Inserting boolean NULL values in BigQuery table — needs a test added

[mdimrankhan112]

For Boolean datatype, Big Query supports True,False and NULL values. However I am not able to insert NULL values using the dataFrame.to_gbq() method.

I tried with Boolean DataFrame column containing True, False, None values and with True, False, NaN values. Nothing seems to work.

Is there any method to do this or atleast a work around.

Thanks in advance!

[max-sixty]

Pandas doesn't support null values in bool columns.

There's an issue to be able to supply a schema; a schema of bool on an object column of True, False and NaN would work

As a hack, you could load a float column with 0, 1, and NaN, and run a query on BQ to cast as a bool

[tswast]

Possibly fixed by #445 which adds support for the "new" boolean dtype, which does support NULL values.

[tswast]

Re-opening, because we need a test case here: https://github.com/googleapis/python-bigquery-pandas/blob/main/tests/system/test_to_gbq.py

[tswast]

Test added:

python-bigquery-pandas/tests/system/test_to_gbq.py

Lines 141 to 144 in 78aa01e

	"boolean_col": pandas.Series(
	[None, True, False],
	dtype="boolean",
	),

as part of

python-bigquery-pandas/tests/system/test_to_gbq.py

Lines 141 to 144 in 78aa01e

	"boolean_col": pandas.Series(
	[None, True, False],
	dtype="boolean",
	),