Pyright type error supplying SQLAlchemy NVARCHAR length

[Philip Nye (philipnye)]

Pyright is raising a type error when I include a length for the NVARCHAR column in the below script:

import urllib.parse
import uuid

import pandas as pd
from sqlalchemy import create_engine
from sqlalchemy.types import NVARCHAR
from sqlalchemy.dialects.mssql import UNIQUEIDENTIFIER

# CREATE DATABASE CONNECTION
connection = create_engine(
    f"mssql+pyodbc:///?odbc_connect={
        urllib.parse.quote_plus(
            'DRIVER=' + <driver details here> + ';'
            'SERVER=' + <server details here> + ';'
            'DATABASE=' + <db details here> + ';'
            'UID=' + <uid details here> + ';'
            'PWD=' + <pwd details here> + ';'
            'Authentication=' + <auth details here>
        )
    }"
)

# CREATE DATAFRAME
df = pd.DataFrame(
    columns=['id', 'name'],
    data=[
        [str(uuid.uuid4()), 'hello'],
        [str(uuid.uuid4()), 'world'],
        [str(uuid.uuid4()), 'foo'],
        [str(uuid.uuid4()), 'bar'],
        [str(uuid.uuid4()), 'baz'],
    ]
)

# SAVE DATA
df.to_sql(
    'test',
    con=connection,
    dtype={
        'id': UNIQUEIDENTIFIER,
        'name': NVARCHAR(length=1024),
    },
    index=False,
)

Specifically, it raises Argument of type "dict[str, type[UNIQUEIDENTIFIER[_UUID_RETURN@UNIQUEIDENTIFIER]] | NVARCHAR]" cannot be assigned to parameter "dtype" of type "DtypeArg | None" in function "to_sql".

If I remove the length parameter on NVARCHAR the error goes away, but I need to be able to specify the length.

I'm fairly new to using type hints. Before I raise a GH issue, is anyone able to tell me if this is an issue in Pyright/an issue in SQLAlchemy/something I've implemented incorrectly.

Thanks in advance.

[Philip Nye (philipnye)]

Now raised as an issue: #10600