Commits

Show all
Author Commit Message Labels Comments Date
Armin Rigo
Include msvcr90.dll in the win32 distribution, for now.
Branches
release-1.6.x
Armin Rigo
Bah.
Branches
release-1.6.x
Armin Rigo
Tweak. (Don't really remember why, but this was needed at some point...)
Branches
release-1.6.x
Maciej Fijalkowski
Added tag release-1.6 for changeset d8ac7d23d3ec
Branches
release-1.6.x
Laurence Tratt
Basic OpenBSD support.
Tags
release-1.6
Branches
release-1.6.x
Armin Rigo
Support for PYTHONDONTWRITEBYTECODE, setting sys.dont_write_bytecode; and when it is set, don't write .pyc files. (transplanted from 6dff8ba8be761d1313c1fc8d3d29221d35c55c85)
Branches
release-1.6.x
Armin Rigo
The constructor can take a different array; but the precise checking of the array type is still valid for the app-level extend() method. Bah. (transplanted from f277de1c5fcdf0e9fa1fd0a86b5149675a2c8524)
Branches
release-1.6.x
Armin Rigo
A bug and fix by mithrandi. No test; I just checked manually with the example shown at https://bugs.pypy.org/issue809 . (transplanted from 6c1a9945506df27e101475242c5df2ff4816f635)
Branches
release-1.6.x
Armin Rigo
Grumble. (transplanted from af690ea765e2ccd6f5143f6f4e46409b7138019a)
Branches
release-1.6.x
Armin Rigo
Can't call these ropenssl methods without releasing the GIL. They may call back RPython code that needs to acquire some locks! Baaaaah**42. Instead make sure not to call them from a __del__, with some enqueue_for_destruction(). (transplanted from ab978bd157c94b51409d330a400b854127e86692)
Branches
release-1.6.x
Armin Rigo
Fix INT/Signed. (transplanted from 6daecc356e1ab105cd2c0eb001904cf1bc6b8414)
Branches
release-1.6.x
Armin Rigo
Win32 fix. (transplanted from a91a0cb0e3e797be2bfd5969c8339013d3d1309d)
Branches
release-1.6.x
Armin Rigo
merge heads
Branches
release-1.6.x
Armin Rigo
Fixes for the previous checkin.
Branches
release-1.6.x
Armin Rigo
The GIL-releasing functions must not be reached by an RPython-level __del__.
Branches
release-1.6.x
Alex Gaynor
Fix for test_posix (how could it have ever worked!).
Branches
release-1.6.x
Armin Rigo
Fix a bug in _sqlite3.py: memory is freed too early
Branches
release-1.6.x
Armin Rigo
Bug fix. Hard to test.
Branches
release-1.6.x
Armin Rigo
Test and fix.
Branches
release-1.6.x
Armin Rigo
Another keepalive_until_here(); this one is probably needed too in theory, but I don't have a particular test case in mind.
Branches
release-1.6.x
Armin Rigo
Baaaah. Missing keepalive_until_here.
Branches
release-1.6.x
Armin Rigo
Test and (90% of a) fix.
Branches
release-1.6.x
Armin Rigo
merge heads
Branches
release-1.6.x
Armin Rigo
FrameBlocks are _immutable_, so don't change the 'previous' attribute.
Branches
release-1.6.x
Maciej Fijalkowski
remove immutable fields for now, they don't obey rules
Branches
release-1.6.x
Armin Rigo
Windows fix.
Branches
release-1.6.x
Maciej Fijalkowski
simply hardcode names for now
Branches
release-1.6.x
Maciej Fijalkowski
use -c names for now
Branches
release-1.6.x
Armin Rigo
Windows fix.
Branches
release-1.6.x
Armin Rigo
Windows bug fix. Shows up as the JIT complaining that it is missing liveness. Quite indirect :-(
Branches
release-1.6.x
  1. Prev
  2. Next