Commit Graph

2839 Commits

Author SHA1 Message Date
Collin Winter 8725dce2ae Issue 5176: special-case string formatting in BINARY_MODULO implementation. This shows a modest (1-3%) speed-up in templating systems, for example. 2009-02-20 19:30:41 +00:00
Mark Dickinson c97c9096ed Issue #4575: fix Py_IS_INFINITY macro to work correctly on x87 FPUs.
It now forces its argument to double before testing for infinity.
2009-02-09 14:18:43 +00:00
Mark Dickinson 9e58a37252 Silence 'arg may be used uninitialized in this function' warning from gcc. 2009-02-08 17:33:11 +00:00
Eric Smith 81fe09344c Implement issue #4285, convert sys.version_info to a named
tuple. Patch by Ross Light.
2009-02-06 00:48:26 +00:00
Raymond Hettinger b516370bcb Issue 1242657: list(obj) can swallow KeyboardInterrupt. 2009-02-02 21:50:13 +00:00
Benjamin Peterson 12554cb224 more flags which only work for function blocks 2009-01-31 23:54:38 +00:00
Benjamin Peterson da9327faf7 add explanatory comment 2009-01-31 23:43:25 +00:00
Mark Hammond 9f4f56d4e8 On Windows, use the Python 'Activation Context' when loading extensions
to avoid problems loading the CRT from a private assembly.  Via bug 4566.
2009-01-27 23:12:23 +00:00
Benjamin Peterson 78821ddf8c fix building the core with --disable-unicode
I changed some bytearray methods to use strings instead of unicode like bytes_repr
Also, bytearray.fromhex() can take strings as well as unicode
2009-01-25 17:15:10 +00:00
Benjamin Peterson e18ef194d9 allow unicode keyword arguments for the ** syntax #4978 2009-01-20 14:21:16 +00:00
Amaury Forgeot d'Arc 59ce042766 #4077: No need to append \n when calling Py_FatalError
+ fix a declaration to make it match the one in pythonrun.h
2009-01-17 20:18:59 +00:00
Martin v. Löwis e3422fae8a Issue #4893: Use NT threading on CE. 2009-01-12 08:11:24 +00:00
Martin v. Löwis 1b3bef21b0 Issue #4915: Port sysmodule to Windows CE. 2009-01-12 07:57:11 +00:00
Martin v. Löwis b3b7d859b3 Issue #4279: Fix build of parsermodule under Cygwin. 2009-01-11 09:26:54 +00:00
Kristján Valur Jónsson abb70e9c9f Issue 4906: Preserve windows error state across PyThread_get_key_value 2009-01-10 12:14:31 +00:00
Kristján Valur Jónsson 0e91938e58 Issue 4293: Make Py_AddPendingCall() thread safe 2009-01-09 20:31:26 +00:00
Kristján Valur Jónsson 3b2a6b819d Issue 3677: Fix import from UNC paths on Windows. 2009-01-09 20:10:59 +00:00
Kristján Valur Jónsson 00f2df495a Issue 3582. Improved thread support and TLS for Windows 2009-01-09 20:03:27 +00:00
Antoine Pitrou e96d4ea4e2 Issue #1180193: When importing a module from a .pyc (or .pyo) file with
an existing .py counterpart, override the co_filename attributes of all
code objects if the original filename is obsolete (which can happen if the
file has been renamed, moved, or if it is accessed through different paths).
Patch by Ziga Seilnacht and Jean-Paul Calderone.
2009-01-06 18:10:47 +00:00
Martin v. Löwis 0fa10b3cd5 Issue #4817: Remove unused function PyOS_GetLastModificationTime. 2009-01-03 17:19:26 +00:00
Benjamin Peterson 796798b29c fix compilation on non-Windows platforms 2009-01-02 20:47:27 +00:00
Martin v. Löwis 5344c99734 Issue #4075: Use OutputDebugStringW in Py_FatalError. 2009-01-02 20:32:55 +00:00
Benjamin Peterson d6baa70b96 welcome to 2009, Python! 2009-01-01 17:29:49 +00:00
Benjamin Peterson 8d5934b25d #4748 lambda generators shouldn't return values 2008-12-27 18:24:11 +00:00
Benjamin Peterson 4caef5c7e2 fix #4720: the format to PyArg_ParseTupleAndKeywords can now start with '|' 2008-12-22 20:51:15 +00:00
Antoine Pitrou d0c3515bc5 Issue #2183: Simplify and optimize bytecode for list comprehensions. 2008-12-17 00:38:28 +00:00
Amaury Forgeot d'Arc ad9b5992e3 #4559: When a context manager's __exit__() method returns an object whose
conversion to bool raises an exception, 'with' loses that exception.

Reviewed by Jeffrey Yasskin.
Already ported to 2.5, will port to 2.6 and 3.0
2008-12-10 23:22:49 +00:00
Jeffrey Yasskin 2d873bd68b Issue 4597: Fix several cases in EvalFrameEx where an exception could be
"raised" without setting x, err, or why to let the eval loop know.
2008-12-08 18:55:24 +00:00
Mark Dickinson 422ce06b9c Issue #4461: Safety check in parsenumber (ast.c) 2008-12-05 17:59:46 +00:00
Jeffrey Yasskin fd8a1ec486 Speed up Python (according to pybench and 2to3-on-itself) by 1-2% by caching
whether any thread has tracing turned on, which saves one load instruction in
the fast_next_opcode path in PyEval_EvalFrameEx().  See issue 4477.
2008-12-03 06:46:45 +00:00
Benjamin Peterson 9e6310d936 always check the return value of NEW_IDENTIFIER 2008-11-25 03:43:14 +00:00
Amaury Forgeot d'Arc dc61901dd2 #3996: On Windows, PyOS_CheckStack is supposed to protect the interpreter from
stack overflow. But doing this, it always crashes when the stack is nearly full.

Reviewed by Martin von Loewis. Will backport to 2.6.
2008-11-22 20:01:18 +00:00
Benjamin Peterson c078f929cb don't segfault when \N escapes are used and unicodedata fails to load
Fixes #4367
2008-11-21 22:27:24 +00:00
Benjamin Peterson 2c98faada6 check for assignment to __debug__ during AST generation
Also, give assignment to None a better error message
2008-11-08 18:38:54 +00:00
Benjamin Peterson dcee09d920 make sure the parser flags and passed onto the compiler
This fixes "from __future__ import unicode_literals" in an exec statment
See #4225
2008-10-31 02:16:05 +00:00
Benjamin Peterson 44a90c95ce move unprefixed error into .c file 2008-10-30 23:59:18 +00:00
Benjamin Peterson 399b1fe8df give a py3k warning when 'nonlocal' is used as a variable name 2008-10-25 02:53:28 +00:00
Benjamin Peterson aff36f1295 fix compiler warning 2008-10-19 13:59:01 +00:00
Armin Ronacher 35e01fbeaa Fixed #4067 by implementing _attributes and _fields for the AST root node. 2008-10-19 08:27:43 +00:00
Christian Heimes 32a66a0410 Fixed a couple more C99 comments and one occurence of inline. 2008-10-02 19:47:50 +00:00
Andrew MacIntyre f8db82242b any platform without HAVE_LOG1P should have DBL_EPSILON in <float.h>
Part of source_os2emx.patch in issue 3868
Reviewed by Amaury Forgeot d'Arc
2008-09-22 14:11:41 +00:00
Amaury Forgeot d'Arc 05e344954d #3743: PY_FORMAT_SIZE_T is designed for the OS "printf" functions, not for
PyString_FromFormat which has an independent implementation, and uses "%zd".

This makes a difference on win64, where printf needs "%Id" to display
64bit values. For example, queue.__repr__ was incorrect.

Reviewed by Martin von Loewis.
2008-09-10 22:04:45 +00:00
Brett Cannon 1eaf0742d8 Move test.test_support.catch_warning() to the warnings module, rename it
catch_warnings(), and clean up the API.

While expanding the test suite, a bug was found where a warning about the
'line' argument to showwarning() was not letting functions with '*args' go
without a warning.

Closes issue 3602.
Code review by Benjamin Peterson.
2008-09-02 01:25:16 +00:00
Benjamin Peterson 17f03ca962 Fix compilation when --without-threads is given #3683
Reviewer: Georg Brandl, Benjamin Peterson
2008-09-01 14:18:30 +00:00
Antoine Pitrou d4ae97bc38 #3668: When PyArg_ParseTuple correctly parses a s* format, but raises an
exception afterwards (for a subsequent parameter), the user code will
not call PyBuffer_Release() and memory will leak.

Reviewed by Amaury Forgeot d'Arc.
2008-08-29 18:39:48 +00:00
Antoine Pitrou 0668c62677 Issue #2534: speed up isinstance() and issubclass() by 50-70%, so as to
match Python 2.5 speed despite the __instancecheck__ / __subclasscheck__
mechanism. In the process, fix a bug where isinstance() and issubclass(),
when given a tuple of classes as second argument, were looking up
__instancecheck__ / __subclasscheck__ on the tuple rather than on each
type object.

Reviewed by Benjamin Peterson and Raymond Hettinger.
2008-08-26 22:42:08 +00:00
Neal Norwitz 18aa388ca0 Fix:
* crashes on memory allocation failure found with failmalloc
 * memory leaks found with valgrind
 * compiler warnings in opt mode which would lead to invalid memory reads
 * problem using wrong name in decimal module reported by pychecker

Update the valgrind suppressions file with new leaks that are small/one-time
leaks we don't care about (ie, they are too hard to fix).

TBR=barry
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall (both debug and opt modes)
  in opt mode:
  valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
    ./python -E -tt ./Lib/test/regrtest.py -uall,-bsddb,-compiler \
                        -x test_logging test_ssl test_multiprocessing
  valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
    ./python -E -tt ./Lib/test/regrtest.py test_multiprocessing
  for i in `seq 1 4000` ; do
    LD_PRELOAD=~/local/lib/libfailmalloc.so FAILMALLOC_INTERVAL=$i \
        ./python -c pass
  done

At least some of these fixes should probably be backported to 2.5.
2008-08-24 05:04:52 +00:00
Benjamin Peterson 80f0ed5bb1 allow keyword args to be passed in after *args #3473 2008-08-19 19:52:46 +00:00
Benjamin Peterson 08336e30ad follup to #3473: don't duplicate the reduce code 2008-08-18 02:01:21 +00:00
Benjamin Peterson 25f2d89f32 uhh PySTEntry->ste_unoptimized has to be exposed too 2008-08-17 02:23:43 +00:00