Suggestion: Make Pandas dependency optional about clevercsv HOT 6 CLOSED

Hi @seperman, thanks for the suggestion! You're correct that the package works without pandas too. Do you have a particular use case in mind where the additional dependency on pandas is problematic? One the one hand I can imagine having a smaller set of dependencies may be useful for some, but for others having the pandas wrappers by default may be very helpful. I'd like to understand both sides well before I make such a change. Let me know! :)

from clevercsv.

Hi @GjjvdBurg
Sorry I didn't get back to you earlier. Yes I do run into type of work that requires data ingestion and the source of data is often a csv. In such cases there is no need for pandas but the CSV sniffing is always required. This is particularly true for ELT pipelines where the transformations happen after the CSV is extracted and loaded.

from clevercsv.

Hi @seperman,

I've created a PR with a potential separation of the package into a core and full version. I'd like to hear your thoughts on this change and whether it addresses your concerns.

Of course, if anyone else seeing this has an opinion on the proposed split, please do let me know.

from clevercsv.

Hi @GjjvdBurg
Thanks for the PR. It looks good. I left some minor comments.

from clevercsv.

Fixed in version 0.6.0. Thanks for the suggestion @seperman!

from clevercsv.

Thanks @GjjvdBurg !

from clevercsv.