Excerpts from Dmitry Tantsur's message of 2018-06-06 18:24:00 +0200: > In Ironic world we run doc8 on README.rst as part of the pep8 job. Maybe we > should make it a common practice?
That seems like it may be a good thing to add, but I don't know that it is sufficient to detect all of the problems that prevent uploading packages because of the README formatting. > > On 06/06/2018 03:35 PM, Jeremy Stanley wrote: > > On 2018-06-06 16:36:45 +0900 (+0900), Akihiro Motoki wrote: > > [...] > >> In addition, unfortunately such checks are not run in project gate, > >> so there is no way to detect in advance. > >> I think we need a way to check this when a change is made > >> instead of detecting an error when a release patch is proposed. > > > > While I hate to suggest yet another Python PTI addition, for my > > personal projects I test every commit (essentially a check/gate > > pipeline job) with: > > > > python setup.py check --restructuredtext --strict > > python setup.py bdist_wheel sdist > > > > ...as proof that it hasn't broken sdist/wheel building nor regressed > > the description-file provided in my setup.cfg. My intent is to add > > other release artifact tests into the same set so that there are no > > surprises come release time. > > > > We sort of address this case in OpenStack projects by forcing sdist > > builds in our standard pep8 jobs, so maybe that would be a > > lower-overhead place to introduce the setup rst check? > > Brainstorming. > > > > > > > > __________________________________________________________________________ > > OpenStack Development Mailing List (not for usage questions) > > Unsubscribe: [email protected]?subject:unsubscribe > > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
