I added a check of Datamodel columns in DataFrameFormatter, I know we're not supposed to do validation in a Formatter. We could do the validation in a dedicated transformer, but I don't see the gain to make it standalone by now. Or we could rename the Transformer with something like DataFramePreparer but I can't find anything smart. If you have a great idea...
DataModel may only contain mandatory columns
Fix #116
I added a check of Datamodel columns in DataFrameFormatter, I know we're not supposed to do validation in a Formatter. We could do the validation in a dedicated transformer, but I don't see the gain to make it standalone by now. Or we could rename the Transformer with something like
DataFramePreparer
but I can't find anything smart. If you have a great idea...