In clipper::group_and_clip(), I reconstruct the user-input template from this list of columns previously extracted from the user-input template. This is possibly bad for a few reasons:
- It's redundant, because the user already created this template themselves
- It nullifies the two template validation checks run in
main.py
- If the user writes:
"{col1}-{col2}", this will fail early in main::main(), but it actually would work with this manual reconstruction of the template, which would write it from ["col1", "col2"] to {col1}_{col2}, e.g. This is good for overall flexibility I guess but, again, is unreachable if the user inputs a bad template.
- It may not even be functioning correctly, I haven't tested it yet
My current opinion is that if there is an easy, more Pythonic way to create these templates for use in clipper.py, then I should implement that. Otherwise, if this works it works, and redundancy is okay for now (product doesn't have to be polished yet).
In
clipper::group_and_clip(), I reconstruct the user-input template from this list of columns previously extracted from the user-input template. This is possibly bad for a few reasons:main.py"{col1}-{col2}", this will fail early inmain::main(), but it actually would work with this manual reconstruction of the template, which would write it from["col1", "col2"]to{col1}_{col2}, e.g. This is good for overall flexibility I guess but, again, is unreachable if the user inputs a bad template.My current opinion is that if there is an easy, more Pythonic way to create these templates for use in
clipper.py, then I should implement that. Otherwise, if this works it works, and redundancy is okay for now (product doesn't have to be polished yet).