Victor Stinner [Mon, 2 Jun 2014 19:57:10 +0000 (21:57 +0200)]
Issue #21233: Rename the C structure "PyMemAllocator" to "PyMemAllocatorEx" to
make sure that the code using it will be adapted for the new "calloc" field
(instead of crashing).
Victor Stinner [Mon, 2 Jun 2014 19:40:22 +0000 (21:40 +0200)]
Issue #21639: Fix a division by zero in tracemalloc on calloc(0, 0). The
regression was introduced recently with the introduction of the new "calloc"
functions (PyMem_RawCalloc, PyMem_Calloc, PyObject_Calloc).
Add also a unit test to check for the non-regression.
Victor Stinner [Mon, 2 Jun 2014 12:18:15 +0000 (14:18 +0200)]
(Merge 3.4) Issue #21634: Fix pystone micro-benchmark: use floor division
instead of true division to benchmark integers instead of floating point
numbers. Set pystone version to 1.2. Patch written by Lennart Regebro.
Victor Stinner [Mon, 2 Jun 2014 12:16:44 +0000 (14:16 +0200)]
Issue #21634: Fix pystone micro-benchmark: use floor division instead of true
division to benchmark integers instead of floating point numbers. Set pystone
version to 1.2. Patch written by Lennart Regebro.
Attempt to fix the "too many open files" errors on several of the
buildbots that the new test_close_fds_when_max_fd_is_lowered test causes.
It now leaves 10 more low fd's available.
Attempt to fix the "too many open files" errors on several of the
buildbots that the new test_close_fds_when_max_fd_is_lowered test causes.
It now leaves 10 more low fd's available.
Don't restrict ourselves to a "max" fd when closing fds before exec()
when we have a way to get an actual list of all open fds from the OS.
Fixes issue #21618: The subprocess module would ignore fds that were
inherited by the calling process and already higher than POSIX resource
limits would otherwise allow. On systems with a functioning /proc/self/fd
or /dev/fd interface the max is now ignored and all fds are closed.
Don't restrict ourselves to a "max" fd when closing fds before exec()
when we have a way to get an actual list of all open fds from the OS.
Fixes issue #21618: The subprocess module would ignore fds that were
inherited by the calling process and already higher than POSIX resource
limits would otherwise allow. On systems with a functioning /proc/self/fd
or /dev/fd interface the max is now ignored and all fds are closed.
Remove the obsolete MAXFD constant and Popen._close_fds() method.
They should have been removed years ago when removing the old pure
Python implementation from the file.
Along the way, dismantle importlib._bootstrap._SpecMethods as it was
no longer relevant and constructing the new function required
partially dismantling the class anyway.
Gregory P. Smith [Fri, 30 May 2014 06:42:47 +0000 (23:42 -0700)]
Fix issue #14315: The zipfile module now ignores extra fields in the central
directory that are too short to be parsed instead of letting a struct.unpack
error bubble up as this "bad data" appears in many real world zip files in the
wild and is ignored by other zip tools.
Gregory P. Smith [Fri, 30 May 2014 06:42:14 +0000 (23:42 -0700)]
Fix issue #14315: The zipfile module now ignores extra fields in the central
directory that are too short to be parsed instead of letting a struct.unpack
error bubble up as this "bad data" appears in many real world zip files in the
wild and is ignored by other zip tools.
Terry Jan Reedy [Thu, 29 May 2014 05:46:26 +0000 (01:46 -0400)]
Issue #21477: Add htests for GrepDialog, UndoDelegator, and configDialog.
Put instructions in a fixed size scrollable Text. Patch by Saimadhav Heblikar.
Serhiy Storchaka [Wed, 28 May 2014 10:00:59 +0000 (13:00 +0300)]
Issue #10203: sqlite3.Row now truly supports sequence protocol. In particulr
it supports reverse() and negative indices. Original patch by Claudiu Popa.
Serhiy Storchaka [Wed, 28 May 2014 09:58:34 +0000 (12:58 +0300)]
Issue #10203: sqlite3.Row now truly supports sequence protocol. In particulr
it supports reverse() and negative indices. Original patch by Claudiu Popa.
Vinay Sajip [Wed, 28 May 2014 07:06:24 +0000 (08:06 +0100)]
Issue #18807: If copying (no symlinks) specified for a venv, then the python interpreter aliases (python, python3) are now created by copying rather than symlinking.