You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

6.5 KiB

Errors

This page lists errors that Redis OM might generate while you're using it, with more context about the error.

E1

In order to query on a list field, you must define the contents of the list with a type annotation, like: orders: List[Order].

You will see this error if you try to use an "IN" query, e.g., await TarotWitch.find(TarotWitch.tarot_cards << "death").all(), on a field that is not a list.

In this example, TarotWitch.tarot_cards is a list, so the query works:

from typing import List

from redis_om import JsonModel, Field

class TarotWitch(JsonModel):
    tarot_cards: List[str] = Field(index=True)

But if tarot_cards was not a list, trying to query with << would have resulted in this error.

E2

You tried sort by {field_name}, but {self.model} does not define that field as sortable.

You tried to sort query results by a field that is not sortable. Here is how you mark a field as sortable:

from typing import List

from redis_om import JsonModel, Field

class Member(JsonModel):
    age: int = Field(index=True, sortable=True)

NOTE: Only an indexed field can be sortable.

E3

You tried to do a full-text search on the field '{field.name}', but the field is not indexed for full-text search. Use the full_text_search=True option.

You can make a full-text search with the module (%) operator. Such a query looks like this:

from redis_om import JsonModel, Field

class Member(JsonModel):
    bio: str = Field(index=True, full_text_search=True, default="")

Member.find(Member.bio % "beaches").all()

If you see this error, it means that the field you are querying (bio in the example) is not indexed for full-text search. Make sure you're marking the field both index=True and full_text_search=True, as in the example.

E4

Only lists and tuples are supported for multi-value fields.

This means that you marked a field as index=True, but the field is not a type that Redis OM can actually index.

Specifically, you probably used a subscripted annotation, like Dict[str, str]. The only subscripted types that OM can index are List and Tuple.

E5

Only equals (=), not-equals (!=), and like() comparisons are supported for TEXT fields.

You are querying a field you marked as indexed for full-text search. You can only query such fields with the operators for equality (==), non-equality (!=), and like (%).

from redis_om import JsonModel, Field

class Member(JsonModel):
    bio: str = Field(index=True, full_text_search=True, default="")

# Equality
Member.find(Member.bio == "Programmer").all()

# Non-equality
Member.find(Member.bio != "Programmer").all()

# Like (full-text search). This stems "programming"
# to find any matching terms with the same stem,
# "program".
Member.find(Member.bio % "programming").all()

E6

You tried to query by a field ({field_name}) that isn't indexed.

You wrote a query using a model field that you did not make indexed. You can only query indexed fields. Here is example code that would generate this error:

from redis_om import JsonModel, Field

class Member(JsonModel):
    first_name: str 
    bio: str = Field(index=True, full_text_search=True, default="")

# Raises a QueryNotSupportedError because we did not make
# `first_name` indexed!
Member.find(Member.first_name == "Andrew").all()

Fix this by making the field indexed:

from redis_om import JsonModel, Field

class Member(JsonModel):
    first_name: str = Field(index=True)
    bio: str = Field(index=True, full_text_search=True, default="")

# Raises a QueryNotSupportedError because we did not make
# `first_name` indexed!
Member.find(Member.first_name == "Andrew").all()

E7

A query expression should start with either a field or an expression enclosed in parentheses.

We got confused trying to parse your query expression. It's not you, it's us! Some code examples might help...

from redis_om import JsonModel, Field

class Member(JsonModel):
    first_name: str = Field(index=True)
    last_name: str = Field(index=True)

    
# Queries with a single operator are usually simple:
Member.find(Member.first_name == "Andrew").all()

# If you want to add multiple conditions, you can AND
# them together by including the conditions one after
# another as arguments.
Member.find(Member.first_name=="Andrew",
            Member.last_name=="Brookins").all()

# Alternatively, you can separate the conditions with
# parenthesis and use an explicit AND.
Member.find(
    (Member.first_name == "Andrew") & ~(Member.last_name == "Brookins")
).all()

# You can't use `!` to say NOT. Instead, use `~`.
Member.find(
    (Member.first_name == "Andrew") & 
    ~(Member.last_name == "Brookins")  # <- Notice, this one is NOT now!
).all()

# Parenthesis are key to building more complex queries,
# like this one.
Member.find(
    ~(Member.first_name == "Andrew")
    & ((Member.last_name == "Brookins") | (Member.last_name == "Smith"))
).all()

# If you're confused about how Redis OM interprets a query,
# use the `tree()` method to visualize the expression tree
# for a `FindQuery`.
query = Member.find(
    ~(Member.first_name == "Andrew")
    & ((Member.last_name == "Brookins") | (Member.last_name == "Smith"))
)
print(query.expression.tree)
"""
           ┌first_name
    ┌NOT EQ┤
    |      └Andrew
 AND┤
    |     ┌last_name
    |  ┌EQ┤
    |  |  └Brookins
    └OR┤
       |  ┌last_name
       └EQ┤
          └Smith
"""

E8

You can only combine two query expressions with AND (&) or OR (|).

The only two operators you can use to combine expressions in a query are & and |. You may have accidentally used another operator, or Redis OM might be confused. Make sure you are using parentheses to organize your query expressions.

If you are trying to use "NOT," you can do that by prefixing a query with the ~ operator, like this:

from redis_om import JsonModel, Field

class Member(JsonModel):
    first_name: str = Field(index=True)
    last_name: str = Field(index=True)

    
# Find people who are not named Andrew.
Member.find(~(Member.first_name == "Andrew")).all()

Note that this form requires parenthesis around the expression that you are "negating." Of course, this example makes more sense with !=:

from redis_om import JsonModel, Field

class Member(JsonModel):
    first_name: str = Field(index=True)
    last_name: str = Field(index=True)

    
# Find people who are not named Andrew.
Member.find(Member.first_name != "Andrew").all()

Still, ~ is useful to negate groups of expressions surrounded by parentheses.