranchlist (a sibling field of SubName).
Using the property path syntax to select fields through arrays (e.g. SubBranchlist.0.Name), there is no "wildcard" index that will allow you to modify Name fields among all objects in the array.
On a somewhat related note, the $ positional operator only applies to the first-matched array element, so you cannot use that to update multiple array elements. In your case, it would not be an option anyway, since you're using the positional operator for the BranchOffice array field.
You can either issue a series of update queries (for each element in SubBranchlist), or consider using $set to modify the entire SubBranchlist array in one query. The downside with using $set is that you'll need to read and write back the entire array, which may be a problem if other, concurrent operations are also issuing updates to the array.
Say I have a stored procedure that returns dataSet from 2 different tables. Example:
SELECT Customers.FirstName, Customers.LastName, SUM(Sales.SaleAmount) AS SalesPerCustomer
FROM Customers LEFT JOIN Sales
ON Customers.CustomerID = Sales.CustomerID
GROUP BY Customers.FirstName, Customers.LastName
Is there any way to get a strongly typed list as a result from this stored procedure ? Something like this:
StoredProcedure sp = myDevDB.GetCustomerSales();
List<MyCustomType> resultSet = sp.ExecuteTypedList<MyCustomType>();
How and where do I define the MyCustomType class ? How do I map its properties to the actual table columns ?
thanks,mehul
I solved it by creating a class (in the same place as all my other classes, but I didn't extend IActiveRecord, it's just a vanilla class).
Make sure the property names have exactly the same name and data type as the ones in the procedure, then call db.sproc(params).ExecuteTypedList().AsQueryable(); and it populated fine.