I'm new to Cassandra and I'm trying to use CQLEngine ORM to update set column which holds UDT but I can't and documentation doesn't say anything about custom types.
My code is;
class MyType(UserType):
val = columns.Text()
point = columns.Integer()
key = columns.Text()
def __init__(self, val, point, key, **values):
super().__init__(**values)
self.val = val
self.point = point
self.key = key
class MyModel(Model):
myid = columns.UUID(primary_key=True)
set_clm = columns.Set(columns.Integer)
mytype = columns.Set(UserDefinedType(MyType))
def __init__(self, set_clm, mytype, **values):
super().__init__(**values)
self.myid = uuid4()
self.set_clm = set_clm
self.mytype = mytype
s = MyModel.objects(myid="2b3adb7d-9e68-49fc-9aa0-26dbec607f9d").update(
mytype__add=set(MyType(val="1", point=2, key="3"))
)
MyModel initially holds NULL in set but when I try to update it, I get the following error:
cassandra.InvalidRequest: Error from server: code=2200 [Invalid query] message="Invalid set literal for mytype: value 'point' is not of type frozen<mytype>"
'point' is not of type frozen<mytype>
-> This part randomly changes whenever I rerun the code (next time I'd run, I'd get the same error for 'val' column etc).
Can anyone help me how I can add a UDT set?
OK. I've solved it. I'm writing it down for people who'd find it on Google.
This is the correct way of adding to a set: mytype__add={MyType(val="1", point=2, key="3")}
and also implement the __hash__
function for MyType such as:
def __hash__():
return hash(self.__repr__())
but with a smarter __hash__
function. It's just an example. Hope it helps to someone else.