I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR. What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs. my 2c, holger
* holger krekel <holger@merlinux.eu> [2015-12-11 14:29:35 +0000]:
I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR.
I agree - I've even been asked on Twitter before from some pytest user about this.
What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs.
I have a more radical view: Do we really want "thanks" in the changelog at all? IMHO they're just unnecessary noise. We already express our gratitude as a PR comment and we give credit by listing people in AUTHORS, so I'd rather not have this in the changelog as well. The job of it, after all, is for someone to view *what changed*. If I want to know *who* changed something, I can easily look it up. Florian -- http://www.the-compiler.org | me@the-compiler.org (Mail/XMPP) GPG: 916E B0C8 FD55 A072 | http://the-compiler.org/pubkey.asc I love long mails! | http://email.is-not-s.ms/
On Fri, Dec 11, 2015 at 12:30 PM holger krekel <holger@merlinux.eu> wrote:
I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR. What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs.
Often when I merge a PR I do so using GitHub's web interface because I'm not at my workstation so I don't have the code checked out, but it is easy enough to also use GitHub's web interface to edit and commit the CHANGELOG file directly to master after the merge, so I'm not against changing the policy regarding the CHANGELOG. About AUTHORS, if we are changing things, IMHO we can delete that file and add a link to the README pointing to https://github.com/pytest-dev/pytest/graphs/contributors or https://github.com/pytest-dev/pytest/network/members. Cheers, Bruno.
my 2c, holger _______________________________________________ pytest-dev mailing list pytest-dev@python.org https://mail.python.org/mailman/listinfo/pytest-dev
On Fri, Dec 11, 2015 at 14:48 +0000, Bruno Oliveira wrote:
On Fri, Dec 11, 2015 at 12:30 PM holger krekel <holger@merlinux.eu> wrote:
I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR. What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs.
Often when I merge a PR I do so using GitHub's web interface because I'm not at my workstation so I don't have the code checked out, but it is easy enough to also use GitHub's web interface to edit and commit the CHANGELOG file directly to master after the merge, so I'm not against changing the policy regarding the CHANGELOG.
About AUTHORS, if we are changing things, IMHO we can delete that file and add a link to the README pointing to https://github.com/pytest-dev/pytest/graphs/contributors or https://github.com/pytest-dev/pytest/network/members.
I think the source distro should directly contain the list of authors. But we can also add links to the github from the web page. As to Florian's view of not thanking: it feels different for some authors to be mentioned by name when a release announce is sent around or in the changelog. Also i think it's interesting to know who changed something. Doing blame/annotate is a different thing for different purposes IMO. best, holger
Cheers, Bruno.
my 2c, holger _______________________________________________ pytest-dev mailing list pytest-dev@python.org https://mail.python.org/mailman/listinfo/pytest-dev
-- about me: http://holgerkrekel.net/about-me/ contracting: http://merlinux.eu
I just list the usernames after the change and if the user contributed for the first time, the fullname as well. See this for example: https://github.com/fschulze/mr.developer/blob/master/CHANGES.rst Regards, Florian Schulze On 11 Dec 2015, at 16:14, holger krekel wrote:
On Fri, Dec 11, 2015 at 14:48 +0000, Bruno Oliveira wrote:
On Fri, Dec 11, 2015 at 12:30 PM holger krekel <holger@merlinux.eu> wrote:
I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR. What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs.
Often when I merge a PR I do so using GitHub's web interface because I'm not at my workstation so I don't have the code checked out, but it is easy enough to also use GitHub's web interface to edit and commit the CHANGELOG file directly to master after the merge, so I'm not against changing the policy regarding the CHANGELOG.
About AUTHORS, if we are changing things, IMHO we can delete that file and add a link to the README pointing to https://github.com/pytest-dev/pytest/graphs/contributors or https://github.com/pytest-dev/pytest/network/members.
I think the source distro should directly contain the list of authors. But we can also add links to the github from the web page.
As to Florian's view of not thanking: it feels different for some authors to be mentioned by name when a release announce is sent around or in the changelog. Also i think it's interesting to know who changed something. Doing blame/annotate is a different thing for different purposes IMO.
best, holger
Cheers, Bruno.
my 2c, holger _______________________________________________ pytest-dev mailing list pytest-dev@python.org https://mail.python.org/mailman/listinfo/pytest-dev
-- about me: http://holgerkrekel.net/about-me/ contracting: http://merlinux.eu _______________________________________________ pytest-dev mailing list pytest-dev@python.org https://mail.python.org/mailman/listinfo/pytest-dev
Can we also change a bit the changelog to have actual links to issues/prs? I often go through the changelog and what to see what the changes actually were. Thanks, -- Ionel Cristian Mărieș, http://blog.ionelmc.ro On Fri, Dec 11, 2015 at 4:29 PM, holger krekel <holger@merlinux.eu> wrote:
I was wondering about the practise to have PR authors add themselves to AUTHORS and CHANGELOG. The latter particularly is a bit odd, i.e. to thank yourself for submitting a PR. What do you all think about having the merger do this last bit (changelog entry)? AUTHORS can also be usually obtained from the history so that contributors can concentrate on tests, code and possibly docs.
my 2c, holger _______________________________________________ pytest-dev mailing list pytest-dev@python.org https://mail.python.org/mailman/listinfo/pytest-dev
participants (5)
-
Bruno Oliveira -
Florian Bruhin -
Florian Schulze -
holger krekel -
Ionel Cristian Mărieș