- 09 May, 2019 1 commit
-
-
Jason Madden authored
Fixes #266
-
- 08 May, 2019 3 commits
-
-
Jason Madden authored
Drop support for Python 3.4
-
Jason Madden authored
-
Jason Madden authored
isAlive -> is_alive. Fixes #263
-
- 07 May, 2019 2 commits
-
-
Jason Madden authored
Also upbate the .travis.yml to pick up modern versions of PyPy and PyPy3
-
Jason Madden authored
-
- 18 Mar, 2019 2 commits
-
-
Marius Gedminas authored
repozo -B with missing -f should show a clear error message
-
Marius Gedminas authored
Fixes #259.
-
- 06 Mar, 2019 3 commits
-
-
Nicolas Wavrant authored
Repozo : add an option to verify on recovery
-
Nicolas Wavrant authored
-
Nicolas Wavrant authored
So if recovering a ZODB fails for any reason, it doesn't leave behind a partial file which may be confused with the recovered file (as it bears same name).
-
- 29 Jan, 2019 2 commits
-
-
Tres Seaver authored
Spellcheck error in ZODB2.rst
-
sblondon authored
Voalitile -> Volatile
-
- 13 Dec, 2018 2 commits
-
-
Nicolas Wavrant authored
To give users hints about their origin if cleaning isn't done correctly
-
Nicolas Wavrant authored
-
- 23 Nov, 2018 1 commit
-
-
Nicolas Wavrant authored
But let the caller opening the file closing it
-
- 25 Oct, 2018 5 commits
-
-
Tres Seaver authored
-
Tres Seaver authored
Prep ZODB 5.5.1 release.
-
Tres Seaver authored
-
Michael Howitz authored
Fix KeyError on releasing resources of a Connection when closing the DB.
-
Michael Howitz authored
-
- 24 Oct, 2018 3 commits
-
-
Michael Howitz authored
-
Michael Howitz authored
-
Michael Howitz authored
-
- 14 Oct, 2018 2 commits
-
-
Jim Fulton authored
Use the regular transaction manager wrapped by the thread-transaction manager so we can call unregisterSynch when close is called from another thread, typically during database shutdown.
-
Jim Fulton authored
Also, use a daemonic thread so that the test runner doesn't hand waiting for the thread to die.
-
- 13 Oct, 2018 4 commits
-
-
Tres Seaver authored
-
Tres Seaver authored
Prep 5.5.0 release.
-
Tres Seaver authored
-
Tres Seaver authored
Avoid most calls to time.sleep() in the tests by mocking time.[gm]time
-
- 03 Oct, 2018 1 commit
-
-
Sune Broendum Woeller authored
-
- 21 Sep, 2018 1 commit
-
-
James Bradshaw authored
grammar (typo)
-
- 31 Aug, 2018 1 commit
-
-
Tres Seaver authored
Lifted from PR #164.
-
- 24 Aug, 2018 1 commit
-
-
Jason Madden authored
This happens on Python 3.7, evidinced by Appveyor (https://ci.appveyor.com/project/jimfulton/zodb/build/1.0.176/job/ehutnhnvbm28w5xl) but not travis because we run coverage on travis, and coverage disables refcounts of None.
-
- 23 Aug, 2018 1 commit
-
-
Jason Madden authored
This also makes us more deterministic. (Based on code I wrote for nti.testing.) On my machine, this speeds up a sequential test run from 4:08 to 1:30. We no longer get much benefit from running the layer tests in parallel, in fact it looks like the overhead actually slows them down slightly, so stop doing that.
-
- 22 Aug, 2018 5 commits
-
-
Jason Madden authored
p64 and p64 are faster and create more informative exceptions.
-
Jason Madden authored
Fixes #216.
-
Jason Madden authored
Add Python 3.7 support.
-
Jason Madden authored
And fix all the deprecations it reports. Fixes #214 and fixes #215 (well there's one remaining deprecation in 'manuel') Based on #217 so that all tests pass.
-
Jason Madden authored
Remove support for 'python setup.py test'. [skip ci]
-