for this rule check, engine tries to merge on IDVAR/IDVARVAL (in addition to keys) as the supp dataset name sets off that this is a relationship dataset, regardless of whether or not we use the Is Relationship flag in the match dataset portion of the rule. The IDVARVAL tries to convert the blank '' idvarval to floats, resulting in the error Els reported. I added filters to check for blank idvar/idvarval so that supp datasets could still be merged without values in these columns to allow this and other checks to occur.
after making it through the dataset_preprocessor.py, the dataframe looks like:
merged.docxRule_underscores.jsonDatasets.json
The rule still needs to be drafted further to perform the second portion of its check but it is correctly connecting parent and supp now
for this rule check, engine tries to merge on IDVAR/IDVARVAL (in addition to keys) as the supp dataset name sets off that this is a relationship dataset, regardless of whether or not we use the Is Relationship flag in the match dataset portion of the rule. The IDVARVAL tries to convert the blank '' idvarval to floats, resulting in the error Els reported. I added filters to check for blank idvar/idvarval so that supp datasets could still be merged without values in these columns to allow this and other checks to occur. after making it through the dataset_preprocessor.py, the dataframe looks like: merged.docx Rule_underscores.json Datasets.json The rule still needs to be drafted further to perform the second portion of its check but it is correctly connecting parent and supp now