zodbtools v0.0.0.dev8

parent b338d004
Zodbtools change history
0.0.0.dev8 (2019-03-06)
- Support using absolute and relative time in tidrange. One example usage is:
``zodb analyze data.fs 2018-01-01T10:30:00Z..yesterday`` ( commit__ ).
__ https://lab.nexedi.com/nexedi/zodbtools/commit/4037002c
- Python3 support progressed ( `commit 1`__, 2__, 3__ ), but zodbtools does not
support python3 yet. The test suite was extended to run python3 ( commit__ )
and also was extended to also run on ZODB with raw extensions from on going
pull request `#183`__ ( commit__ ).
__ https://lab.nexedi.com/nexedi/zodbtools/commit/d6bde57c
__ https://lab.nexedi.com/nexedi/zodbtools/commit/f16ccfd4
__ https://lab.nexedi.com/nexedi/zodbtools/commit/b338d004
__ https://lab.nexedi.com/nexedi/zodbtools/commit/eaa3aec7
__ https://github.com/zopefoundation/ZODB/pull/183
__ https://lab.nexedi.com/nexedi/zodbtools/commit/c50bfb00
0.0.0.dev7 (2019-01-11)
include COPYING LICENSE-ZPL.txt README.rst CHANGELOG.rst tox.ini
recursive-include zodbtools/test/testdata *.fs *.index *.ok
recursive-include zodbtools/test/testdata *.fs *.index *.ok *.txt
......@@ -8,7 +8,7 @@ def readfile(path):
name = 'zodbtools',
version = '0.0.0.dev7',
version = '0.0.0.dev8',
description = 'ZODB-related utilities',
long_description = '%s\n----\n\n%s' % (
readfile('README.rst'), readfile('CHANGELOG.rst')),
  • mentioned in merge request nexedi/zodbtools!8 (merged)

    Toggle commit list
  • @jerome, thanks for this. I have only minor suggestions:

    • the rest of the changelog uses (commit) not ( commit ) (note the extra spaces in between braces and content. It would be good to use consistent style throughout the whole document;
    • I'm not a native english speaker, but one thing caught my eye: ... from on going pull ... (note the space in between on and going). Maybe that should be changed to ... from ongoing pull ... (note no space in ongoing).

    Thanks also for catching *.txt not being explicitly listed there in MANIFEST.

  • Thanks @kirr for these suggestions. You are right, "ongoing" was a typo, I fixed all this and commited the result in nexedi/zodbtools@bcaf3984

    Thanks also for catching *.txt not being explicitly listed there in MANIFEST.

    I remembered I had to do this, but I think that this kind of mistakes are anyway catched when following the wendelin release procedure. The step of unpacking the dist tarball and running tests again should have failed.

    One mistake that I did first was something wrong with the rst syntax, a missing blank line. I only noticed when uploading a first time on test pypi. To prevent thisi, I'd like to suggest an amendment to the wendelin release procedure, to include a twine check step, which warns about these mistakes. I submitted a new version at https://nexedi.erp5.net/web_page_module/9202/ . I also changed lab.nexedi.cn to lab.nexedi.com in that page, I hope that's also OK.

  • @jerome, thanks for feedback and fixes. *.txt was not caught earlier, because setuptools, by defaufault, include MANIFEST.txt or some other simialr file. With our input vector txt neeeds to be included explicitly and it is generally better that we have all files included into the manifest explicitly without relying on setuptools implicit behaviour.

    I welcome twine check and updating the release howto. I will have a closer look at your edits a bit later today.

  • @jerome, thanks again. I've reviewed your edits on the howto and they look good. To review how RST rendering looks I can also suggest the following:

    1. do rst2html CHANGELOG.rst x.html and check how x.html looks in the browser;
    2. when the release commit is pushed onto lab.nexedi.com, even in some personal place in draft state, it is possible to go to that project/tree/sha, and there go to changelog, similarly to https://lab.nexedi.com/nexedi/zodbtools/blob/master/CHANGELOG.rst . Then it also shows the rendering.

    All this can help to verify not only that we have correct rst syntax, but that the document also looks right visually.

    Thanks once again for the how-to update. Could you please mark the older 001 version as no longer shared (= Archived) and make 002 shared alive?


  • Thanks @kirr ah yes rst2html is also a good way. There's also an online version at http://rst.ninjs.org/

    Because you confirmed, I shared the version 002 ( which automatically archives old version of document with same reference, this is how document management works in ERP5 ). For the records, I had to do this with my "super user" account. My normal user is not member of wendelin project so I don't have permissions to do this - and no permission to view this document either. I feel this is not confidential at all and could even be made in "published" state, but I left it as it was before.

  • @jerome, yes, thanks for feedback. Yes, this is not confidential at all and could be imho made public. I've tried to adjust classification and Follow-ups but the document is still not yet publicly visible. Unfortunately I don't have rights to "publish alive", but probably the adjustments should be enough for regular Nexedians to see the document. Please feel free to ammend the access settings if it is not the case and if you can make it public.

  • @kirr yes, only "published" / "published alive" state are realy public. Apparently, only users with "Knowledge Manager" assignments can make documents public.

    I made it public since we both agree. This how to is good, it should not be hidden.

  • @jerome, thanks. I confirm that https://www.nexedi.com/P-WIA-Wendelin.Core.Release.Howto now works, thanks for publishing it. Should we drop P-WIA from the reference, since the document is not P-WIA specific?

  • I feel it's OK to keep it as P-WIA for now, because the document uses wendelin.core in the commands and some steps are specific to wendelin ( the way of running tests). We could make a general guideline with this one and some other "how to make release" documents we have for ERP5 and SlapOS.

  • @jerome, thanks for feedback. Ok, let's keep it as is for now.

Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment