Commit Graph

3176 Commits

Author SHA1 Message Date
Benjamin Peterson a95e977e41 decrement offset when it points to a newline (#10186 followup) 2010-10-29 03:28:14 +00:00
Antoine Pitrou 07e20ef50b Issue #5437: A preallocated MemoryError instance should not hold traceback
data (including local variables caught in the stack trace) alive infinitely.
2010-10-28 22:56:58 +00:00
Victor Stinner 07e4f1565b sys_update_path(): update sys.path even if argc==0 2010-10-25 17:37:23 +00:00
Georg Brandl 08be72d0aa Add a new warning gategory, ResourceWarning, as discussed on python-dev. It is silent by default,
except when configured --with-pydebug.

Emit this warning from the GC shutdown procedure, rather than just printing to stderr.
2010-10-24 15:11:22 +00:00
Benjamin Peterson d4519c14ca tighten loop 2010-10-24 03:41:46 +00:00
Benjamin Peterson 503d6c5ae9 remove broken code accounting an offset the size of the line #10186 2010-10-24 02:52:05 +00:00
Benjamin Peterson 8035bc5c04 follow up to #9778: define and use an unsigned hash type 2010-10-23 16:20:50 +00:00
Antoine Pitrou 9583cac633 Issue #10089: Add support for arbitrary -X options on the command-line.
They can be retrieved through a new attribute `sys._xoptions`.
2010-10-21 13:42:28 +00:00
Benjamin Peterson b8ffb60ec6 fix uninitialized struct member #10152 2010-10-20 21:25:23 +00:00
Victor Stinner e474309bb7 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.
2010-10-19 00:05:51 +00:00
Benjamin Peterson 8f67d0893f make hashes always the size of pointers; introduce Py_hash_t #9778 2010-10-17 20:54:53 +00:00
Victor Stinner c049982ea5 compiler_error(): use PyUnicode_DecodeFSDefault() to decode the filename,
instead of utf-8 in strict mode.
2010-10-17 19:16:33 +00:00
Victor Stinner 15a71cdad2 PyErr_SyntaxLocationEx() uses PyUnicode_DecodeFSDefault(), instead of
PyUnicode_FromString(), to decode the filename.
2010-10-17 19:03:16 +00:00
Victor Stinner ed7916dd00 find_module(): use FS encoding to display the missing __init__ warning 2010-10-17 02:07:09 +00:00
Victor Stinner 49d3f2514b _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
2010-10-17 01:24:53 +00:00
Victor Stinner 168e117e0a 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.
2010-10-16 23:16:16 +00:00
Victor Stinner 0a1b8cba90 _Py_wrealpath() uses _Py_char2wchar() to decode the result, to support
surrogate characters.
2010-10-16 22:55:47 +00:00
Victor Stinner 350147b5ca _Py_wreadlink(): catch _Py_char2wchar() failure 2010-10-16 22:52:09 +00:00
Victor Stinner 3f711f4a3e _Py_wreadlink() uses _Py_char2wchar() to decode the result, to support
surrogate characters.
2010-10-16 22:47:37 +00:00
Victor Stinner 2f2ed1f36c Fix ast_error_finish() and err_input(): filename can be NULL
Fix my previous commit (r85569).
2010-10-16 13:42:53 +00:00
Victor Stinner 4c7c8c3023 Issue #9713, #10114: Parser functions (eg. PyParser_ASTFromFile) expects
filenames encoded to the filesystem encoding with surrogateescape error handler
(to support undecodable bytes), instead of UTF-8 in strict mode.
2010-10-16 13:14:10 +00:00
Benjamin Peterson 230b20684f don't identify the toplevel namespace by name #9997 2010-10-16 03:45:45 +00:00
Benjamin Peterson 294a9fcba6 fix refleak 2010-10-16 03:12:39 +00:00
Barry Warsaw 8cf4eae522 First (uncontroversial) part of issue 9807.
* Expose the build flags to Python as sys.abiflags
* Shared library libpythonX.Y<abiflags>.so
* python-config --abiflags
* Make two distutils tests that failed with --enable-shared (even before this
  patch) succeed.
* Fix a few small style issues.
2010-10-16 01:04:07 +00:00
Victor Stinner 88bd891e6c Fix imp_cache_from_source(): Decode make_compiled_pathname() result from the
filesystem encoding instead of utf-8.

imp_cache_from_source() encodes the input path to filesystem encoding and this
path is passed to make_compiled_pathname().
2010-10-15 22:46:07 +00:00
Victor Stinner 1a5630326f imp_load_module() uses PyUnicode_FSConverter() to support surrogates in module
path
2010-10-15 22:43:10 +00:00
Victor Stinner 3ea23ddabf imp.cache_from_source() uses PyUnicode_FSConverter() to support surrogates in
module path
2010-10-15 20:34:32 +00:00
Victor Stinner 8dbf629bbd imp.load_dynamic() uses PyUnicode_FSConverter() to support surrogates
in the library path.
2010-10-15 12:48:01 +00:00
Victor Stinner f3170ccef8 Use locale encoding if Py_FileSystemDefaultEncoding is not set
* 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
2010-10-15 12:04:23 +00:00
Victor Stinner 6a4aff10f0 redecode_filename(): don't need to initialize variables 2010-10-15 11:16:59 +00:00
Victor Stinner 5d1e3438cd Mark _Py_char2wchar() input argument as constant 2010-10-15 11:15:54 +00:00
Antoine Pitrou b86680e299 Explicitly close some files (from issue #10093) 2010-10-14 21:15:17 +00:00
Victor Stinner f4061dac60 _Py_wgetcwd() decodes the path using _Py_char2wchar() to support surrogates 2010-10-14 12:37:19 +00:00
Victor Stinner 8f6b6b0cc3 Issue #9992: Remove PYTHONFSENCODING environment variable. 2010-10-13 22:02:27 +00:00
Victor Stinner a00064576a ceval.c: catch recursion error on _PyUnicode_AsString(co->co_filename) 2010-10-13 10:48:55 +00:00
Antoine Pitrou 19f8edc39d Issue #10062: Allow building on platforms which do not have sem_timedwait. 2010-10-10 08:37:22 +00:00
Victor Stinner 6672d0c5bb fileutils.c: document which encodings are used 2010-10-07 22:53:43 +00:00
Victor Stinner 015f4d87ab _Py_wrealpath() requires the size of the output buffer 2010-10-07 22:29:53 +00:00
Victor Stinner a4a759515e _Py_stat() and _Py_fopen(): avoid PyUnicode_AsWideCharString() on Windows
On Windows, Py_UNICODE is wchar_t, so we can avoid the expensive Py_UNICODE*
=> wchar_t* conversion.
2010-10-07 22:23:10 +00:00
Victor Stinner b306d7594f Fix fileutils for Windows
* 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
2010-10-07 22:09:40 +00:00
Victor Stinner e7c8083bf1 Ooops, fileutils.c contains twice the same code
I suppose that I reapplied my local patch creating Python/fileutils.c whereas
the file already existed.
2010-10-07 21:55:44 +00:00
Victor Stinner 4e31443c4d Create fileutils.c/.h
* _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
2010-10-07 21:45:39 +00:00
Victor Stinner 7ae7c87b05 _wrealpath() and _Py_wreadlink() support surrogates (PEP 383)
Use _Py_wchar2char() to support surrogate characters in the input path.
2010-10-07 11:06:49 +00:00
Victor Stinner beb4135b8c PyUnicode_AsWideCharString() takes a PyObject*, not a PyUnicodeObject*
All unicode functions uses PyObject* except PyUnicode_AsWideChar(). Fix the
prototype for the new function PyUnicode_AsWideCharString().
2010-10-07 01:02:42 +00:00
Victor Stinner c08ec9fdba Create a subfunction for PySys_SetArgvEx()
Create sys_update_path() static function. Do nothing if argc==0.
2010-10-06 22:44:06 +00:00
Amaury Forgeot d'Arc 90774dd47b #9060 Let platforms without dup2() compile the replacement fonction without error. 2010-10-05 22:15:37 +00:00
Brian Curtin 748cacee46 Remove an unreferenced variable. len is no longer needed. 2010-09-29 19:09:33 +00:00
Victor Stinner c39211f51e Issue #9630: Redecode filenames when setting the filesystem encoding
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.
2010-09-29 16:35:47 +00:00
Victor Stinner 255dfdb5ce Issue #9979: Use PyUnicode_AsWideCharString() in import.c
Don't truncate path if it is too long anymore, and allocate fewer memory (but
allocate it on the heap, not on the stack).
2010-09-29 10:28:51 +00:00
Brett Cannon 5305a998d5 Since __import__ is not designed for general use, have its docstring point
people towards importlib.import_module().

Closes issue #7397.
2010-09-27 21:08:38 +00:00