Clickhouse SQL cell bug with Int8 column type

Problems and bugs
1

Hello,
There is old problem with Int8 column type

Example

CREATE OR REPLACE TABLE test
(
    id UInt64,
    problem_type_field Int8
)
ENGINE = MergeTree
ORDER BY id;

INSERT INTO test (id, problem_type_field) Values (1,10);
INSERT INTO test (id, problem_type_field) Values (2,20);
SELECT * FROM test;

Returning

Traceback (most recent call last):
  at cell 10, line 4
  at cell 8, line 22, in _datalore_internal_python_execute_sql_query(sql_query, database_id, jwt, parameters, unsafe_parameters)
  at /opt/python/envs/minimal/lib/python3.8/site-packages/pandas/core/frame.py, line 746, in __init__(self, data, index, columns, dtype, copy)
  at /opt/python/envs/minimal/lib/python3.8/site-packages/pandas/core/internals/construction.py, line 510, in nested_data_to_arrays(data, columns, index, dtype)
  at /opt/python/envs/minimal/lib/python3.8/site-packages/pandas/core/internals/construction.py, line 875, in to_arrays(data, columns, dtype)
  at /opt/python/envs/minimal/lib/python3.8/site-packages/pandas/core/internals/construction.py, line 972, in _finalize_columns_and_data(content, columns, dtype)
ValueError: 2 columns passed, passed data had 1 columns

If convert field to Int16, works fine

SELECT id, toInt16(problem_type_field) FROM test;

Datalore Free Enterprise Docker 2023.5.1

2

Hello Timur,

Thank you for the report.
We will investigate this issue.

3

This issue has been fixed.

4

Fixed for “Free Enterprise Docker” ? Or only in cloud version?

5

The cloud Datalore has this fix already. A new Enterprise release (2023.6) with the fix will be available soon.