accept numpy range

[pbronez]

@max-hoffman looks like this still fails on Python 3.9. Also, would be good to update the CI checks to include Python 3.10. That was released in October 2021... Python 3.10.4 is current.

[pbronez]

@max-hoffman looks like this still fails on Python 3.9.

Looking at the build logs, there was a problem with typed_ast. That library is obsolete as of Python 3.8 but Black uses it for backwards compatibility. The failed build used black (20.8b1), latest version is black (22.3.0). Maybe try loosening that constraint? The older version of black is pinned in doltpy's pyproject.toml

[max-hoffman]

Thank you for your interest @pbronez! Yes we can bump the black version. I have found that the downsides of a stale stable black is preferable to the codebase being subtly rewritten on unrelated PRs. In this case we'll bump.

The most recent active development of Doltpy was in 2020. We would encourage you to either use a sql-client interface, like python-mysql-connector or doltcli, which does not have dependencies. The pandas/numpy integration removed in doltcli amounts to a csv -> pd.read_csv helper function. We try to support preexisting applications with a Doltpy dependence, but we are not currently investing in expanding into new Python version support.

Feel free to ask us more questions on our Discord if you would like.

[codecov-commenter]

Codecov Report

Merging #178 (b23a86b) into main (d7c8596) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##             main     #178   +/-   ##
=======================================
  Coverage   42.88%   42.88%           
=======================================
  Files          23       23           
  Lines         977      977           
=======================================
  Hits          419      419           
  Misses        558      558           

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update d7c8596...b23a86b. Read the comment docs.

[max-hoffman]

@pbronez After more investigation, we will not be able to test 3.9 and 3.10 in CI. If you find any functional deficiencies not satisfied by Doltpy or our preferred interfaces, please let us know on Discord or file an issue and we will respond as soon as we see it. We do appreciate your interest and patience! But we are selling an application database. Shipping a database as fast as MySQL and covering every SQL interface gap is our all consuming focus for the time being.

[timsehn]

The additional problem here is that doltpy includes a ton of additional scope that we implemented before you could do almost everything through the SQL interface. We can't deprecate it because it may be in use but we only want to go so far at making it work with the latest and greatest Python.