[Biopython] Volunteers for making Biopython release 1.77

Thu May 14 11:10:37 UTC 2020

Excellent news, thank you Chris.

Two passes of LaTex is the norm when using labels and references, three
passes if you use references to page numbers. The Makefile should do that.
Hevea includes its own loop which is a littles easier to use.

As to the outstanding issues, the BLAST DB v5 test update seems easy enough
- could you review this pull request Chris?

https://github.com/biopython/biopython/issues/2863
https://github.com/biopython/biopython/pull/2890

Updating the BLAST wrappers would be nice, but it seems we might just skip
that (and silence the test warnings about the BLAST wrappers being out of
date) if we're serious about dropping the Bio.Application framework:

https://github.com/biopython/biopython/issues/2184
https://github.com/biopython/biopython/issues/2877

Oh, and can we close the UniProt/SwissProt parser issue or not?:

https://github.com/biopython/biopython/issues/2417

Other suggestions are https://github.com/biopython/biopython/pull/2399 which
has some merge conflicts and needs a review.

Also if I can finish the documentation
https://github.com/biopython/biopython/pull/2334 would be nice to include
too.

Right now I still think only #2863/#2890 for the BLAST test, and #2417 for
UniProt/SwissProt are blockers.

Peter

On Thu, May 14, 2020 at 8:16 AM Daley, Christopher <dalchr28 at evergreen.edu>
wrote:

> I fetched the latest commits just now and ran the full tests, including
> the tar-ball and documentation and it all seems to be in order. I do a lot
> of Latex work, so pretty well set up in that area although I had to install
> Hevea from APT. The version in the Ubuntu 18.04 repos is 2.30 which is
> pretty recent and seems to output the Tutorial nicely enough although it
> took two passes of the latex compiler.
> It looks all set from this end, barring the pull requests - are there any
> blocking issues or PRs?
> Chris
> ------------------------------
> *From:* Peter Cock <p.j.a.cock at googlemail.com>
> *Sent:* Tuesday, May 12, 2020 7:52 AM
> *To:* Daley, Christopher <dalchr28 at evergreen.edu>
> *Cc:* Michiel de Hoon <mjldehoon at yahoo.com>; biopython at biopython.org <
> biopython at biopython.org>
> *Subject:* Re: [Biopython] Volunteers for making Biopython release 1.77
>
> Excellent - Markus said on GitHub that the NCBI tests did complete
> recently but sometimes took up to 2 hours. This is something we will
> probably have to reduce as that is not practical.
>
> If you haven't already, I suggest you next try building the tar-ball and
> wheels locally, installing them, and running the tests from an unzipped
> tar-ball. The reason we do this is to catch problems with the manifest
> (e.g. new test files not being included in the tar-ball).
>
> Also are you setup to build the Tutorial? Latex is easy to install (just a
> large download), but Hevea can be a pain. I did spend some effort trying to
> get that into conda, but didn't succeed.
>
> Peter
>
> On Mon, May 11, 2020 at 3:20 PM Daley, Christopher <dalchr28 at evergreen.edu>
> wrote:
>
> I've stood up a Docker container with all of the dependencies and
> successfully run the full online test suite with the exception of the
> NCBI_qblast test which seems to be timing out.
>
> I'll make a pull request to update the Docker images once I test them a
> bit more locally.
>
> Chris
>
>
> ------------------------------
> *From:* Peter Cock <p.j.a.cock at googlemail.com>
> *Sent:* Saturday, May 9, 2020 10:47 AM
> *To:* Daley, Christopher <dalchr28 at evergreen.edu>
> *Cc:* Michiel de Hoon <mjldehoon at yahoo.com>; biopython at biopython.org <
> biopython at biopython.org>
> *Subject:* Re: [Biopython] Volunteers for making Biopython release 1.77
>
> Good news I guess: Bio.PopGen.FDist was deprecated in Release 1.68,
> removed in Release 1.70. So we don't need to worry about fdist2 anymore.
>
> Perhaps you will have some pull requests to update the Docker images?
>
> Peter
>
> On Sat, May 9, 2020 at 6:40 PM Daley, Christopher <dalchr28 at evergreen.edu>
> wrote:
>
> Glad it was a simple fix.
>
> I started building a Docker container to run the full tests locally based
> on the https://github.com/biopython/biopython_docker
> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fbiopython%2Fbiopython_docker&data=02%7C01%7Cdalchr28%40evergreen.edu%7C0f2d9d6e9164401aa66e08d7f684241d%7C22adcff7c06f49a68f2050711c40ddaa%7C0%7C0%7C637248919720142489&sdata=Bxr5BeYAh7J8wj8VUy6bLwBcwz4Cne3j5nnkmGe0RwI%3D&reserved=0>
> repo, updated to Ubuntu 18.04 and Python 3.6. In doing so, I discovered
> that the download link for fdist2 is dead and there doesn't seem to be any
> other source for it online. Checking the other links it also seems like
> http://xxmotif.genzentrum.lmu.de
> <https://nam02.safelinks.protection.outlook.com/?url=http%3A%2F%2Fxxmotif.genzentrum.lmu.de%2Findex.php&data=02%7C01%7Cdalchr28%40evergreen.edu%7C0f2d9d6e9164401aa66e08d7f684241d%7C22adcff7c06f49a68f2050711c40ddaa%7C0%7C0%7C637248919720142489&sdata=Is7cTyw%2BAieycEyj5VWm8cDoZaUziCrGG8OvYGfmTKg%3D&reserved=0>
> is down as well. The Bioconda package points to this github repo:
> https://github.com/soedinglab/xxmotif
> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fsoedinglab%2Fxxmotif&data=02%7C01%7Cdalchr28%40evergreen.edu%7C0f2d9d6e9164401aa66e08d7f684241d%7C22adcff7c06f49a68f2050711c40ddaa%7C0%7C0%7C637248919720152483&sdata=%2BntvyTrH3Drw0DeZLhCksmBkRURRwcWWgqF7fz5%2BwyU%3D&reserved=0>
>
> Any thoughts on how to proceed?
> Chris
>
> ------------------------------
> *From:* Peter Cock <p.j.a.cock at googlemail.com>
> *Sent:* Friday, May 8, 2020 1:34 PM
> *To:* Daley, Christopher <dalchr28 at evergreen.edu>
> *Cc:* Michiel de Hoon <mjldehoon at yahoo.com>; biopython at biopython.org <
> biopython at biopython.org>
> *Subject:* Re: [Biopython] Volunteers for making Biopython release 1.77
>
> Good work on this documentation fix Chris - something I was hoping we'd
> fixed for the next release:
>
> https://github.com/biopython/biopython/issues/2502
> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fbiopython%2Fbiopython%2Fissues%2F2502&data=02%7C01%7Cdalchr28%40evergreen.edu%7C0f2d9d6e9164401aa66e08d7f684241d%7C22adcff7c06f49a68f2050711c40ddaa%7C0%7C0%7C637248919720162478&sdata=OwY6vorDjyNvrN3Lhm1OXe5ECBeqVfhEaiYugyREXcs%3D&reserved=0>
>
> Another important task (which on re-reading might be made even more
> explicit) is running all the tests locally, including the online tests
> (since we skip those on TravisCI and AppVeyor). This occasionally throws up
> nasty surprises:
>
> https://biopython.org/wiki/Building_a_release
> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbiopython.org%2Fwiki%2FBuilding_a_release&data=02%7C01%7Cdalchr28%40evergreen.edu%7C0f2d9d6e9164401aa66e08d7f684241d%7C22adcff7c06f49a68f2050711c40ddaa%7C0%7C0%7C637248919720162478&sdata=stgShVLrb%2BuUsjXenxp2PQeFaWFQa4OKqX67Y6CzQQI%3D&reserved=0>
>
> Peter
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.open-bio.org/pipermail/biopython/attachments/20200514/d4ede239/attachment-0001.htm>