Fix failing unit tests for pandas and mlflow compatibility #1463
Fix failing unit tests for pandas and mlflow compatibility #1463skrawcz merged 10 commits intoapache:mainfrom
Conversation
This update introduces the tag for nodes in graph visualizations, allowing for human-readable labels while maintaining valid Python identifiers as function names. The changes include: - Documentation updates to explain the usage of in visualizations. - Modifications to the graph creation logic to utilize when available. - New tests to ensure is correctly applied in visualizations and that HTML characters are properly escaped. - Example functions demonstrating the use of in a new resource file. This feature improves the readability of visualizations for stakeholders while keeping the codebase Pythonic.
…special characters are properly escaped in the label.
This update improves the handling of the display_name tag in graph visualizations by adding support for cases where display_name is a list. The first element of the list will now be used as the display name. Additionally, new tests have been added to verify this functionality, ensuring that the correct display names are rendered in the graph output. This enhancement contributes to better readability and usability of visualizations.
- Update PeriodIndex API to use from_fields() for pandas 2.2+ compatibility - Add pytest.importorskip for mlflow tests on Python 3.14+ Fixes apache#1450
Some more unit test failures -- seems like another library release update that changed dependencies... created #1466 |
|
and another test failure for python 3.14 -- need to exclude ray.
created created #1467 |
2 participants
pd.PeriodIndex()topd.PeriodIndex.from_fields()for pandas 2.2+pytest.importorskip("mlflow")for Python 3.14+ where mlflow isn't installedFixes #1450