SNOW-2102704: described columns do not match query columns

[padhia]

Python version

3.12

Operating system and processor architecture

macOS-15.4.1-arm64-arm-64bit

Installed packages

asn1crypto==1.5.1
boto3==1.36.21
botocore==1.36.21
build==1.2.2.post1
certifi==2025.1.31
cffi==1.17.1
charset-normalizer==3.4.1
cloudpickle==3.1.1
cryptography==44.0.2
filelock==3.18.0
idna==3.10
iniconfig==2.1.0
jmespath==1.0.1
packaging==24.2
platformdirs==4.3.7
pluggy==1.5.0
protobuf==5.29.4
pycparser==2.22
PyJWT==2.10.1
pyOpenSSL==25.0.0
pyproject_hooks==1.2.0
pytest==8.3.5
python-dateutil==2.9.0.post0
pytz==2025.2
PyYAML==6.0.2
requests==2.32.3
s3transfer==0.11.2
setuptools==78.1.0.post0
six==1.17.0
snowflake-connector-python==3.15.0
snowflake-snowpark-python==1.31.1
sortedcontainers==2.4.0
tomlkit==0.13.2
typing_extensions==4.13.0
tzlocal==5.3.1
urllib3==2.3.0
wheel==0.45.1

What did you do?

import os
from snowflake.connector import connect

with connect(connection_name=os.environ["SNOWFLAKE_DEFAULT_CONNECTION_NAME"]) as cnx, cnx.cursor() as csr:
    csr.describe("show applications")
    describe_columns = set(c.name for c in csr.description)

    csr.execute("show applications")
    execute_columns = set(c.name for c in csr.description)

    print(describe_columns ^ execute_columns)

What did you expect to see?

Metadata from the SQL should match regardless of whether the query is run or just described.

Note: This happens only with selected statements, such as show applications. The problem may lie with the underlying SQL engine; if so please let the appropriate team know.

Can you set logging to DEBUG and collect the logs?

import logging
import os

for logger_name in ('snowflake.connector',):
    logger = logging.getLogger(logger_name)
    logger.setLevel(logging.DEBUG)
    ch = logging.StreamHandler()
    ch.setLevel(logging.DEBUG)
    ch.setFormatter(logging.Formatter('%(asctime)s - %(threadName)s %(filename)s:%(lineno)d - %(funcName)s() - %(levelname)s - %(message)s'))
    logger.addHandler(ch)

[sfc-gh-dszmolka]

I remember discussing this internally and coming to the conclusion that the columns are coming from Snowflake backend; and driver just consumes them blindly. We discussed the possible way forward (with the backend team)

Do you require any further assistance on this one from the driver side? (This is the driver library repository)

[padhia]

As per our communication, I'll report this to the SQL team. Thank you