]>
granicus.if.org Git - python/log
Georg Brandl [Tue, 19 Oct 2010 18:54:25 +0000 (18:54 +0000)]
#10092: Properly reset locale in Locale*Calendar classes. The context manager was buggy because setlocale() returns the *new* locale, not the old. Also add a test for this.
Alexander Belopolsky [Tue, 19 Oct 2010 17:43:50 +0000 (17:43 +0000)]
Fixed the docstring for calendar.isleap() function.
Thanks Boštjan Mejak for the patch.
Vinay Sajip [Tue, 19 Oct 2010 15:26:24 +0000 (15:26 +0000)]
logging: Added _logRecordClass, getLogRecordClass, setLogRecordClass to increase flexibility of LogRecord creation.
Senthil Kumaran [Tue, 19 Oct 2010 04:39:35 +0000 (04:39 +0000)]
Fix Issue10140 - Tools/scripts/pathfix.py: add option to preserve timestamps
Victor Stinner [Tue, 19 Oct 2010 00:05:51 +0000 (00:05 +0000)]
initfsencoding(): get_codeset() failure is now a fatal error
Don't fallback to utf-8 anymore to avoid mojibake. I never got any error from
his function.
Victor Stinner [Mon, 18 Oct 2010 22:45:54 +0000 (22:45 +0000)]
zipimport: remove arbitrary length limit from message formats
PyErr_Format() and PyUnicode_FromFormat() are able to allocate the right buffer
size and to catch memory allocation failures.
Victor Stinner [Mon, 18 Oct 2010 22:34:46 +0000 (22:34 +0000)]
zipimport: document archive encoding; fix indentation
Victor Stinner [Mon, 18 Oct 2010 21:44:33 +0000 (21:44 +0000)]
Revert r85699 and r85701 (zipimport): fullname is a module name, not a path
UTF-8 is just fine for module names.
Victor Stinner [Mon, 18 Oct 2010 21:21:02 +0000 (21:21 +0000)]
zipimport: fix "can't find module ..." error message
I cannot use %U: fullname is a bytes object, not an unicode object. %A format
cannot be used, it adds 'b' (bytes) prefix. So create cant_find_module()
function to decode the filename and raise the error message.
Victor Stinner [Mon, 18 Oct 2010 20:59:24 +0000 (20:59 +0000)]
PyUnicode_FromFormatV(): Fix %A format
It was not completly implemented. Add a test.
Victor Stinner [Mon, 18 Oct 2010 20:44:08 +0000 (20:44 +0000)]
zipimport: find_module(), is_package() and get_source() supports surrogates
Use PyUnicode_FSConverter to support surrogates in the full name.
Victor Stinner [Mon, 18 Oct 2010 20:40:59 +0000 (20:40 +0000)]
zipimport: pass path size to make_filename()
Don't hardcode path size in make_filename().
Barry Warsaw [Mon, 18 Oct 2010 17:09:07 +0000 (17:09 +0000)]
One more change for issue 9807, by Matthias Klose.
This adds $ABIFLAGS to the static library, since it should match the shared
library name. Also, include the abiflags in python-config --libs.
Alexander Belopolsky [Mon, 18 Oct 2010 15:11:33 +0000 (15:11 +0000)]
Reverted inadvertent change that slipped into r85695
Alexander Belopolsky [Mon, 18 Oct 2010 14:43:38 +0000 (14:43 +0000)]
Issue #10117: Tools/scripts/reindent.py now accepts source files that
use encoding other than ASCII or UTF-8. Source encoding is preserved
when reindented code is written to a file.
Georg Brandl [Mon, 18 Oct 2010 12:24:53 +0000 (12:24 +0000)]
Fix hash function type.
Victor Stinner [Mon, 18 Oct 2010 12:15:34 +0000 (12:15 +0000)]
get_code_from_data() uses the filesystem encoding to encode the module path,
instead of utf-8.
Victor Stinner [Mon, 18 Oct 2010 12:13:46 +0000 (12:13 +0000)]
zipimport: read_directory() uses cp437 or utf-8 (in strict mode), depending on
the unicode flag, to decode the filename, instead of the filesystem encoding.
Use the same choice than the zipfile module.
Hirokazu Yamamoto [Mon, 18 Oct 2010 12:13:18 +0000 (12:13 +0000)]
Issue #5117: Case normalization was needed on ntpath.relpath(). And
fixed root directory issue on posixpath.relpath(). (Ported working fixes
from ntpath)
Victor Stinner [Mon, 18 Oct 2010 12:09:02 +0000 (12:09 +0000)]
zipimport: get_module_code() returns modpath as a Unicode object
... instead of a char*. Encode the module path to the fileystem encoding
(for PyImport_ExecCodeModuleEx) instead of utf-8.
Victor Stinner [Mon, 18 Oct 2010 12:03:25 +0000 (12:03 +0000)]
zipimporter_load_module() doesn't destroy mod on error
PyImport_AddModule() returns a borrowed reference. Don't display "import ... #
loaded from Zip ..." on error.
Victor Stinner [Mon, 18 Oct 2010 11:44:21 +0000 (11:44 +0000)]
zipimport: encode the prefix to the fileystem encoding
Victor Stinner [Mon, 18 Oct 2010 11:40:40 +0000 (11:40 +0000)]
zipimport, get_module_code(): avoid useless _PyUnicode_AsString()
Victor Stinner [Mon, 18 Oct 2010 11:39:05 +0000 (11:39 +0000)]
zipimport: fix indentation
Georg Brandl [Mon, 18 Oct 2010 07:35:09 +0000 (07:35 +0000)]
Remove more unneeded casts to hashfunc.
Georg Brandl [Mon, 18 Oct 2010 07:32:48 +0000 (07:32 +0000)]
Remove unneeded casts to hashfunc.
Georg Brandl [Mon, 18 Oct 2010 07:30:06 +0000 (07:30 +0000)]
Fix type of hash function.
Georg Brandl [Mon, 18 Oct 2010 07:27:55 +0000 (07:27 +0000)]
Fix compiler warning about unused static function.
Georg Brandl [Mon, 18 Oct 2010 05:06:18 +0000 (05:06 +0000)]
Fix compiler warnings about formatting pid_t as an int, by always casting to long.
R. David Murray [Mon, 18 Oct 2010 01:14:06 +0000 (01:14 +0000)]
#678250: Make mmap flush a noop on ACCESS_READ and ACCESS_COPY.
Patch by Sébastien Sablé. This solves a test_mmap failure on AIX.
R. David Murray [Mon, 18 Oct 2010 00:15:31 +0000 (00:15 +0000)]
#4499: silence compiler warning on AIX.
Patch from ActiveState.
R. David Murray [Sun, 17 Oct 2010 23:12:16 +0000 (23:12 +0000)]
Tighten up 'byte string' wording in base64 docs.
R. David Murray [Sun, 17 Oct 2010 22:46:45 +0000 (22:46 +0000)]
#1343: Add short_empty_elements option to XMLGenerator.
Patch and tests by Neil Muller.
Skip Montanaro [Sun, 17 Oct 2010 22:22:24 +0000 (22:22 +0000)]
Note the resolution of issue 9778.
Benjamin Peterson [Sun, 17 Oct 2010 21:27:01 +0000 (21:27 +0000)]
-1 is reserved for errors
Benjamin Peterson [Sun, 17 Oct 2010 21:20:58 +0000 (21:20 +0000)]
fix prototype
Benjamin Peterson [Sun, 17 Oct 2010 21:12:18 +0000 (21:12 +0000)]
fix strict aliasing warnings
Benjamin Peterson [Sun, 17 Oct 2010 20:54:53 +0000 (20:54 +0000)]
make hashes always the size of pointers; introduce Py_hash_t #9778
Victor Stinner [Sun, 17 Oct 2010 20:17:41 +0000 (20:17 +0000)]
test_httpservers: Python CGI scripts have to be encoded to utf-8
And not the locale encoding. With this commit, the test finally pass on Windows
with a non-ascii path :-)
Victor Stinner [Sun, 17 Oct 2010 20:13:36 +0000 (20:13 +0000)]
test_httpservers: Fix skip test check, the python executable path have to be
encodable to utf-8, not to the file1 encoding.
Benjamin Peterson [Sun, 17 Oct 2010 20:13:05 +0000 (20:13 +0000)]
use helper hash unimplemented function
Martin v. Löwis [Sun, 17 Oct 2010 19:48:29 +0000 (19:48 +0000)]
Fix T_BOOL bug of issue 8845.
Victor Stinner [Sun, 17 Oct 2010 19:46:36 +0000 (19:46 +0000)]
test_httpservers: skip CGI tests if Python executable path is not encodable
Gregory P. Smith [Sun, 17 Oct 2010 19:40:59 +0000 (19:40 +0000)]
Remove the external non-python.org hyperlink and point people to
Tools/gdb/libpython.py.
Martin v. Löwis [Sun, 17 Oct 2010 19:36:54 +0000 (19:36 +0000)]
Drop OPT:Olimit, as it confuses SunPRO (and other compilers).
Victor Stinner [Sun, 17 Oct 2010 19:35:30 +0000 (19:35 +0000)]
libpython: implementation of os.fsencode() with surrogateescape error handler
Victor Stinner [Sun, 17 Oct 2010 19:34:51 +0000 (19:34 +0000)]
run_file(): encode the filename with PyUnicode_EncodeFSDefault() instead of
PyUnicode_AsUTF8String()
Victor Stinner [Sun, 17 Oct 2010 19:16:33 +0000 (19:16 +0000)]
compiler_error(): use PyUnicode_DecodeFSDefault() to decode the filename,
instead of utf-8 in strict mode.
Victor Stinner [Sun, 17 Oct 2010 19:03:16 +0000 (19:03 +0000)]
PyErr_SyntaxLocationEx() uses PyUnicode_DecodeFSDefault(), instead of
PyUnicode_FromString(), to decode the filename.
Gregory P. Smith [Sun, 17 Oct 2010 18:59:56 +0000 (18:59 +0000)]
The py_decref macro was incorrect and unnecessary.
Gregory P. Smith [Sun, 17 Oct 2010 18:38:04 +0000 (18:38 +0000)]
* Applys part of the patch from http://bugs.python.org/issue3631 to add
a py_decref macro, fixup the pyo macro and reuse it and avoid a memory
leak introduced by the pylocals macro.
* Adds a note about gdb 7 python debugging support with links for
more info on that.
Matthias Klose [Sun, 17 Oct 2010 13:22:33 +0000 (13:22 +0000)]
- Accept Oracle Berkeley DB 5.0 and 5.1 as backend for the dbm extension.
Senthil Kumaran [Sun, 17 Oct 2010 11:42:21 +0000 (11:42 +0000)]
Syntax fixes for examples in the Doc/includes
Georg Brandl [Sun, 17 Oct 2010 11:36:28 +0000 (11:36 +0000)]
#9730: fix example.
Georg Brandl [Sun, 17 Oct 2010 11:29:07 +0000 (11:29 +0000)]
Fix documentation of dis.opmap direction.
Georg Brandl [Sun, 17 Oct 2010 11:23:56 +0000 (11:23 +0000)]
Make twisted example a bit more logical.
Georg Brandl [Sun, 17 Oct 2010 11:13:37 +0000 (11:13 +0000)]
Port changes to pickle docs apparently lost in py3k.
Georg Brandl [Sun, 17 Oct 2010 11:06:14 +0000 (11:06 +0000)]
#9237: document sys.call_tracing().
Georg Brandl [Sun, 17 Oct 2010 11:03:22 +0000 (11:03 +0000)]
#5121: fix claims about default values leading to segfaults.
Georg Brandl [Sun, 17 Oct 2010 10:59:41 +0000 (10:59 +0000)]
#9204: remove mentions of removed types in the types module.
Georg Brandl [Sun, 17 Oct 2010 10:53:54 +0000 (10:53 +0000)]
Fix capitalization.
Senthil Kumaran [Sun, 17 Oct 2010 10:52:12 +0000 (10:52 +0000)]
Fix Issue10119 - test_urllibnet failure when using support.transient_internet.
Georg Brandl [Sun, 17 Oct 2010 10:51:45 +0000 (10:51 +0000)]
#5962: clarify sys.exit() vs. threads.
Georg Brandl [Sun, 17 Oct 2010 10:44:11 +0000 (10:44 +0000)]
#9138: reword introduction to classes in Python.
Georg Brandl [Sun, 17 Oct 2010 10:38:20 +0000 (10:38 +0000)]
#9117: fix syntax for class definition.
Georg Brandl [Sun, 17 Oct 2010 10:34:28 +0000 (10:34 +0000)]
Some markup and style fixes in argparse docs.
Georg Brandl [Sun, 17 Oct 2010 10:28:04 +0000 (10:28 +0000)]
#9112: document error() and exit() methods of ArgumentParser.
Georg Brandl [Sun, 17 Oct 2010 10:24:54 +0000 (10:24 +0000)]
#9105: move pickle warning to a bit more prominent location.
Georg Brandl [Sun, 17 Oct 2010 10:22:28 +0000 (10:22 +0000)]
Remove second parser module example; it referred to non-readily-available example files, and this kind of discovery is much better done with the AST nowadays anyway.
Georg Brandl [Sun, 17 Oct 2010 10:15:50 +0000 (10:15 +0000)]
Make file names consistent.
Georg Brandl [Sun, 17 Oct 2010 10:14:38 +0000 (10:14 +0000)]
#9086: correct wrong terminology about linking with pythonXY.dll.
Georg Brandl [Sun, 17 Oct 2010 10:09:06 +0000 (10:09 +0000)]
#5212: md5 weaknesses do not affect hmac, so remove the note about that.
Georg Brandl [Sun, 17 Oct 2010 10:07:29 +0000 (10:07 +0000)]
Fix copy-paste error in example.
Georg Brandl [Sun, 17 Oct 2010 10:05:13 +0000 (10:05 +0000)]
#459007: merge info from PC/getpathp.c and using/windows.rst to document the forming of sys.path under Windows.
Georg Brandl [Sun, 17 Oct 2010 09:46:11 +0000 (09:46 +0000)]
#8968: add actual name of token constants.
Georg Brandl [Sun, 17 Oct 2010 09:43:35 +0000 (09:43 +0000)]
Fix hmac docs: it takes and returns bytes, except for hexdigest().
Georg Brandl [Sun, 17 Oct 2010 09:37:54 +0000 (09:37 +0000)]
#8855: add shelve security warning.
Georg Brandl [Sun, 17 Oct 2010 09:33:24 +0000 (09:33 +0000)]
#8811: small fixes to sqlite3 docs.
Georg Brandl [Sun, 17 Oct 2010 09:23:05 +0000 (09:23 +0000)]
#8686: remove potentially confusing wording that does not add any value.
Georg Brandl [Sun, 17 Oct 2010 09:19:03 +0000 (09:19 +0000)]
#8556: use less confusing mapping key in example.
Amaury Forgeot d'Arc [Sun, 17 Oct 2010 08:34:22 +0000 (08:34 +0000)]
Fix compilation warning in _ctypes module on Window
Amaury Forgeot d'Arc [Sun, 17 Oct 2010 08:26:13 +0000 (08:26 +0000)]
On Windows, remove compilation warnings with the zlib and gzip modules.
Georg Brandl [Sun, 17 Oct 2010 06:32:59 +0000 (06:32 +0000)]
#10058: tweak wording about exception returns.
Georg Brandl [Sun, 17 Oct 2010 06:21:59 +0000 (06:21 +0000)]
Note that maxtasksperchild is new in 3.2.
Gregory P. Smith [Sun, 17 Oct 2010 03:09:12 +0000 (03:09 +0000)]
Avoid hanging the test on netbsd5.
Gregory P. Smith [Sun, 17 Oct 2010 02:57:19 +0000 (02:57 +0000)]
skip test_itimer_virtual on NetBSD to prevent the test suite from hanging.
Gregory P. Smith [Sun, 17 Oct 2010 02:14:36 +0000 (02:14 +0000)]
Fix multiprocessing Semaphore's on netbsd5. SEM_VALUE_MAX is defined
as (~0U) on NetBSD which was causing it to appear as -1 when used as
a signed int for _multprocessing.SemLock.SEM_VALUE_MAX. This works
around the problem by substituting INT_MAX on systems where it appears
negative when used as an int.
Victor Stinner [Sun, 17 Oct 2010 02:07:09 +0000 (02:07 +0000)]
find_module(): use FS encoding to display the missing __init__ warning
Victor Stinner [Sun, 17 Oct 2010 01:41:09 +0000 (01:41 +0000)]
test_cmd_line_script: format paths with ascii() instead of repr()
Fix the test if the native filesystem encoding is not utf-8 (eg. cp1250 on
Windows).
Benjamin Peterson [Sun, 17 Oct 2010 01:30:26 +0000 (01:30 +0000)]
use assertion methods
Benjamin Peterson [Sun, 17 Oct 2010 01:29:11 +0000 (01:29 +0000)]
disable the garbage collector while collecting traces, so that __del__s don't get caught
Benjamin Peterson [Sun, 17 Oct 2010 01:25:19 +0000 (01:25 +0000)]
remove rather pointless test
Victor Stinner [Sun, 17 Oct 2010 01:24:53 +0000 (01:24 +0000)]
_PyImport_FixupExtension() and _PyImport_FindExtension() uses FS encoding
* Rename _PyImport_FindExtension() to _PyImport_FindExtensionUnicode():
the filename becomes a Unicode object instead of byte string
* Rename _PyImport_FixupExtension() to _PyImport_FixupExtensionUnicode():
the filename becomes a Unicode object instead of byte string
Gregory P. Smith [Sun, 17 Oct 2010 00:43:10 +0000 (00:43 +0000)]
applying netbsd-wizs-mod.patch from issue5510 -
fixes for netbsd (and dragonflybsd?)
Gregory P. Smith [Sun, 17 Oct 2010 00:17:24 +0000 (00:17 +0000)]
fix for netbsd.
Victor Stinner [Sat, 16 Oct 2010 23:46:43 +0000 (23:46 +0000)]
test_subprocess: use surrogateescape error handler to write shell scripts
test_args_string() and test_call_string() create shell scripts including the
path to the Python executable: use surrogateescape to encode paths including
surrogate characters.
Victor Stinner [Sat, 16 Oct 2010 23:38:07 +0000 (23:38 +0000)]
Oops, fix my previous commit (r85583) on calculate_path()
path value may be changed, so keep a copy in a new variable.
Victor Stinner [Sat, 16 Oct 2010 23:34:22 +0000 (23:34 +0000)]
calculate_path(): use _Py_char2wchar() to decode the PATH environment variable,
to support surrogate characters
Victor Stinner [Sat, 16 Oct 2010 23:16:16 +0000 (23:16 +0000)]
Add an optional size argument to _Py_char2wchar()
_Py_char2wchar() callers usually need the result size in characters. Since it's
trivial to compute it in _Py_char2wchar() (O(1) whereas wcslen() is O(n)), add
an option to get it.
Victor Stinner [Sat, 16 Oct 2010 22:55:47 +0000 (22:55 +0000)]
_Py_wrealpath() uses _Py_char2wchar() to decode the result, to support
surrogate characters.