https://hg.dopieralski.pl/cpython-lru_cache-weakref/ en-us cpython-lru_cache-weakref: Objects/unicodeobject.c history Objects/unicodeobject.c revision history Issue #19674: inspect.signature() now produces a correct signature https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/78ec18f5cb45/Objects/unicodeobject.c for some builtins.]]> Larry Hastings <larry@hastings.org> Sat, 23 Nov 2013 15:37:55 -0800 Issue #19730: Argument Clinic now supports all the existing PyArg https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/760ccd78e874/Objects/unicodeobject.c "format units" as legacy converters, as well as two new features:
"self converters" and the "version" directive.]]> Larry Hastings <larry@hastings.org> Sat, 23 Nov 2013 14:54:00 -0800 Issue #19619: Blacklist non-text codecs in method API https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/d68df99d7a57/Objects/unicodeobject.c
str.encode, bytes.decode and bytearray.decode now use an
internal API to throw LookupError for known non-text encodings,
rather than attempting the encoding or decoding operation and
then throwing a TypeError for an unexpected output type.

The latter mechanism remains in place for third party non-text
encodings.]]> Nick Coghlan <ncoghlan@gmail.com> Fri, 22 Nov 2013 22:39:36 +1000 ssue #19183: Implement PEP 456 'secure and interchangeable hash algorithm'. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/adb471b9cba1/Objects/unicodeobject.c Python now uses SipHash24 on all major platforms.]]> Christian Heimes <christian@cheimes.de> Wed, 20 Nov 2013 11:46:18 +0100 Add _PyUnicodeWriter_WriteASCIIString() function https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/d1ca05428c38/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Tue, 19 Nov 2013 12:54:53 +0100 Issue #12892: The utf-16* and utf-32* codecs now reject (lone) surrogates. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/0d9624f2ff43/Objects/unicodeobject.c
The utf-16* and utf-32* encoders no longer allow surrogate code points
(U+D800-U+DFFF) to be encoded.
The utf-32* decoders no longer decode byte sequences that correspond to
surrogate code points.
The surrogatepass error handler now works with the utf-16* and utf-32* codecs.

Based on patches by Victor Stinner and Kang-Hao (Kenny) Lu.]]> Serhiy Storchaka <storchaka@gmail.com> Tue, 19 Nov 2013 11:32:41 +0200 Issue #19581: Change the overallocation factor of _PyUnicodeWriter on Windows https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/093b9838a41c/Objects/unicodeobject.c
On Windows, a factor of 50% gives best performances.]]> Victor Stinner <victor.stinner@gmail.com> Mon, 18 Nov 2013 21:08:39 +0100 Argument Clinic: rename "self" to "module" for module-level functions. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/64c853a88a10/Objects/unicodeobject.c Larry Hastings <larry@hastings.org> Mon, 18 Nov 2013 09:32:13 -0800 #17806: Added keyword-argument support for "tabsize" to str/bytes.expandtabs(). https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/82b58807f481/Objects/unicodeobject.c Ezio Melotti <ezio.melotti@gmail.com> Sat, 16 Nov 2013 19:10:57 +0200 Close #17828: better handling of codec errors https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/854a2cea31b9/Objects/unicodeobject.c
- output type errors now redirect users to the type-neutral
convenience functions in the codecs module
- stateless errors that occur during encoding and decoding
will now be automatically wrapped in exceptions that give
the name of the codec involved]]> Nick Coghlan <ncoghlan@gmail.com> Wed, 13 Nov 2013 23:49:21 +1000 _Py_normalize_encoding(): explain how the value 6 was computed https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/9c929b9e0a2a/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Thu, 07 Nov 2013 23:12:23 +0100 Fix _Py_normalize_encoding(): ensure that buffer is big enough to store "utf-8" https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/99afa4c74436/Objects/unicodeobject.c if the input string is NULL]]> Victor Stinner <victor.stinner@gmail.com> Thu, 07 Nov 2013 13:33:36 +0100 Issue #19512: add _PyUnicode_CompareWithId() function https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/77bebcf5c4cf/Objects/unicodeobject.c
_PyUnicode_CompareWithId() is faster than PyUnicode_CompareWithASCIIString()
when both strings are equal and interned.

Add also _PyId_builtins identifier for "builtins" common string.]]> Victor Stinner <victor.stinner@gmail.com> Thu, 07 Nov 2013 00:46:04 +0100 Issue #19424: PyUnicode_CompareWithASCIIString() normalizes memcmp() result https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/494f736f5945/Objects/unicodeobject.c to -1, 0, 1]]> Victor Stinner <victor.stinner@gmail.com> Mon, 04 Nov 2013 11:28:26 +0100 Issue #16286: remove duplicated identity check from unicode_compare() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/da9c6e4ef301/Objects/unicodeobject.c
Move the test to PyUnicode_Compare()]]> Victor Stinner <victor.stinner@gmail.com> Mon, 04 Nov 2013 11:27:14 +0100 Issue #16286: optimize PyUnicode_RichCompare() for identical strings (same https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/5fa291435740/Objects/unicodeobject.c pointer) for any operator, not only Py_EQ and Py_NE.

Code of bytes_richcompare() and PyUnicode_RichCompare() is now closer.]]> Victor Stinner <victor.stinner@gmail.com> Mon, 04 Nov 2013 11:23:05 +0100 Issue #16286: write a new subfunction bytes_compare_eq() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/536a7c09c7fd/Objects/unicodeobject.c
* cleanup bytes_richcompare()
* PyUnicode_RichCompare(): replace a test with a XOR]]> Victor Stinner <victor.stinner@gmail.com> Mon, 04 Nov 2013 11:08:10 +0100 Issue #19424: Fix a compiler warning on comparing signed/unsigned size_t https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/2ed8d500e113/Objects/unicodeobject.c
Patch written by Zachary Ware.]]> Victor Stinner <victor.stinner@gmail.com> Sun, 03 Nov 2013 13:53:12 +0100 Issue #19424: Fix a compiler warning https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/52ec6a3eeda5/Objects/unicodeobject.c
memcmp() just takes raw pointers]]> Victor Stinner <victor.stinner@gmail.com> Wed, 30 Oct 2013 18:27:13 +0100 Issue #19424: Optimize PyUnicode_CompareWithASCIIString() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/34e166d60f37/Objects/unicodeobject.c
Use fast memcmp() instead of a loop using the slow PyUnicode_READ() macro.
strlen() is still necessary to check Unicode string containing null bytes.]]> Victor Stinner <victor.stinner@gmail.com> Tue, 29 Oct 2013 23:31:50 +0100 Issue #19437: Fix _PyUnicode_New() (constructor of legacy string), set all https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/22ce0b7b1eaf/Objects/unicodeobject.c attributes before checking for error. The destructor expects all attributes to
be set. It is now safe to call Py_DECREF(unicode) in the constructor.]]> Victor Stinner <victor.stinner@gmail.com> Tue, 29 Oct 2013 19:31:43 +0100 Issue #18609: Add a fast-path for "iso8859-1" encoding https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/31f9c1481cfa/Objects/unicodeobject.c
On AIX, the locale encoding may be "iso8859-1", which was not a known syntax of
the legacy ISO 8859-1 encoding.

Using a C codec instead of a Python codec is faster but also avoids tricky
issues during Python startup or complex code.]]> Victor Stinner <victor.stinner@gmail.com> Tue, 29 Oct 2013 11:34:05 +0100 Issue #18408: Fix PyUnicode_AsUTF8AndSize(), raise MemoryError exception on https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/e1d51c42e5a1/Objects/unicodeobject.c memory allocation failure]]> Victor Stinner <victor.stinner@gmail.com> Tue, 29 Oct 2013 01:28:23 +0100 Issue #1772673: The type of `char*` arguments now changed to `const char*`. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/3055e61f1e66/Objects/unicodeobject.c Serhiy Storchaka <storchaka@gmail.com> Sat, 19 Oct 2013 21:03:34 +0300 Issue #19279: UTF-7 decoder no more produces illegal strings. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/7dde9c553f16/Objects/unicodeobject.c Serhiy Storchaka <storchaka@gmail.com> Sat, 19 Oct 2013 20:39:28 +0300 Issue #16612: Add "Argument Clinic", a compile-time preprocessor https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/8fde1a2c94dc/Objects/unicodeobject.c for C files to generate argument parsing code. (See PEP 436.)]]> Larry Hastings <larry@hastings.org> Sat, 19 Oct 2013 00:09:25 -0700 Close #18780: %-formatting now prints value for int subclasses with %d, %i, and %u codes. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/33727fbb4668/Objects/unicodeobject.c Ethan Furman <ethan@stoneleaf.us> Sat, 31 Aug 2013 10:18:55 -0700 Issue #18722: Remove uses of the "register" keyword in C code. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/e7f6cef7a4cc/Objects/unicodeobject.c Antoine Pitrou <solipsis@pitrou.net> Tue, 13 Aug 2013 20:18:52 +0200 merge https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/f1cb19264a1f/Objects/unicodeobject.c Raymond Hettinger <python@rcn.com> Sun, 04 Aug 2013 11:51:35 -0700 Check return value of PyType_Ready(&EncodingMapType) https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/a6891949b54a/Objects/unicodeobject.c CID 486654]]> Christian Heimes <christian@cheimes.de> Sat, 20 Jul 2013 14:57:28 +0200 Issue #18408: Don't check unicode consistency in _PyUnicode_HAS_UTF8_MEMORY() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/ef0dbe00760e/Objects/unicodeobject.c and _PyUnicode_HAS_WSTR_MEMORY() macros

These macros are called in unicode_dealloc(), whereas the unicode object can be
"inconsistent" if the creation of the object failed.

For example, when unicode_subtype_new() fails on a memory allocation,
_PyUnicode_CheckConsistency() fails with an assertion error because data is
NULL.]]> Victor Stinner <victor.stinner@gmail.com> Mon, 15 Jul 2013 18:22:47 +0200 Issue #18408: _PyUnicodeWriter_Finish() now clears its buffer attribute in all https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/d605c6b8095c/Objects/unicodeobject.c cases, so _PyUnicodeWriter_Dealloc() can be called after finish.]]> Victor Stinner <victor.stinner@gmail.com> Tue, 09 Jul 2013 00:37:24 +0200 Issue #18408: Fix _PyUnicodeWriter_Finish(): clear writer->buffer, https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/df8b40593a08/Objects/unicodeobject.c so _PyUnicodeWriter_Dealloc() can be called on the writer after finish.]]> Victor Stinner <victor.stinner@gmail.com> Mon, 08 Jul 2013 22:29:55 +0200 Issue #18203: Fix _Py_DecodeUTF8_surrogateescape(), use PyMem_RawMalloc() as _Py_char2wchar() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/31a635303e55/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 07 Jul 2013 22:57:45 +0200 Issue #18203: Replace malloc() with PyMem_RawMalloc() at Python initialization https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/18bb92b0c458/Objects/unicodeobject.c
* Replace malloc() with PyMem_RawMalloc()
* Replace PyMem_Malloc() with PyMem_RawMalloc() where the GIL is not held.
* _Py_char2wchar() now returns a buffer allocated by PyMem_RawMalloc(), instead
of PyMem_Malloc()]]> Victor Stinner <victor.stinner@gmail.com> Sun, 07 Jul 2013 16:25:15 +0200 Fix ref leak in error case of unicode find, count, formatlong https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/cad335e0b205/Objects/unicodeobject.c CID 983315: Resource leak (RESOURCE_LEAK)
CID 983316: Resource leak (RESOURCE_LEAK)
CID 983317: Resource leak (RESOURCE_LEAK)]]> Christian Heimes <christian@cheimes.de> Sat, 29 Jun 2013 21:33:36 +0200 Fix ref leak in error case of unicode index https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/ec036132cc28/Objects/unicodeobject.c CID 983319 (#1 of 2): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable substring going out of scope leaks the storage it points to.]]> Christian Heimes <christian@cheimes.de> Sat, 29 Jun 2013 21:21:37 +0200 Fix ref leak in error case of unicode rindex and rfind https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/d0fb17896c95/Objects/unicodeobject.c CID 983320: Resource leak (RESOURCE_LEAK)
CID 983321: Resource leak (RESOURCE_LEAK)
leaked_storage: Variable substring going out of scope leaks the storage it points to.]]> Christian Heimes <christian@cheimes.de> Sat, 29 Jun 2013 21:17:34 +0200 Fix memory leak in endswith https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/787b0f99d77b/Objects/unicodeobject.c CID 1040368 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable substring going out of scope leaks the storage it points to.]]> Christian Heimes <christian@cheimes.de> Sat, 29 Jun 2013 20:41:06 +0200 Issue #18184: PyUnicode_FromFormat() and PyUnicode_FromFormatV() now raise https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/42def600210e/Objects/unicodeobject.c OverflowError when an argument of %c format is out of range.]]> Serhiy Storchaka <storchaka@gmail.com> Sun, 23 Jun 2013 20:21:16 +0300 merge 3.3 (#18183) https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/668aba845fb2/Objects/unicodeobject.c Benjamin Peterson <benjamin@python.org> Mon, 10 Jun 2013 09:24:01 -0700 Issue #9566: Fix compiler warning on Windows 64-bit https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/5dcbd5d8d004/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Wed, 05 Jun 2013 00:21:31 +0200 Issue #17237: Fix crash in the ASCII decoder on m68k. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/201ae2d02328/Objects/unicodeobject.c Antoine Pitrou <solipsis@pitrou.net> Sat, 11 May 2013 15:59:37 +0200 Fix uninitialized value in charmap_decode_mapping() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/1c1f60ccf305/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Tue, 07 May 2013 01:01:31 +0200 Issue #7330: Implement width and precision (ex: "%5.3s") for the format string https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/9e0f1c3bf9b6/Objects/unicodeobject.c of PyUnicode_FromFormat() function, original patch written by Ysj Ray.]]> Victor Stinner <victor.stinner@gmail.com> Mon, 06 May 2013 23:11:54 +0200 Partial revert of changeset 9744b2df134c https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/cb392bce91e6/Objects/unicodeobject.c
PyUnicode_Append() cannot call directly resize_compact(): I forgot that a
string can be ready *and* not compact (a legacy string can also be ready).]]> Victor Stinner <victor.stinner@gmail.com> Thu, 18 Apr 2013 09:41:34 +0200 Split PyUnicode_DecodeCharmap() into subfunction for readability https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/3e02d70cd07b/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Thu, 18 Apr 2013 01:44:27 +0200 Fix bug in Unicode decoders related to _PyUnicodeWriter https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/94d1c3bdb79c/Objects/unicodeobject.c
Bug introduced by changesets 7ed9993d53b4 and edf029fc9591.]]> Victor Stinner <victor.stinner@gmail.com> Thu, 18 Apr 2013 00:25:28 +0200 Fix typo in unicode_decode_call_errorhandler_writer() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/7eb52460c999/Objects/unicodeobject.c
Bug introduced by changeset 7ed9993d53b4.]]> Victor Stinner <victor.stinner@gmail.com> Wed, 17 Apr 2013 23:58:16 +0200 Close #17694: Add minimum length to _PyUnicodeWriter https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/edf029fc9591/Objects/unicodeobject.c
* Add also min_char attribute to _PyUnicodeWriter structure (currently unused)
* _PyUnicodeWriter_Init() has no more argument (except the writer itself):
min_length and overallocate must be set explicitly
* In error handlers, only enable overallocation if the replacement string
is longer than 1 character
* CJK decoders don't use overallocation anymore
* Set min_length, instead of preallocating memory using
_PyUnicodeWriter_Prepare(), in many decoders
* _PyUnicode_DecodeUnicodeInternal() checks for integer overflow]]> Victor Stinner <victor.stinner@gmail.com> Wed, 17 Apr 2013 23:02:17 +0200 Cleanup PyUnicode_Contains() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/a88310d86455/Objects/unicodeobject.c
* No need to double-check that strings are ready: test already done by
PyUnicode_FromObject()
* Remove useless kind variable (use kind1 instead)]]> Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 19:22:47 +0200 Minor change: fix character in do_strip() for the ASCII case https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/46f781a8d64f/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 19:17:42 +0200 Cleanup PyUnicode_Append() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/9744b2df134c/Objects/unicodeobject.c
* Check also that right is a Unicode object
* call directly resize_compact() instead of unicode_resize() for a more
explicit error handling, and to avoid testing some properties twice
(ex: unicode_modifiable())]]> Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 19:13:03 +0200 PyUnicode_Join(): move use_memcpy test out of the loop to cleanup and optimize the code https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/f7bc775f7ef3/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 18:56:46 +0200 Optimize repr(str): use _PyUnicode_FastCopyCharacters() when no character is escaped https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/870fddd176c8/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 18:45:39 +0200 Optimize ascii(str): don't encode/decode repr if repr is already ASCII https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/ff89957e3204/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 18:44:10 +0200 Add _PyUnicodeWriter_WriteCharInline() https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/0ee785c9d1b4/Objects/unicodeobject.c Victor Stinner <victor.stinner@gmail.com> Sun, 14 Apr 2013 02:35:33 +0200 Issue #16061: Speed up str.replace() for replacing 1-character strings. https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/d396e0716bf4/Objects/unicodeobject.c Serhiy Storchaka <storchaka@gmail.com> Sat, 13 Apr 2013 22:45:04 +0300 Close #17693: Rewrite CJK decoders to use the _PyUnicodeWriter API instead of https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/bcecf3910162/Objects/unicodeobject.c the legacy Py_UNICODE API.

Add also a new _PyUnicodeWriter_WriteChar() function.]]> Victor Stinner <victor.stinner@gmail.com> Thu, 11 Apr 2013 22:09:04 +0200 Issue #17615: On Windows (VS2010), Performances of wmemcmp() to compare Unicode https://hg.dopieralski.pl/cpython-lru_cache-weakref/log/b3168643677b/Objects/unicodeobject.c strings are not convincing. For UCS2 (16-bit wchar_t type), use a dummy loop
instead of wmemcmp(). The dummy loop is as fast, or a little bit faster.

wchar_t is only 16-bit long on Windows. wmemcmp() is still used for 32-bit
wchar_t.]]> Victor Stinner <victor.stinner@gmail.com> Tue, 09 Apr 2013 23:53:26 +0200