Cleans up language_level_df

grambank / rgrambank

R package to access and analyse Grambank's CLDF data

Apache License 2.0

4 stars 1 forks source link

Cleans up language_level_df #22

Closed SimonGreenhill closed 1 year ago

SimonGreenhill commented 1 year ago

refactors to match coding style
removes unneeded method (drop_question_marks). People should do this themselves.
ensures that we get a LanguageTable as input.
fixes the tests

HedvigS commented 1 year ago

Reviewing now.

removes unneeded method (drop_question_marks). People should do this themselves.

I disagree, I think this is a good place to do this. It has a large effect on "singular_least_missing_data", and I'd rather there was a way of addressing this here rather than leaving users to not realise this and then backtrack.

HedvigS commented 1 year ago

Re 5310bf4, does "refactor" means standardise whitespaces? Or is there something that went missing here?

HedvigS commented 1 year ago

I was going to add to this function by first matching Language_ID in ValueTable to ID in LanguageTable and then do the stuff with Glottocode and Language_level_ID. That way it can be used on cldf-datasets that don't happen to be blessed with having Glottocodes as Language_IDs ^^!

I'll do that once this is merged.