Use rows library to convert values

[turicas]

jarbas could use some library to automatically identify and convert data types (from strings) instead of doing it itself (like on this serialize method). The rows library does it automatically and will probably remove a lot of jarbas code if used here, so the import management commands will be easier to write and maintain.

This issue strongly related to https://github.com/datasciencebr/serenata-toolbox/issues/101

cc @cuducos

[cuducos]

I think it's a great idea to use rows to better handle data input here in Jarbas!

[cuducos]

[Just pasting From (closed) #198]

rows already deals with a lot of common issues related to reading a CSV file and serializing data. With that in mind maybe we ca reduce the LOC to maintain in this project if we use rows to deal with:

• [x] ~importing and serializing reimbursement data (basically handling this transformation)~
• [ ] importing and serializing suspicions data (basically handling this transformation)
• [ ] WIP #263 importing and serializing companies data (basically handling this transformation and paying attention to the creation of Activity data)

[Ilozuluchris]

@cuducos,I think I can do this.

[cuducos]

Sounds great @Ilozuluchris — I got started one of these days (check #242), so you can either:

• Cover the scripts I haven't covered yet (if I remember right, suspicions and company importion)
• Continue from where I left opening a PR to my branch
• Throw away my code and do it all by yourself

Third option sounds like exaggerating, but might make sense in one or another scenario. Anyway I strongly recommend you to follow the advices @turicas left as a code review to my WIP PR over there ; )

[annacruz]

This issue still need to be done?

[cuducos]

Yep… there’s two WIP but still at least one case to be covered: take a look in the discussion and drafts on #242 and #262 and count on me ; )