Skip to content

BigQuery: Pin pyarrow version in tests.#8784

Closed
tswast wants to merge 2 commits intogoogleapis:masterfrom
tswast:pyarrow
Closed

BigQuery: Pin pyarrow version in tests.#8784
tswast wants to merge 2 commits intogoogleapis:masterfrom
tswast:pyarrow

Conversation

@tswast
Copy link
Contributor

@tswast tswast commented Jul 25, 2019

This is a workaround for a possible regression in pyarrow handling of NaN vs
NULL in float columns.

This is a workaround for a possible regression in pyarrow handling of NaN vs
NULL in float columns.
@tswast tswast requested review from a team and tseaver July 25, 2019 19:46
@googlebot googlebot added the cla: yes This human has signed the Contributor License Agreement. label Jul 25, 2019
Copy link
Contributor

@tseaver tseaver left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we need an open issue to remind us to revert this change when PyArrow releases a fix?

Copy link
Contributor

@tseaver tseaver left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We need to pin for unit tests, particularly.

@tswast tswast closed this Jul 25, 2019
@tswast
Copy link
Contributor Author

tswast commented Jul 25, 2019

Found some documentation about this behavior: https://arrow.apache.org/docs/python/data.html#none-values-and-nan-handling

Since we are converting from pandas (which uses NaN as a placeholder for NULL for numpy reasons), this behavior is expected. I'll open a new PR that updates the test.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

cla: yes This human has signed the Contributor License Agreement.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants