Thank you very much for the time spent on this and the feedback!
My answers are below:
If an idea gains traction it can then be discussed and honed to the point of becoming a solid proposal to put to python-dev as appropriate.
So, no PEP at the moment.
2. Not that particular - to be honest:
- processing data from third-parties (reports / API responses)
- enriching data, retrieved from database, to be passed on to UI (serialization)
- code reuse -- to the extent of reusing code at expression level (e.g. to store trimming and replacing particular symbols under a separate conversion to be reused everywhere)
- also this becomes even more important when you have a few functions to reuse some code and you need to call them for each item where there are thousands of them
3. in TODO list
4. the library populates line_cache in debug mode, so this allows to normally debug using pdb (if black is installed, it gets even better :) )
5. at the bottom of TODO list
On Sun, 2020-12-27 at 16:27 +0000, Anton Abrosimov wrote:
1. I think this is too complex for the stdlib. One tool should do one thing. What about the PEP for this project?
2. This is very particular. For those who often convert data in different ways, but do not use pandas, attrs, SQL...
3. What about `typing`?
4. OTF code generation (if I understood correctly) is an debugging hell.
5. `pylint` has "several complaints" about the code.