Commit Graph

2784 Commits

Author SHA1 Message Date
Gregory P. Smith d868be8805 Adds a sanity check to avoid a *very rare* infinite loop due to a corrupt tls
key list data structure in the thread startup path.

This change is a companion to r60148 which already successfully dealt with a
similar issue on thread shutdown.

In particular this loop has been observed happening from this call path:
 #0  in find_key ()
 #1  in PyThread_set_key_value ()
 #2  in _PyGILState_NoteThreadState ()
 #3  in PyThreadState_New ()
 #4  in t_bootstrap ()
 #5  in pthread_start_thread ()

I don't know how this happens but it does, *very* rarely.  On more than
one hardware platform.  I have not been able to reproduce it manually.
(A flaky mutex implementation on the system in question is one hypothesis).

As with r60148, the spinning we managed to observe in the wild was due to a
single list element pointing back upon itself.
2008-08-04 07:33:37 +00:00
Amaury Forgeot d'Arc 39fd672dfe #3479: unichr(2**32) used to return u'\x00'.
The argument was fetched in a long, but PyUnicode_FromOrdinal takes an int.

(why doesn't gcc issue a truncation warning in this case?)
2008-07-31 21:28:03 +00:00
Amaury Forgeot d'Arc 246daedd11 #2542: now that issubclass() may call arbitrary code,
make sure that PyErr_ExceptionMatches returns 0 when an exception occurs there.
2008-07-31 00:42:16 +00:00
Antoine Pitrou c2cc80c64e Raymond's patch for #1819: speedup function calls with named parameters
(35% faster according to pybench)
2008-07-25 22:13:52 +00:00
Georg Brandl 1ad108db05 #3378: in case of no memory, don't leak even more memory. :) 2008-07-19 10:08:55 +00:00
Eric Smith f032a00271 Fix issue 3411: default float format spec fails on negative numbers. 2008-07-19 00:24:05 +00:00
Jesse Noller 5e62ca4fea Apply patch for 874900: threading module can deadlock after fork 2008-07-16 20:03:47 +00:00
Mark Dickinson 64b7e501f4 Issue #3360: Fix incorrect parsing of "020000000000.0". 2008-07-16 09:40:03 +00:00
Nick Coghlan b028f50911 Fix issue 3221 by emitting a RuntimeWarning instead of raising SystemError when the parent module can't be found during an absolute import (likely due to non-PEP 361 aware code which sets a module level __package__ attribute) 2008-07-13 14:52:36 +00:00
Amaury Forgeot d'Arc 2252d11c08 #3342: In tracebacks, printed source lines were not indented since r62555.
#3343: Py_DisplaySourceLine should be a private function. Rename it to _Py_DisplaySourceLine.
2008-07-11 21:45:06 +00:00
Robert Schuppenies 4762902998 Added garbage collector overhead and optional default return value to
sys.getsizeof.
2008-07-10 17:13:55 +00:00
Robert Schuppenies d2cd86ddd5 Fixed Issue3122 and extended sys.getsizeof tests for built-in types. 2008-07-10 13:43:26 +00:00
Gregory P. Smith 2fe77060eb - Issue #2862: Make int and float freelist management consistent with other
freelists.  Changes their CompactFreeList apis into ClearFreeList apis and
  calls them via gc.collect().
2008-07-06 03:35:58 +00:00
Georg Brandl 3c0fd5616f Fix tabs. 2008-07-05 10:07:18 +00:00
Raymond Hettinger afae11ea88 Update comment on prediction macros. 2008-07-05 02:11:55 +00:00
Georg Brandl aa76d77e2b Add a comment about incref'ing w. 2008-07-01 20:56:03 +00:00
Amaury Forgeot d'Arc bdd941fac3 #3242: fix a crash in "print", if sys.stdout is set to a custom object,
whose write() method installs another sys.stdout.

Will backport.
2008-07-01 20:38:04 +00:00
Benjamin Peterson 175e4d9663 #3219 repeated keyword arguments aren't allowed in function calls anymore 2008-07-01 19:34:52 +00:00
Brett Cannon dea1b5653f warnings.warn_explicit() did not have the proper TypeErrors in place to prevent
bus errors or SystemError being raised. As a side effect of fixing this, a bad
DECREF that could be triggered when 'message' and 'category' were both None was
fixed.

Closes issue 3211. Thanks JP Calderone for the bug report.
2008-06-27 00:31:13 +00:00
Raymond Hettinger 9c437af4eb Revert 64424, 64438, and 64439. 2008-06-24 22:46:07 +00:00
Eric Smith 65fe47b931 Modified interface to _Py_[String|Unicode]InsertThousandsGrouping, in anticipation of fixing issue 3140. 2008-06-24 00:42:10 +00:00
Raymond Hettinger e3ae655edf Make bin() implementation parallel oct() and hex() so that int/long subclasses can override or so that other classes can support. 2008-06-20 04:18:15 +00:00
Raymond Hettinger 305480c9dc Issue 3116: fix quadratic behavior in marshal.dumps(). 2008-06-16 01:42:40 +00:00
Benjamin Peterson 114f7e5fff #1683 prevent forking from interfering in threading storage
This should prevent some test_multiprocessing failures
2008-06-13 00:09:47 +00:00
Gregory P. Smith 9d53457e59 Merge in release25-maint r60793:
Added checks for integer overflows, contributed by Google. Some are
 only available if asserts are left in the code, in cases where they
 can't be triggered from Python code.
2008-06-11 07:41:16 +00:00
Gregory P. Smith dd96db63f6 This reverts r63675 based on the discussion in this thread:
http://mail.python.org/pipermail/python-dev/2008-June/079988.html

Python 2.6 should stick with PyString_* in its codebase.  The PyBytes_* names
in the spirit of 3.0 are available via a #define only.  See the email thread.
2008-06-09 04:58:54 +00:00
Benjamin Peterson f4fcdb6b8c warn about parameter tuple unpacking 2008-06-08 23:00:00 +00:00
Benjamin Peterson d5efd20406 Warn about assigning to Py3k keywords (True and False) 2008-06-08 22:52:37 +00:00
Benjamin Peterson cbd78133fa add an ast_warn helper function to make adding those Py3k warnings easier 2008-06-08 15:45:23 +00:00
Benjamin Peterson 2fe3ef8750 change Py3k backquote warning to a SyntaxWarning and add a test 2008-06-08 02:05:33 +00:00
Ronald Oussoren 5640ce2f1e MacOS X: Enable 4-way universal builds
This patch adds a new configure argument on OSX:
        --with-universal-archs=[32-bit|64-bit|all]

When used with the --enable-universalsdk option this controls which
CPU architectures are includes in the framework. The default is 32-bit,
meaning i386 and ppc. The most useful alternative is 'all', which includes
all 4 CPU architectures supported by MacOS X (i386, ppc, x86_64 and ppc64).

This includes limited support for the Carbon bindings in 64-bit mode as well,
limited because (a) I haven't done extensive testing and (b) a large portion
of the Carbon API's aren't available in 64-bit mode anyway.

I've also duplicated a feature of Apple's build of python: setting the
environment variable 'ARCHFLAGS' controls the '-arch' flags used for building
extensions using distutils.
2008-06-05 12:58:24 +00:00
Georg Brandl 7a6de8b0f4 Some style nits. Also clarify in the docstrings what __sizeof__ does. 2008-06-01 16:42:16 +00:00
Robert Schuppenies 51df064767 Issue #2898: Added sys.getsizeof() to retrieve size of objects in bytes. 2008-06-01 16:16:17 +00:00
Martin v. Löwis 6495c8da8f Typo: encoding -> codeset. 2008-06-01 08:19:02 +00:00
Martin v. Löwis b12d8579e3 Move sys_stream and sys_isatty out of the have-langinfo block. 2008-06-01 08:06:17 +00:00
Martin v. Löwis 99815892f6 New environment variable PYTHONIOENCODING. 2008-06-01 07:20:46 +00:00
Eric Smith dc13b79a38 Refactor and clean up str.format() code (and helpers) in advance of optimizations. 2008-05-30 18:10:04 +00:00
Raymond Hettinger 65856600ed Issue 2784: fix leaks in exception exit. 2008-05-30 06:37:27 +00:00
Gregory P. Smith 933d3731de Fix issue2588: Do not execute str[size-1] = '\0' when a 0 size is
passed in.  (The assert won't prevent this in non-debug builds).
2008-05-26 22:07:28 +00:00
Gregory P. Smith 23921f00e8 Fix issue2589: there was a potential integer overflow leading to
memory corruption on esoteric platforms and incorrect behavior on
normal platforms.
2008-05-26 21:16:34 +00:00
Benjamin Peterson dacde0d6ae turn PyErr_WarnPy3k into a macro 2008-05-26 17:43:53 +00:00
Christian Heimes 593daf545b Renamed PyString to PyBytes 2008-05-26 12:51:38 +00:00
Christian Heimes 3497f94476 First step of the C API rename:
renamed Include/bytesobject.h to Include/bytearrayobject.h
renamed Include/stringobject.h to Include/bytesobject.h
added Include/stringobject.h with aliases
2008-05-26 12:29:14 +00:00
Martin v. Löwis 4dd019fde3 Patch #2488: Add sys.maxsize. 2008-05-20 08:11:19 +00:00
Benjamin Peterson c6d64ec83f revert 63425 over Guido's Febuary message about this, that I missed 2008-05-17 20:09:42 +00:00
Benjamin Peterson d7943cb71d fix spelling 2008-05-17 19:51:10 +00:00
Benjamin Peterson 79a922d6df add Py3k warnings to oct and hex. backport hex behavior (because it's not different) 2008-05-17 19:21:20 +00:00
Georg Brandl 9c5b51541c Fix a refleak in the _warnings module. 2008-05-13 21:32:03 +00:00
Benjamin Peterson b9030f4f0d #2196 hasattr now allows SystemExit and KeyboardInterrupt to propagate 2008-05-12 00:41:23 +00:00
Eric Smith cf537ff39e Addresses issue 2802: 'n' formatting for integers.
Adds 'n' as a format specifier for integers, to mirror the same
specifier which is already available for floats.  'n' is the same as
'd', but inserts the current locale-specific thousands grouping.

I added this as a stringlib function, but it's only used by str type,
not unicode.  This is because of an implementation detail in
unicode.format(), which does its own str->unicode conversion.  But the
unicode version will be needed in 3.0, and it may be needed by other
code eventually in 2.6 (maybe decimal?), so I left it as a stringlib
implementation.  As long as the unicode version isn't instantiated,
there's no overhead for this.
2008-05-11 19:52:48 +00:00