Release PyGMT v0.3.0 #845

willschlitzer · 2021-02-06T07:17:52Z

Release: v0.3.0
Scheduled Date: 2021/02/15
Pull request due date: 2021/02/13

Priority PRs/issues to complete prior to release

Wrap subplot (Wrap subplot using with statement #822)
Wrap grd2cpt (Wrap grd2cpt #803)
Wrap grdinfo aliases (Wrap grdinfo aliases #799)
Move plotting functions to separate modules (Move plotting functions to separate modules #808)
Add affiliation and ORCID to AUTHORS.md (Add affiliation and ORCID to AUTHORS.md #857)
Fail to install the dev version following the install instructions (Fail to install the dev version following the install instructions #856)

Before release:

Reserve a DOI on Zenodo by clicking on "New Version" (@weiji14)
Update Changelog (Changelog entry for v0.3.0 #887) (@willschlitzer)
Add a new entry in doc/_static/version_switch.js for documentation switcher
Update citation information https://github.com/GenericMappingTools/pygmt#citing-pygmt

Release:

Go to GitHub Release and make a release, this will automatically create a tag too (@willschlitzer)
Manually upload the pygmt-vX.Y.Z.zip file to https://zenodo.org/deposit, make sure you file it under the correct reserved DOI (@weiji14)

After release:

Update conda-forge pygmt-feedstock [Usually done automatically by conda-forge's bot] (pygmt v0.3.0 conda-forge/pygmt-feedstock#14)
Bump PyGMT version on https://github.com/GenericMappingTools/try-gmt (@seisman) (GenericMappingTools/try-gmt@bd8090c)
Announce the release on:
- GMT forum (https://forum.generic-mapping-tools.org/t/pygmt-v0-3-0-released)
- GMT website (News) (@michaelgrund) (Update index.rst to announce release of PyGMT 0.3.0 website#95)
- ResearchGate link
- ~~Twitter~~ (feel free to tweet the release anyone!)

Party 🎉 (don't tick before all other checkboxes are ticked!)

The text was updated successfully, but these errors were encountered:

weiji14 · 2021-02-06T09:20:47Z

Cool, cheers for taking this up @willschlitzer! This v0.3.0 seems much bigger one than the previous v0.2.1 (#665), and looking through the draft changelog at https://github.com/GenericMappingTools/pygmt/releases (viewable only to PyGMT team members), I think we might need to reduce the number of items by adding skip-changelog labels to trim out less important commits, particularly for the non-user facing 'maintenance' ones.

Once those last features are wrapped up (hopefully by Monday/Tuesday), please start curating a changelog (see previous one at #683), and give a shoutout if you need any help with things.

willschlitzer · 2021-02-09T07:52:35Z

@GenericMappingTools/python For updating the citation info, what do I put in place for "uieda_leonardo_2020_4025418"? I'm assuming this is some sort of DOI-like reference that @leouieda set up; will it be different for the new package? Also, what is the proper order to put the authors in? I know that AUTHORS.md uses alphabetical order, but this doesn't seem to do that.

weiji14 · 2021-02-09T10:05:23Z

@GenericMappingTools/python For updating the citation info, what do I put in place for "uieda_leonardo_2020_4025418"? I'm assuming this is some sort of DOI-like reference that @leouieda set up; will it be different for the new package?

I've reserved a new doi at doi.org/10.5281/zenodo.4522136, so it should be uieda_leonardo_2021_4522136 for v0.3.0.

Also, what is the proper order to put the authors in? I know that AUTHORS.md uses alphabetical order, but this doesn't seem to do that.

It's roughly by order of commits/lines of code/outreach contributions. Leo's definitely first since he started it, and Paul the guru is last, but the rest we can discuss in #726.

P.S. The 12 Feb date is looking a bit too optimistic. Maybe push it out to 16 Feb or so.

willschlitzer · 2021-02-09T16:11:55Z

@GenericMappingTools/python For updating the citation info, what do I put in place for "uieda_leonardo_2020_4025418"? I'm assuming this is some sort of DOI-like reference that @leouieda set up; will it be different for the new package?

I've reserved a new doi at doi.org/10.5281/zenodo.4522136, so it should be uieda_leonardo_2021_4522136 for v0.3.0.

Just so I understand it, when I upload the PyGMT zip (that's just downloading the entire repository, correct?) to Zenodo, I'll list the DOI as "10.5281/zenodo.4522136", correct?

Also, what is the proper order to put the authors in? I know that AUTHORS.md uses alphabetical order, but this doesn't seem to do that.

It's roughly by order of commits/lines of code/outreach contributions. Leo's definitely first since he started it, and Paul the guru is last, but the rest we can discuss in #726.
I'll discuss the author order over at #726. Regarding the ORCID, it seems like I can copy everyone from the previous release, and just add myself and @michaelgrund. Is there anything else I'm forgetting?

P.S. The 12 Feb date is looking a bit too optimistic. Maybe push it out to 16 Feb or so.

Is this delay mostly because of wrapping subplot? I'm fine with a delay, but I do want PyGMT v0.3 done before the GMT 6.2 release. My personal preference is that it is released on 15 Feb (a US holiday) at the latest, otherwise I may have some delays during the week and not get it done until the following weekend.

weiji14 · 2021-02-09T17:57:01Z

Just so I understand it, when I upload the PyGMT zip (that's just downloading the entire repository, correct?) to Zenodo, I'll list the DOI as "10.5281/zenodo.4522136", correct?

Yes, but I'm actually the only one who can do the upload to Zenodo (see #601). You can still list the DOI in the changelog though.

Is this delay mostly because of wrapping subplot? I'm fine with a delay, but I do want PyGMT v0.3 done before the GMT 6.2 release. My personal preference is that it is released on 15 Feb (a US holiday) at the latest, otherwise I may have some delays during the week and not get it done until the following weekend.

Ok, 15 Feb (UTC) is fine (it'll be 16 Feb for me in NZ :P). Just didn't want to overwork Dongdong over CNY ;)

willschlitzer · 2021-02-09T19:15:58Z

Just so I understand it, when I upload the PyGMT zip (that's just downloading the entire repository, correct?) to Zenodo, I'll list the DOI as "10.5281/zenodo.4522136", correct?

Yes, but I'm actually the only one who can do the upload to Zenodo (see #601). You can still list the DOI in the changelog though.

Sounds good! I'll send you the ORCIDs for @michaelgrund and me. Do you need anything else from me for this?

Is this delay mostly because of wrapping subplot? I'm fine with a delay, but I do want PyGMT v0.3 done before the GMT 6.2 release. My personal preference is that it is released on 15 Feb (a US holiday) at the latest, otherwise I may have some delays during the week and not get it done until the following weekend.

Ok, 15 Feb (UTC) is fine (it'll be 16 Feb for me in NZ :P). Just didn't want to overwork Dongdong over CNY ;)

Sounds good!

seisman · 2021-02-09T22:36:58Z

15 Feb sounds a good time to me.

seisman · 2021-02-11T16:52:02Z

I've reserved a new doi at doi.org/10.5281/zenodo.4522136, so it should be uieda_leonardo_2021_4522136 for v0.3.0.

Are there any rules for the bibtex name? Is it better to use pygmt_2021_4522136?

weiji14 · 2021-02-11T22:19:10Z

I've reserved a new doi at doi.org/10.5281/zenodo.4522136, so it should be uieda_leonardo_2021_4522136 for v0.3.0.

Are there any rules for the bibtex name? Is it better to use pygmt_2021_4522136?

The uieda_leonardo_* name is just copied from https://zenodo.org/record/4253459/export/hx, but it's also completely arbitrary and people might change it in their reference manager anyway I guess. Using pygmt_2021_4522136 sounds good.

michaelgrund · 2021-02-12T07:47:27Z

I've reserved a new doi at doi.org/10.5281/zenodo.4522136, so it should be uieda_leonardo_2021_4522136 for v0.3.0.

Are there any rules for the bibtex name? Is it better to use pygmt_2021_4522136?

As @weiji14 mentioned the bibtex name is completely arbitrary, in principle it's possible to select what ever you want but I agree that pygmt_2021_4522136 (and for future releases pygmt_202X_*) would be a good choice.

weiji14 · 2021-02-14T20:29:14Z

Ok, subplot wrapper is finally done (woohoo)! I realize there's a lot of really good documentation improvements in the pipeline, and I'm happy for one or two to sneak in, but let's not waste time with the changelog at #887.

We can always make a 'quick' v0.3.1 release once #631 is fixed, and throw in a couple of minor bug fixes along the way.

weiji14 · 2021-02-15T09:46:29Z

Ok, changelog merged and published at https://www.pygmt.org/dev/changes.html#release-v0-3-0-2021-02-15! @willschlitzer, once you publish the release, I'll work on the Zenodo zipfile upload, and then it'll be the post release announcements.

weiji14 · 2021-02-15T10:12:51Z

Let's edit the forum announcement together at https://hackmd.io/@pygmt/v0-3-0_announcement. For reference, see previous one at https://forum.generic-mapping-tools.org/t/pygmt-v0-2-1-released.

weiji14 · 2021-02-15T11:09:19Z

Argh, PyPI release isn't correct, we released https://pypi.org/project/pygmt/0.3.1.dev0/#history instead. Something wrong with the PR #679, I think this line:

pygmt/.github/workflows/publish-to-pypi.yml

Line 46 in f9ec676

sed --in-place "s/pep440/pep440-pre/g" setup.cfg

~~Might need to do a force push and re-do the release...~~ Edit: No, can't do a force push as the Zenodo archive is uploaded already. We'll need to manually build and make a upload a PyPI release @seisman... Will help you out in the morning. I guess we could just download the 0.3.1.dev0 wheel/tar.gz and reupload it as 0.3.0?

seisman · 2021-02-15T13:51:05Z

We'll need to manually build and make a upload a PyPI release

I just made an upload of v0.3.0 (https://pypi.org/project/pygmt/0.3.0/).

Here is the command I'm using:

python -m pip install setuptools wheel
python setup.py sdist bdist_wheel
twine upload dist/*

I guess we could just download the 0.3.1.dev0 wheel/tar.gz and reupload it as 0.3.0?

No, we can't. The version information is written in the packages, so I need to rebuild them.

seisman · 2021-02-15T13:54:13Z

I haven't deleted v0.3.1.dev0 yet (https://pypi.org/project/pygmt/0.3.1.dev0/). I will delete it after you confirm that the v0.3.0 release works well.

I tried the following command, and it install PyGMT v0.3.0 correctly. So it looks good to me.

# cleanup the cache
python -m pip cache purge
# install PyGMT
pip install pygmt

michaelgrund · 2021-02-15T17:36:56Z

Let's edit the forum announcement together at https://hackmd.io/@pygmt/v0-3-0_announcement. For reference, see previous one at https://forum.generic-mapping-tools.org/t/pygmt-v0-2-1-released.

I already added some notes, partly influenced by the previous release.

weiji14 · 2021-02-15T18:12:24Z

I just made an upload of v0.3.0 (https://pypi.org/project/pygmt/0.3.0/).

Phew, you're a lifesaver @seisman! Was panicking a bit with all the release drama just now!

I haven't deleted v0.3.1.dev0 yet (https://pypi.org/project/pygmt/0.3.1.dev0/). I will delete it after you confirm that the v0.3.0 release works well.

I tried the following command, and it install PyGMT v0.3.0 correctly. So it looks good to me.
# cleanup the cache
python -m pip cache purge
# install PyGMT
pip install pygmt

Yep, I checked the tar.gz file and installed the wheel. Seems pretty much the same as 0.3.1.dev0 but pygmt.show_versions() gives v0.3.0 (which is good!). I think we can delete the 0.3.1.dev0 from PyPI now.

seisman · 2021-02-15T18:17:24Z

I think we can delete the 0.3.1.dev0 from PyPI now.

Done!

weiji14 · 2021-02-15T18:28:10Z

Let's edit the forum announcement together at https://hackmd.io/@pygmt/v0-3-0_announcement. For reference, see previous one at https://forum.generic-mapping-tools.org/t/pygmt-v0-2-1-released.

I already added some notes, partly influenced by the previous release.

Thanks @michaelgrund! I see that @seisman and @liamtoney's added some sentences as well. I'll polish the notes up and post it to the GMT forum later. If someone wants to start a PR to the GMT website (see previous one at GenericMappingTools/website#69) that would be terrific!

michaelgrund · 2021-02-15T18:43:44Z

Let's edit the forum announcement together at https://hackmd.io/@pygmt/v0-3-0_announcement. For reference, see previous one at https://forum.generic-mapping-tools.org/t/pygmt-v0-2-1-released.

I already added some notes, partly influenced by the previous release.

Thanks @michaelgrund! I see that @seisman and @liamtoney's added some sentences as well. I'll polish the notes up and post it to the GMT forum later. If someone wants to start a PR to the GMT website (see previous one at GenericMappingTools/website#69) that would be terrific!

Added a PR with the updates for the website @weiji14 😉

weiji14 · 2021-02-15T21:03:44Z

Just tried out the try-gmt demo at https://github.com/GenericMappingTools/try-gmt. Not sure why the version is showing up as v0.0.0 (pulled from conda-forge)?

seisman · 2021-02-15T21:13:31Z

Just tried out the try-gmt demo at GenericMappingTools/try-gmt. Not sure why the version is showing up as v0.0.0 (pulled from conda-forge)?

I also noticed that. If you download and extract the tarball from conda-forge, you will see that the metadata in the package is v0.0.0.

seisman · 2021-02-15T21:19:54Z

Just tried out the try-gmt demo at GenericMappingTools/try-gmt. Not sure why the version is showing up as v0.0.0 (pulled from conda-forge)?

I also noticed that. If you download and extract the tarball from conda-forge, you will see that the metadata in the package is v0.0.0.

We are not alone (conda-forge/pint-feedstock#32) 😃

weiji14 · 2021-02-15T21:32:10Z

Ah ok, setuptools_scm again 😅. Do you want to submit a PR to https://github.com/conda-forge/pygmt-feedstock based on that? I'm just polishing up the release announcement now.

seisman · 2021-02-15T21:35:39Z

Do you want to submit a PR to conda-forge/pygmt-feedstock based on that?

OK.

weiji14 · 2021-02-15T23:19:16Z

Alright, conda-forge versioning issue was fixed in conda-forge/pygmt-feedstock#15 (and @willschlitzer, we've added you as a maintainer there).

Forum announcement has also been posted on https://forum.generic-mapping-tools.org/t/pygmt-v0-3-0-released! What's left should just be the ResearchGate announcement (~~any takers?~~ Edit: done at https://www.researchgate.net/project/PyGMT-A-Python-interface-for-the-Generic-Mapping-Tools/update/602b43dbfcc1e200015cbd62). Maybe a twitter/instagram post as well? Did we manage to set up a social media account @meghanrjones?

seisman · 2021-02-16T20:29:04Z

Ping @meghanrjones about the social media accounts.

If we still don't have one, then we can close the issue.

maxrjones · 2021-02-16T20:32:11Z

Ping @meghanrjones about the social media accounts.

If we still don't have one, then we can close the issue.

I have not yet. I don't think Leo has either. I will be sure to get one set up in time for the next release (sorry, social media is not my expertise).

liamtoney · 2021-02-16T20:49:36Z

Ping @meghanrjones about the social media accounts.
If we still don't have one, then we can close the issue.

I have not yet. I don't think Leo has either. I will be sure to get one set up in time for the next release (sorry, social media is not my expertise).

Perhaps we could query the forum for a volunteer? Some folks really like this sort of thing (I don't have time for it)

Matplotlib has a cool Insta: https://www.instagram.com/matplotart/?hl=en. We could also set up a submission system.

weiji14 · 2021-02-16T20:55:47Z

Perhaps we could query the forum for a volunteer? Some folks really like this sort of thing (I don't have time for it)

Matplotlib has a cool Insta: https://www.instagram.com/matplotart/?hl=en. We could also set up a submission system.

Cool, I think we can move the social media discussion to another thread. I'll mark this Release v0.3.0 issue as closed, but feel free anyone to find a way to spread the word (tweet or otherwise)!

willschlitzer · 2021-02-17T12:52:16Z

@GenericMappingTools/python-contributors not sure why we closed this; I haven't seen sufficient evidence that the "Party" criteria have been met!

On a serious note; great team work everyone! Glad we got this done!

willschlitzer added the maintenance Boring but important stuff for the core devs label Feb 6, 2021

weiji14 pinned this issue Feb 6, 2021

weiji14 added this to the 0.3.0 milestone Feb 6, 2021

weiji14 added this to To do in Release v0.3.x Feb 12, 2021

weiji14 moved this from To do to In progress in Release v0.3.x Feb 14, 2021

weiji14 mentioned this issue Feb 15, 2021

Ensure proper non-dev version string when publishing to PyPI #900

Merged

5 tasks

weiji14 mentioned this issue Feb 16, 2021

Update release process in template post v0.3.0 #907

Merged

5 tasks

weiji14 closed this as completed Feb 16, 2021

Release v0.3.x automation moved this from In progress to Done Feb 16, 2021

seisman unpinned this issue Feb 17, 2021

weiji14 mentioned this issue Mar 9, 2021

Release PyGMT v0.3.1 #987

Closed

23 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Release PyGMT v0.3.0 #845

Release PyGMT v0.3.0 #845

willschlitzer commented Feb 6, 2021 •

edited by weiji14

Loading

weiji14 commented Feb 6, 2021

willschlitzer commented Feb 9, 2021

weiji14 commented Feb 9, 2021 •

edited

Loading

willschlitzer commented Feb 9, 2021

weiji14 commented Feb 9, 2021 •

edited

Loading

willschlitzer commented Feb 9, 2021

seisman commented Feb 9, 2021

seisman commented Feb 11, 2021

weiji14 commented Feb 11, 2021

michaelgrund commented Feb 12, 2021

weiji14 commented Feb 14, 2021 •

edited

Loading

weiji14 commented Feb 15, 2021

weiji14 commented Feb 15, 2021

weiji14 commented Feb 15, 2021 •

edited

Loading

seisman commented Feb 15, 2021

seisman commented Feb 15, 2021

michaelgrund commented Feb 15, 2021

weiji14 commented Feb 15, 2021

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 •

edited

Loading

michaelgrund commented Feb 15, 2021

weiji14 commented Feb 15, 2021

seisman commented Feb 15, 2021

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 •

edited

Loading

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 •

edited

Loading

seisman commented Feb 16, 2021

maxrjones commented Feb 16, 2021

liamtoney commented Feb 16, 2021

weiji14 commented Feb 16, 2021 •

edited

Loading

willschlitzer commented Feb 17, 2021

Release PyGMT v0.3.0 #845

Release PyGMT v0.3.0 #845

Comments

willschlitzer commented Feb 6, 2021 • edited by weiji14 Loading

weiji14 commented Feb 6, 2021

willschlitzer commented Feb 9, 2021

weiji14 commented Feb 9, 2021 • edited Loading

willschlitzer commented Feb 9, 2021

weiji14 commented Feb 9, 2021 • edited Loading

willschlitzer commented Feb 9, 2021

seisman commented Feb 9, 2021

seisman commented Feb 11, 2021

weiji14 commented Feb 11, 2021

michaelgrund commented Feb 12, 2021

weiji14 commented Feb 14, 2021 • edited Loading

weiji14 commented Feb 15, 2021

weiji14 commented Feb 15, 2021

weiji14 commented Feb 15, 2021 • edited Loading

seisman commented Feb 15, 2021

seisman commented Feb 15, 2021

michaelgrund commented Feb 15, 2021

weiji14 commented Feb 15, 2021

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 • edited Loading

michaelgrund commented Feb 15, 2021

weiji14 commented Feb 15, 2021

seisman commented Feb 15, 2021

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 • edited Loading

seisman commented Feb 15, 2021

weiji14 commented Feb 15, 2021 • edited Loading

seisman commented Feb 16, 2021

maxrjones commented Feb 16, 2021

liamtoney commented Feb 16, 2021

weiji14 commented Feb 16, 2021 • edited Loading

willschlitzer commented Feb 17, 2021

willschlitzer commented Feb 6, 2021 •

edited by weiji14

Loading

weiji14 commented Feb 9, 2021 •

edited

Loading

weiji14 commented Feb 9, 2021 •

edited

Loading

weiji14 commented Feb 14, 2021 •

edited

Loading

weiji14 commented Feb 15, 2021 •

edited

Loading

weiji14 commented Feb 15, 2021 •

edited

Loading

weiji14 commented Feb 15, 2021 •

edited

Loading

weiji14 commented Feb 15, 2021 •

edited

Loading

weiji14 commented Feb 16, 2021 •

edited

Loading