Yikun commented on a change in pull request #34687: URL: https://github.com/apache/spark/pull/34687#discussion_r754789113
########## File path: python/pyspark/pandas/tests/data_type_ops/testing_utils.py ########## @@ -49,8 +49,15 @@ def numeric_pdf(self): dtypes = [np.int32, int, np.float32, float] sers = [pd.Series([1, 2, 3], dtype=dtype) for dtype in dtypes] sers.append(pd.Series([decimal.Decimal(1), decimal.Decimal(2), decimal.Decimal(3)])) + sers.append(pd.Series([1, 2, np.nan], dtype=float)) pdf = pd.concat(sers, axis=1) - pdf.columns = [dtype.__name__ for dtype in dtypes] + ["decimal"] + pdf.columns = [dtype.__name__ for dtype in dtypes] + [ + "decimal", + "float_nan", + ] + if LooseVersion(pd.__version__) >= LooseVersion("1.3.0"): Review comment: - pandas on spark with old versions of pandas fails same with/without this PR. (it's very weird, https://gist.github.com/Yikun/6b88920652fc535b336a03746fe3b04f), I added note: `# Skip decimal_nan test before v1.3.0, it not supported by pandas on spark yet.` - pandas on spark with v1.3.0+ of pandas passed with this PR. - old versions of pandas (native pandas) support decimal. This PR only enable the test case of pandas on spark with pandas v1.3.0. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org