summaryrefslogtreecommitdiffstats
path: root/Python
Commit message (Collapse)AuthorAgeFilesLines
* imp.load_dynamic() uses PyUnicode_FSConverter() to support surrogatesVictor Stinner2010-10-151-6/+6
| | | | in the library path.
* Use locale encoding if Py_FileSystemDefaultEncoding is not setVictor Stinner2010-10-151-258/+0
| | | | | | | | * PyUnicode_EncodeFSDefault(), PyUnicode_DecodeFSDefaultAndSize() and PyUnicode_DecodeFSDefault() use the locale encoding instead of UTF-8 if Py_FileSystemDefaultEncoding is NULL * redecode_filenames() functions and _Py_code_object_list (issue #9630) are no more needed: remove them
* redecode_filename(): don't need to initialize variablesVictor Stinner2010-10-151-1/+1
|
* Mark _Py_char2wchar() input argument as constantVictor Stinner2010-10-151-1/+1
|
* Explicitly close some files (from issue #10093)Antoine Pitrou2010-10-141-0/+6
|
* _Py_wgetcwd() decodes the path using _Py_char2wchar() to support surrogatesVictor Stinner2010-10-141-3/+11
|
* Issue #9992: Remove PYTHONFSENCODING environment variable.Victor Stinner2010-10-131-16/+6
|
* ceval.c: catch recursion error on _PyUnicode_AsString(co->co_filename)Victor Stinner2010-10-131-0/+4
|
* Issue #10062: Allow building on platforms which do not have sem_timedwait.Antoine Pitrou2010-10-101-1/+2
|
* fileutils.c: document which encodings are usedVictor Stinner2010-10-071-6/+26
|
* _Py_wrealpath() requires the size of the output bufferVictor Stinner2010-10-072-3/+4
|
* _Py_stat() and _Py_fopen(): avoid PyUnicode_AsWideCharString() on WindowsVictor Stinner2010-10-071-18/+6
| | | | | On Windows, Py_UNICODE is wchar_t, so we can avoid the expensive Py_UNICODE* => wchar_t* conversion.
* Fix fileutils for WindowsVictor Stinner2010-10-071-8/+7
| | | | | | * Don't define _Py_wstat() on Windows, Windows has its own _wstat() function with a different API (the stat buffer has another type) * Include windows.h
* Ooops, fileutils.c contains twice the same codeVictor Stinner2010-10-071-380/+0
| | | | | I suppose that I reapplied my local patch creating Python/fileutils.c whereas the file already existed.
* Create fileutils.c/.hVictor Stinner2010-10-073-93/+759
| | | | | | | * _Py_fopen() and _Py_stat() come from Python/import.c * (_Py)_wrealpath() comes from Python/sysmodule.c * _Py_char2wchar(), _Py_wchar2char() and _Py_wfopen() come from Modules/main.c * (_Py)_wstat(), (_Py)_wgetcwd(), _Py_wreadlink() come from Modules/getpath.c
* _wrealpath() and _Py_wreadlink() support surrogates (PEP 383)Victor Stinner2010-10-071-3/+4
| | | | Use _Py_wchar2char() to support surrogate characters in the input path.
* PyUnicode_AsWideCharString() takes a PyObject*, not a PyUnicodeObject*Victor Stinner2010-10-071-3/+3
| | | | | All unicode functions uses PyObject* except PyUnicode_AsWideChar(). Fix the prototype for the new function PyUnicode_AsWideCharString().
* Create a subfunction for PySys_SetArgvEx()Victor Stinner2010-10-061-78/+94
| | | | Create sys_update_path() static function. Do nothing if argc==0.
* #9060 Let platforms without dup2() compile the replacement fonction without ↵Amaury Forgeot d'Arc2010-10-051-0/+1
| | | | error.
* Remove an unreferenced variable. len is no longer needed.Brian Curtin2010-09-291-1/+0
|
* Issue #9630: Redecode filenames when setting the filesystem encodingVictor Stinner2010-09-291-0/+258
| | | | | | | | | | | | | | Redecode the filenames of: - all modules: __file__ and __path__ attributes - all code objects: co_filename attribute - sys.path - sys.meta_path - sys.executable - sys.path_importer_cache (keys) Keep weak references to all code objects until initfsencoding() is called, to be able to redecode co_filename attribute of all code objects.
* Issue #9979: Use PyUnicode_AsWideCharString() in import.cVictor Stinner2010-09-291-18/+16
| | | | | Don't truncate path if it is too long anymore, and allocate fewer memory (but allocate it on the heap, not on the stack).
* Since __import__ is not designed for general use, have its docstring pointBrett Cannon2010-09-271-2/+6
| | | | | | people towards importlib.import_module(). Closes issue #7397.
* revert r85003, poorly considered; breaks testsBenjamin Peterson2010-09-251-1/+1
|
* don't count keyword arguments as positional #9943Benjamin Peterson2010-09-251-1/+1
|
* add column offset to all syntax errorsBenjamin Peterson2010-09-204-33/+56
|
* add PyErr_SyntaxLocationEx, to support adding a column offsetBenjamin Peterson2010-09-201-1/+17
|
* Issue #9901: Destroying the GIL in Py_Finalize() can fail if some otherAntoine Pitrou2010-09-201-5/+8
| | | | | threads are still running. Instead, reinitialize the GIL on a second call to Py_Initialize().
* issue 9786 Native TLS support for pthreadsKristján Valur Jónsson2010-09-203-1/+49
| | | | PyThread_create_key now has a failure mode that the applicatino can detect.
* PyImport_Import was using the old import hack of sticking a dummy value intoBrett Cannon2010-09-191-3/+13
| | | | | | | fromlist to get __import__ to return the module desired. Now it uses the proper approach of fetching the module from sys.modules. Closes issue #9252. Thanks to Alexander Belopolsky for the bug report.
* Issue #9828: Destroy the GIL in Py_Finalize(), so that it gets properlyAntoine Pitrou2010-09-133-4/+34
| | | | | re-created on a subsequent call to Py_Initialize(). The problem (a crash) wouldn't appear in 3.1 or 2.7 where the GIL's structure is more trivial.
* Fix incorrect comment regarding MAGIC and TAG in import.cNick Coghlan2010-09-111-2/+5
|
* typoBenjamin Peterson2010-09-101-2/+2
|
* use Py_REFCNTBenjamin Peterson2010-09-101-2/+2
|
* remove gil_drop_request in --without-threadsBenjamin Peterson2010-09-101-4/+13
|
* use DISPATCH() instead of continueBenjamin Peterson2010-09-101-1/+1
|
* Issue #9632: Remove sys.setfilesystemencoding() function: use PYTHONFSENCODINGVictor Stinner2010-09-102-44/+0
| | | | | | environment variable to set the filesystem encoding at Python startup. sys.setfilesystemencoding() creates inconsistencies because it is unable to reencode all filenames in all objects.
* bump magic number for DELETE_DEREFBenjamin Peterson2010-09-101-1/+2
|
* #4617: Previously it was illegal to delete a name from the localAmaury Forgeot d'Arc2010-09-103-24/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | namespace if it occurs as a free variable in a nested block. This limitation of the compiler has been lifted, and a new opcode introduced (DELETE_DEREF). This sample was valid in 2.6, but fails to compile in 3.x without this change:: >>> def f(): ... def print_error(): ... print(e) ... try: ... something ... except Exception as e: ... print_error() ... # implicit "del e" here This sample has always been invalid in Python, and now works:: >>> def outer(x): ... def inner(): ... return x ... inner() ... del x There is no need to bump the PYC magic number: the new opcode is used for code that did not compile before.
* Fix Issue #9752: MSVC compiler warning due to undefined functionDaniel Stutzbach2010-09-091-4/+7
| | | | (Patch by Jon Anglin)
* Issue #9804: ascii() now always represents unicode surrogate pairs asAntoine Pitrou2010-09-091-6/+20
| | | | | | a single `\UXXXXXXXX`, regardless of whether the character is printable or not. Also, the "backslashreplace" error handler now joins surrogate pairs into a single character on UCS-2 builds.
* PEP 3149: Try to load the extension with the SOABI before tryingMatthias Klose2010-09-081-1/+1
| | | | to load the one without the SOABI in the name.
* Issue #9797: pystate.c wrongly assumed that zero couldn't be a validAntoine Pitrou2010-09-081-8/+7
| | | | thread-local storage key.
* Issue #9225: Remove the ROT_FOUR and DUP_TOPX opcode, the latter replacedAntoine Pitrou2010-09-044-46/+17
| | | | | by the new (and simpler) DUP_TOP_TWO. Performance isn't changed, but our bytecode is a bit simplified. Patch by Demur Rumed.
* _warnings exposed two variables with the name 'default_action' andBrett Cannon2010-09-041-2/+2
| | | | | | | | | | | 'once_registry'. This is bad as the warnings module had variables named 'defaultaction' and 'onceregistry' which are what people should be looking at (technically those variables shouldn't be mucked with as they are undocumented, but we all know better than to believe that isn't happening). So the variables from _warnings have been renamed to come off as private and to avoid confusion over what variable should be used. Closes issue #9766. Thanks to Antoine Pitrou for the discovery.
* PEP 3149 is accepted.Barry Warsaw2010-09-031-6/+13
| | | | http://mail.python.org/pipermail/python-dev/2010-September/103408.html
* Issue #9549: sys.setdefaultencoding() and PyUnicode_SetDefaultEncoding()Antoine Pitrou2010-09-011-20/+0
| | | | | are now removed, since their effect was inexistent in 3.x (the default encoding is hardcoded to utf-8 and cannot be changed).
* only catch AttributeError in hasattr() #9666Benjamin Peterson2010-08-241-8/+5
|
* Issue 8403: Don't mask KeyboardInterrupt during peephole operation.Raymond Hettinger2010-08-221-4/+8
|
* Add tests for r84209 (crashes in the Ast builder)Amaury Forgeot d'Arc2010-08-191-3/+3
| | | | Also remove one tab, and move a check closer to the possible failure.
generated by cgit v0.12 at 2024-10-20 08:10:18 (GMT)