Re: [Biopython] Biopython 1.76 plans
Peter Cock <[email protected]> Fri, 20 Dec 2019 15:09:37 +0000
| Newsgroups | gmane.comp.python.bio.general |
|---|---|
| Message-ID | <CAKVJ-_7EVpCEqxuiMz0T0xiyitpQ+tSEz=dMi8iH+20c04eoHA@mail.gmail.com> |
OK, release announcement is out on the blog and twitter: https://www.open-bio.org/2019/12/20/biopython-1-76-released/ https://pypi.org/project/biopython/1.76/ Enjoy! Peter On Fri, Dec 20, 2019 at 2:35 PM Peter Cock <[email protected]> wrote: > > The tar-ball, Linux, Mac and Windows binaries are on PyPY. > > Note that since PyPI now blocks having both tar-ball and zip file, > would there be any objection to dropping the zip file in future? > Currently we build it and put both the tar-ball and zip are on > biopython.org > > Next I will turn off TravisCI and AppVeyor testing under Python 2.7 > and Python 3.5 (an immediate benefit here is that should about > halve the time AppVeyor takes as it runs sequentially). > > Peter > > On Fri, Dec 20, 2019 at 1:11 PM Peter Cock <[email protected]> wrote: > > > > The release is in progress, Tutorial and API documentation live: > > > > http://biopython.org/DIST/docs/tutorial/Tutorial.html > > http://biopython.org/DIST/docs/tutorial/Tutorial.pdf > > https://biopython.org/docs/1.76/api/ > > > > I recall no objections to the idea, so I am not intending to > > update the epydoc API docs, > > > > https://biopython.org/DIST/docs/api/ > > > > A small post-release project will be to replace those pages > > with redirection stubs to the Sphinx generated output. > > > > Peter > > > > On Wed, Dec 18, 2019 at 8:06 PM Peter Cock <[email protected]> wrote: > > > > > > A brief discussion with Chris about next steps in 2020 once this release is done. > > > > > > Peter > > > > > > ---------- Forwarded message --------- > > > From: Chris Rands <[email protected]> > > > Date: Wed, 18 Dec 2019 at 18:03 > > > Subject: Re: [Biopython] Biopython 1.76 plans > > > To: Peter Cock <[email protected]> > > > > > > > > > Thanks Peter, I like your plan, and please do fwd to the mailing list if you think it's of broad interest > > > > > > Chris > > > > > > ________________________________ > > > From: Peter Cock <[email protected]> > > > Sent: 18 December 2019 6:09 PM > > > To: Chris Rands <[email protected]> > > > Subject: Re: [Biopython] Biopython 1.76 plans > > > > > > Shall we stay on the list? My plan: > > > > > > (1) Release Biopython 1.76 with Python 2.7 and 3.5 support > > > (2) Turn off testing under Python 2.7 and 3.5 > > > > > > Then tackle the low handing fruit (in no particular order): > > > > > > (3) Remove Bio._py3k compatibility shim > > > (4) Look for and remove other Python 2 specific workarounds > > > (5) Remove no longer needed __future__ imports etc > > > (6) Remove all the Python 2/3 overhead in the C code > > > > > > Also, where sensible and after appropriate discussion, take > > > advantage of new language or standard library features in > > > Python 3.6+ as appropriate (e.g. sorted dictionaries by default > > > without needing to import OrderedDict from collections). > > > > > > I hadn't thought much about f-strings (beyond that I'm not > > > used to them and that makes me dislike them), but that could > > > be a good example. > > > > > > Peter > > > > > > On Wed, Dec 18, 2019 at 4:52 PM Chris Rands <[email protected]> wrote: > > > > > > > > Dear Peter, > > > > > > > > Apologies if this has already been discussed, but is the plan to merely passively drop support for Python 2/Python 3.5 or to also actively update existing code? > > > > > > > > For example, in 2020, would you merge a PR converting all string formatting to f-strings (supported since Python 3.6) or removing all `from __future__ import print_function` lines? > > > > > > > > Best, Chris > > > > > > > > > > > > ________________________________ > > > > From: Biopython <[email protected]> on behalf of Peter Cock <[email protected]> > > > > Sent: 18 December 2019 5:16 PM > > > > To: Biopython Mailing List <[email protected]> > > > > Subject: [Biopython] Biopython 1.76 plans > > > > > > > > Dear Biopythoneers, > > > > > > > > As has been discussed earlier, we hope to release Biopython 1.76 this > > > > month (December 2019). This will be a significant milestone as our > > > > final release to support Python 2 which is now officially at end of > > > > life: > > > > > > > > https://mailman.open-bio.org/pipermail/biopython/2019-November/016735.html > > > > > > > > https://python3statement.org/ > > > > > > > > Additionally, I am proposing this also be the final release to support > > > > Python 3.5: > > > > > > > > https://mailman.open-bio.org/pipermail/biopython/2019-December/016737.html > > > > > > > > We had a couple of issues reported with Biopython 1.75 post release, > > > > mainly from the Debian packaging team covering C code doctests and C > > > > code on alternative CPUs - which are now being tested via TravisCI. > > > > > > > > Otherwise I am not aware of any reported issues which would be a > > > > release blocker. > > > > > > > > If no one has any objections, I propose to do the release this Friday, > > > > 20 December 2019. > > > > > > > > Any thoughts? > > > > > > > > Thank you, > > > > > > > > Peter > > > > _______________________________________________ > > > > Biopython mailing list - [email protected] > > > > https://mailman.open-bio.org/mailman/listinfo/biopython _______________________________________________ Biopython mailing list - [email protected] https://mailman.open-bio.org/mailman/listinfo/biopython