-
Notifications
You must be signed in to change notification settings - Fork 28
Open
Labels
bugSomething isn't workingSomething isn't workingpr-size: smallMinimal code updateMinimal code updatetriage doneIssues that are triaged by dev team and are in investigation.Issues that are triaged by dev team and are in investigation.under development
Description
Describe the bug
From this discussion: https://www.reddit.com/r/MicrosoftFabric/comments/1p5oduh/comment/nquk0e7/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button
Polars data frames returns the following error:
ComputeError: could not append value: 2013-01-01 of type: date to the builder; make sure that all rows have the same schema or consider increasing `infer_schema_length`
it might also be that a value overflows the data-type's capacity
To reproduce
import struct
from os import getenv
from dotenv import load_dotenv
import polars as pl
from pyodbc import connect
def main():
load_dotenv()
# Establish a connection to the SQL Server database
connection = connect(getenv("SQL_CONNECTION_STRING"))
tables = ["Customers", "Invoices", "Orders"]
for table in tables:
query = f"SELECT TOP 5 * FROM Sales.{table}"
pl_data = pl.read_database(query=query, connection=connection)
print(pl_data)
# Close the connection
connection.close()
if __name__ == "__main__":
main()Expected behavior
Script would run without error
Further technical details
Python version: any
SQL Server version: any
Operating system: any
Additional context
The database is the World Wide Importers Sample database
Metadata
Metadata
Assignees
Labels
bugSomething isn't workingSomething isn't workingpr-size: smallMinimal code updateMinimal code updatetriage doneIssues that are triaged by dev team and are in investigation.Issues that are triaged by dev team and are in investigation.under development