Release 0.710 planning

[gvanrossum]

I plan to release mypy 0.710 coming Wed (June 19) and cut the release branch for 0.710 today (June 17). This will be the first release where we're confident that the new semantic analyzer is better than the old one. You still need to enable it with --new-semantic-analyzer -- we'll likely flip this flag in the next release (0.720) and delete the old semantic analyzer in the release after that (0.730).

Please add comments to this issue if you have any PR you'd like to be included. Ideally those PRs should be landed before the release branch is cut, to minimize cherry-picking.

[gvanrossum]

Branch created: https://github.com/python/mypy/tree/release-0.710

(UPDATE: Moved to a more recent commit, to actually include the typeshed sync at #7010.)

[ilevkivskyi]

Ideally it would be great to have python/typeshed#3070 and #7013 in the release. Assuming everyone is OK with both PRs I can make the necessary syncs tomorrow morning Dublin time (unless someone does it sooner).

[gvanrossum]

I think I may have to delay the release anyway, the pin move failed (see Slack).

[gvanrossum]

OK, #7013 (and hence python/typeshed#3070) have been cherry-picked into the release branch. I will prepare new pin moves.

[gvanrossum]

Also, #6919 was cherry-picked, since it turns out #7013 depends on it. (I had to force-push the release branch.)

[gvanrossum]

The release is out.

• PyPI: https://pypi.org/project/mypy/0.710/
• Blog: https://mypy-lang.blogspot.com/2019/06/mypy-0710-released-new-semantic-analyzer.html
• Website: http://mypy-lang.org/

I'm still hoping someone will test the Windows installer so I can check that one of my internal checklist.

[JelleZijlstra]

In our codebase python/typeshed#2878 led to a number of new false positives—perhaps we should back out that change.

[JukkaL]

@JelleZijlstra We saw a few false positives caused by that PR as well. Reverting the change sounds reasonable to me.