The GitHub Actions job "Required Checks" on texera.git/fix/numpy-scalar-coercion has failed. Run started by GitHub user xuang7 (triggered by xuang7).
Head commit for run: 7e301370d66abade86e8f80bfc8ce6d05382f09e / eugenegujing <[email protected]> fix(pyamber): bound numpy integers into LONG by int64 range, not the float64 window Follow-up to a PR review finding. cast_to_schema() range-checked both integral floats and numpy integers against INTEGRAL_TYPE_RANGES, whose LONG bound (2**53-1) exists only because float64 cannot exactly represent integers above 2**53. numpy integers are exact, so that cap wrongly rejected valid int64 values in (2**53, 2**63) — e.g. database/snowflake IDs and large sums from pandas reductions — while the same value as a plain Python int is accepted. Add NUMPY_INTEGRAL_RANGES (INT -> int32, LONG -> int64) for the numpy.integer path; integral floats keep the 2**53 window. Tests updated accordingly. Report URL: https://github.com/apache/texera/actions/runs/29110759837 With regards, GitHub Actions via GitBox
