Repository navigation
Release PyGMT v0.5.0 #1576
Description
Release: v0.5.0
Scheduled Date: 2021/10/29
Pull request due date: 2021/10/27
Priority PRs/issues to complete prior to release
- Add missing aliases to sphdistance #1516
- Add missing aliases to pygmt.grdgradient #1515
- Add missing aliases to pygmt.xyz2grd #1506
- Add missing aliases to grd2xyz #1537
- Add function to import seafloor crustal age dataset #1471
- Increase minimum thumbnail height in style.css to fit text #1575
- Allow non-string color when input data is a matrix or a file for plot and plot3d #1526
- Wrap grdvolume #1299
- Update region and projection standard docstrings #1510
- Support Python 3.10 #1591
- Wrap project #1122
- Wrap gmtselect #1429
- Allow GMTDataArrayAccessor to work on sliced datacubes #1581
- Add CITATION.cff file for PyGMT #1592
Before release:
- Run
grep --include="*.py" -r 'remove_version="v0.5.0"' pygmtfrom the base of the repository to check if any deprecations and related tests should be removed in this version - Reserve a DOI on Zenodo by clicking on "New Version"
- Finish up 'Changelog entry for v0.5.0' Pull Request:
- Add a new entry in
doc/_static/version_switch.jsfor documentation switcher - Update
CITATION.cffand BibTeX at https://github.com/GenericMappingTools/pygmt#citing-pygmt - Add the documentation link https://github.com/GenericMappingTools/pygmt#documentation-for-other-versions
- Add compatibility information https://github.com/GenericMappingTools/pygmt#compatibility-with-gmt-and-pythonnumpy-versions
- Copy draft changelog from Release Drafter and edit it to look nice (see maintainers guide for details)
- Add a new entry in
Release:
- At the PyGMT release page on GitHub:
- Edit the draft release notes with the finalized changelog
- Set the tag version and release title to v0.5.0
- Make a release by clicking the 'Publish Release' button, this will automatically create a tag too
- Manually upload the pygmt-v0.5.0.zip and baseline-images.zip files to https://zenodo.org/deposit, ensure that it is filed under the correct reserved DOI
After release:
- Update conda-forge pygmt-feedstock [Usually done automatically by conda-forge's bot] pygmt v0.5.0 conda-forge/pygmt-feedstock#19
- Bump PyGMT version on https://github.com/GenericMappingTools/try-gmt (after conda-forge update) (Bump to PyGMT v0.5.0 try-gmt#31)
- Announce the release on:
- GMT forum (do this announcement first! draft on https://hackmd.io/@pygmt. requires moderator status) done
- [Major/Minor releases only] GMT website (News)
- ResearchGate (after forum announcement)
- Twitter (after forum announcement) https://twitter.com/gmt_dev/status/1454390118668718094
- Party 🎉 (don't tick before all other checkboxes are ticked!)
Activity
willschlitzer commentedon Oct 7, 2021
@GenericMappingTools/pygmt-maintainers Please let me know if I have forgotten any priority pull requests.
weiji14 commentedon Oct 12, 2021
Due to an unforeseen family emergency, we'll need to push this release out by a week or two and assign a new release manager. Ideally we keep this to October still, so how about somewhere between 22-29 October? Any preferences @GenericMappingTools/pygmt-maintainers?
willschlitzer commentedon Oct 13, 2021
My vote is October 27-29; I can stay the release manager if the delay is that long.
willschlitzer commentedon Oct 15, 2021
I changed the release date to October 29th; if I should make it a week earlier please let me know!
weiji14 commentedon Oct 19, 2021
Ok, so we've got a reply from pyOpenSci (xref #677), and there's an impression that the disclaimer on our front README.md is a bit too scary:
Maybe we should tone down on the wording of this disclaimer for PyGMT v0.5.0, as I suggested at pyOpenSci/software-submission#43 (comment). How do people feel about PyGMT's stability, especially considering that we have a deprecation policy in place at https://github.com/GenericMappingTools/pygmt/blob/v0.4.1/doc/maintenance.md#backwards-compatibility-and-deprecation-policy?
willschlitzer commentedon Oct 21, 2021
I think it's a good idea to reword or remove this; our deprecation policy is at odds with a statement making it seem like every release may introduce breaking changes. The fact that we have changelogs and release announcement that discuss changes is sufficient, in my opinion, to inform users if there is a change that may break some of their code.
I think PyGMT has been very stable; since I came to the project 11 months ago. I don't think there have been any user-facing changes that would break older code. If the goal is to get PyGMT to v1.0, as the pyOpenSci reply indicates should be the goal for a paper, then I think we first need to come to a decision on some potentially breaking changes (#1318, #593, and #356 come to mind).
willschlitzer commentedon Oct 28, 2021
@weiji14 Do we have a DOI for this upcoming release?
weiji14 commentedon Oct 28, 2021
Yep, just reserved it. DOI for PyGMT v0.5.0 will be
10.5281/zenodo.5607255.weiji14 commentedon Oct 29, 2021
Forum announcement being drafted at https://hackmd.io/@pygmt/v0-5-0_announcement. Feel free anyone to edit the document!
michaelgrund commentedon Oct 30, 2021
Made some tiny modifications in the text and adjusted the link to the 0.5.0 changelog. All in all the whole text describes the new features etc. quite well. Great job!
weiji14 commentedon Oct 30, 2021
Thanks @michaelgrund! Do you or @willschlitzer want to go ahead and make the forum announcement? I think you both have moderator status now, so are able to post in the Announcements section. I can handle the Twitter tweet, and maybe someone can do the ResearchGate one.
michaelgrund commentedon Oct 30, 2021
Forum announcement is done.
Will also do the ResearchGate announcement in a few minutes. Updates are also posted on ResearchGate. Also created a PR to update the website news.maxrjones commentedon Nov 1, 2021
Closing this since all the boxes are checked. Great work, everyone! 🎉