mir-dataset-loaders / mirdata

Python library for working with Music Information Retrieval datasets
https://mirdata.readthedocs.io/en/stable/
BSD 3-Clause "New" or "Revised" License
351 stars 59 forks source link

#580 python39 and remove python3.6 #586

Closed guillemcortes closed 8 months ago

guillemcortes commented 1 year ago

Reference issue: https://github.com/mir-dataset-loaders/mirdata/issues/580

codecov[bot] commented 1 year ago

Codecov Report

Merging #586 (afe9f95) into master (496eb4a) will increase coverage by 0.02%. The diff coverage is 100.00%.

:exclamation: Current head afe9f95 differs from pull request most recent head 18559f9. Consider uploading reports for the commit 18559f9 to get more accurate results

@@            Coverage Diff             @@
##           master     #586      +/-   ##
==========================================
+ Coverage   96.96%   96.99%   +0.02%     
==========================================
  Files          58       58              
  Lines        6990     6987       -3     
==========================================
- Hits         6778     6777       -1     
+ Misses        212      210       -2     
guillemcortes commented 1 year ago

Should we mention somewhere that we discontinue the support to Python3.6 but that if someone is interested on using that they can get mirdata<=0.3.6 (or 0.3.7 actually) which is the last one compatible with Python3.6?

I think we can include this into the next release notes. If this is too hidden, we can maybe add a changelog.md file with the info for easy tracking?

Probably that fact that the 3.6 test is still there is because the circleci config right? Maybe we could merge and see what happens in the other open PR?

yeah, let's see if @magdalenafuentes got more info, but yeah, it is strange behavior

EDIT: I see that at soundata there's a changelog.rst > https://github.com/soundata/soundata/blob/main/docs/source/changelog.rst that points to github release notes. We could just do the same, if you agree.

guillemcortes commented 8 months ago

Closing the PR since this has ben already addressed in #596