fbb1c5ee06
Issue #26494 : Fixed crash on iterating exhausting iterators.
...
Affected classes are generic sequence iterators, iterators of str, bytes,
bytearray, list, tuple, set, frozenset, dict, OrderedDict, corresponding
views and os.scandir() iterator.
2016-03-30 20:40:02 +03:00
3625af5f21
Moved misplaced functions to the section for C API functions.
2016-03-27 01:15:07 -07:00
2c257ab0f8
Responsibility for argument checking belongs in set.__init__() rather than set.__new__().
...
See dict.__new__() and list.__new__() for comparison. Neither of those examine or touch
args or kwds. That work is done in the __init__() methods.
2016-03-26 04:10:11 -07:00
b72e21b9ab
Speed-up construction of empty sets by approx 12-14%.
2016-03-25 02:29:59 -07:00
f963c13597
longobject.c: fix compilation warning on Windows 64-bit
...
We know that Py_SIZE(b) is -1 or 1 an so fits into the sdigit type.
2016-03-23 18:36:54 +01:00
0aed3a4ebc
_PyMem_DebugFree(): fix compiler warning on Windows
...
Don't return a void value.
2016-03-23 11:30:43 +01:00
fac395681f
Optimize bytes.replace(b'', b'.')
...
Issue #26574 : Optimize bytes.replace(b'', b'.') and
bytearray.replace(b'', b'.'): up to 80% faster. Patch written by Josh Snider.
2016-03-21 10:38:58 +01:00
ad524375af
Fail if PyMem_Malloc() is called without holding the GIL
...
Issue #26563 : Debug hooks on Python memory allocators now raise a fatal error
if functions of the PyMem_Malloc() family are called without holding the GIL.
2016-03-16 12:12:53 +01:00
ffcf1a54d3
Oops, revert unwanted change used to create an example
...
Issue #26564 .
2016-03-15 22:49:40 +01:00
0611c26a58
On memory error, dump the memory block traceback
...
Issue #26564 : _PyObject_DebugDumpAddress() now dumps the traceback where a
memory block was allocated on memory block. Use the tracemalloc module to get
the traceback.
2016-03-15 22:22:13 +01:00
c4aec3628b
Check the GIL in PyObject_Malloc()
...
Issue #26558 : The debug hook of PyObject_Malloc() now checks that the GIL is
held when the function is called.
2016-03-14 22:26:53 +01:00
34be807ca4
Add PYTHONMALLOC env var
...
Issue #26516 :
* Add PYTHONMALLOC environment variable to set the Python memory
allocators and/or install debug hooks.
* PyMem_SetupDebugHooks() can now also be used on Python compiled in release
mode.
* The PYTHONMALLOCSTATS environment variable can now also be used on Python
compiled in release mode. It now has no effect if set to an empty string.
* In debug mode, debug hooks are now also installed on Python memory allocators
when Python is configured without pymalloc.
2016-03-14 12:04:26 +01:00
0f04bc7959
merge 3.5 ( closes #26478 )
2016-03-03 22:10:52 -08:00
cfc2a1fc70
merge 3.4 ( closes #26478 )
2016-03-03 22:08:01 -08:00
f11b25b081
properly use the ObjArgs variant of CallMethod in dictview binary operations ( closes #26478 )
2016-03-03 22:05:36 -08:00
d2dc15b26b
Merge 3.5 (issue #25888 )
2016-03-02 11:31:06 -05:00
c724bae51c
coroutines: Error when awaiting on coroutine that's being awaited
...
Issue #25888
2016-03-02 11:30:46 -05:00
f2192855dd
Merge 3.5
2016-03-01 22:07:53 +01:00
337986740f
Issue #26464 : Fix unicode_fast_translate() again
...
Initialize i variable if the string is non-ASCII.
2016-03-01 21:59:58 +01:00
3d9d77a3dc
Merge 3.5
2016-03-01 21:30:50 +01:00
6c9aa8f2bf
Fix str.translate()
...
Issue #26464 : Fix str.translate() when string is ASCII and first replacements
removes character, but next replacement uses a non-ASCII character or a string
longer than 1 character. Regression introduced in Python 3.5.0.
2016-03-01 21:30:30 +01:00
3e48b38dff
Merge 3.5 (issue #25887 )
2016-02-13 18:00:31 -05:00
77c96813ab
Issue #25887 : Raise a RuntimeError when a coroutine is awaited more than once.
2016-02-13 17:59:05 -05:00
e0b23095ee
Issues #26289 and #26315 : Optimize floor/modulo div for single-digit longs
...
Microbenchmarks show 2-2.5x improvement. Built-in 'divmod' function
is now also ~10% faster.
-m timeit -s "x=22331" "x//2;x//-3;x//4;x//5;x//-6;x//7;x//8;x//-99;x//100;"
with patch: 0.321 without patch: 0.633
-m timeit -s "x=22331" "x%2;x%3;x%-4;x%5;x%6;x%-7;x%8;x%99;x%-100;"
with patch: 0.224 without patch: 0.66
Big thanks to Serhiy Storchaka, Mark Dickinson and Victor Stinner for
thorow code reviews and algorithms improvements.
2016-02-11 10:26:27 -05:00
7e3a91a5fc
Issue #26136 : Upgrade the generator_stop warning to DeprecationWarning
...
Patch by Anish Shah.
2016-02-10 04:40:48 +00:00
d2962f145a
Issue #25949 : __dict__ for an OrderedDict instance is now created only when
...
needed.
2016-02-08 16:39:05 +02:00
a0fcaca4e1
Issue #26288 : Fix comment
2016-02-06 12:21:33 -05:00
186c30b7ae
Issue #26288 : Optimize PyLong_AsDouble.
2016-02-05 19:40:01 -05:00
f50215412c
Add early-out for the common case where kwds is NULL (gives 1.1% speedup).
2016-02-04 02:46:16 -08:00
ce5179fcba
Issue #23601 : Use small object allocator for dict key objects
2016-01-31 08:56:21 -08:00
5b96f17b1c
Merge 3.5
2016-01-27 17:01:13 +01:00
5bc03a6d4d
Fix resize_compact()
...
Issue #26217 : resize_compact() must set wstr_length to 0 after freeing the wstr
string. Otherwise, an assertion fails in _PyUnicode_CheckConsistency().
2016-01-27 16:56:53 +01:00
0dceb91866
Tests versus zero are more compact than tests versus -1.
2016-01-25 10:33:30 -08:00
3cdd5fb970
code_richcompare() now uses the constants types
...
Issue #25843 : When compiling code, don't merge constants if they are equal but
have a different types. For example, "f1, f2 = lambda: 1, lambda: 1.0" is now
correctly compiled to two different functions: f1() returns 1 (int) and f2()
returns 1.0 (int), even if 1 and 1.0 are equal.
Add a new _PyCode_ConstantKey() private function.
2016-01-22 12:33:12 +01:00
efb2413ce8
code_richcompare() now uses the constants types
...
Issue #25843 : When compiling code, don't merge constants if they are equal but
have a different types. For example, "f1, f2 = lambda: 1, lambda: 1.0" is now
correctly compiled to two different functions: f1() returns 1 (int) and f2()
returns 1.0 (int), even if 1 and 1.0 are equal.
Add a new _PyCode_ConstantKey() private function.
2016-01-22 12:33:12 +01:00
9f78939552
Issue #26107 : Fix typo in Objects/lnotab_notes.txt
...
Double parenthesis
2016-01-21 18:12:29 +01:00
f3914eb16d
co_lnotab supports negative line number delta
...
Issue #26107 : The format of the co_lnotab attribute of code objects changes to
support negative line number delta.
Changes:
* assemble_lnotab(): if line number delta is less than -128 or greater than
127, emit multiple (offset_delta, lineno_delta) in co_lnotab
* update functions decoding co_lnotab to use signed 8-bit integers
- dis.findlinestarts()
- PyCode_Addr2Line()
- _PyCode_CheckLineNumber()
- frame_setlineno()
* update lnotab_notes.txt
* increase importlib MAGIC_NUMBER to 3361
* document the change in What's New in Python 3.6
* cleanup also PyCode_Optimize() to use better variable names
2016-01-20 12:16:21 +01:00
b56837a033
Merge 3.5
...
Issue #26154 : Add a new private _PyThreadState_UncheckedGet() function.
2016-01-20 11:19:46 +01:00
bfd316e750
Add _PyThreadState_UncheckedGet()
...
Issue #26154 : Add a new private _PyThreadState_UncheckedGet() function which
gets the current thread state, but don't call Py_FatalError() if it is NULL.
Python 3.5.1 removed the _PyThreadState_Current symbol from the Python C API to
no more expose complex and private atomic types. Atomic types depends on the
compiler or can even depend on compiler options. The new function
_PyThreadState_UncheckedGet() allows to get the variable value without having
to care of the exact implementation of atomic types.
Changes:
* Replace direct usage of the _PyThreadState_Current variable with a call to
_PyThreadState_UncheckedGet().
* In pystate.c, replace direct usage of the _PyThreadState_Current variable
with the PyThreadState_GET() macro for readability.
* Document also PyThreadState_Get() in pystate.h
2016-01-20 11:12:38 +01:00
4918b47c64
Issue #25935 : Garbage collector now breaks reference loops with OrderedDict.
2016-01-19 14:48:42 +02:00
d205d0145c
Issue #25935 : Garbage collector now breaks reference loops with OrderedDict.
2016-01-19 14:46:25 +02:00
2c028492d4
merge 3.5 ( #25731 )
2016-01-18 21:18:35 -08:00
9ad11544bf
set tp_new from the class in the hierarchy that actually owns the descriptor ( closes #25731 )
...
Debugging by Eryk Sun.
2016-01-18 21:11:18 -08:00
3b63dc3aa9
- merge 3.5
2016-01-14 14:05:21 +01:00
17210f50d2
dictobject.c(dict_sizeof): Make it static again.
2016-01-14 14:04:59 +01:00
3d220cc4d7
As per further discussion, re-enable the typeobject.c guard for picklability.
...
upstreams such as Cython will have to adjust.
2016-01-11 15:51:27 -05:00
09880c89e9
Comment out some tests that won't pass now that we've reverted the
...
picklability regression. Also, as per further discussion, remove the
regressing code.
2016-01-11 15:14:53 -05:00
ec71f1779f
Issue #22995 : [UPDATE] Comment out the one of the pickleability tests in
...
_PyObject_GetState() due to regressions observed in Cython-based projects.
2016-01-11 13:49:40 -05:00
885e1939be
Issue #22995 : [UPDATE] Comment out the one of the pickleability tests in
...
_PyObject_GetState() due to regressions observed in Cython-based projects.
2016-01-11 13:24:02 -05:00
576f132b98
Issue #20440 : Cleaning up the code by using Py_SETREF.
2016-01-05 21:27:54 +02:00
3e47a1337c
merge 3.5
2016-01-01 11:56:35 -06:00
4e3dd51396
merge 3.4
2016-01-01 11:56:16 -06:00
630329e4ea
merge 3.3
2016-01-01 11:55:47 -06:00
0e617e22f0
remove some copyright notices supserseded by the toplevel ones
2016-01-01 11:53:47 -06:00
288cb25f1a
Issue #25961 : Disallowed null characters in the type name.
...
Simplified testing for null characters in __name__ setter.
2015-12-30 21:41:53 +02:00
42bf8fc901
Issue #25961 : Disallowed null characters in the type name.
...
Simplified testing for null characters in __name__ setter.
2015-12-30 21:40:49 +02:00
4e6aad1f7a
Clean up float.as_integer_ratio().
2015-12-29 22:55:48 +02:00
0d250bc119
Issue #25971 : Optimized creating Fractions from floats by 2 times and from
...
Decimals by 3 times.
Unified error messages in float.as_integer_ratio(), Decimal.as_integer_ratio(),
and Fraction constructors.
2015-12-29 22:34:23 +02:00
1ed017ae92
Issue #20440 : Cleaning up the code by using Py_SETREF and Py_CLEAR.
...
Old code is correct, but with Py_SETREF and Py_CLEAR it can be cleaner.
This patch doesn't fix bugs and hence there is no need to backport it.
2015-12-27 15:51:32 +02:00
726fc139a5
Issue #20440 : More use of Py_SETREF.
...
This patch is manually crafted and contains changes that couldn't be handled
automatically.
2015-12-27 15:44:33 +02:00
191321d11b
Issue #20440 : More use of Py_SETREF.
...
This patch is manually crafted and contains changes that couldn't be handled
automatically.
2015-12-27 15:41:34 +02:00
bdb908ea54
Issue #20440 : Applied yet one patch for using Py_SETREF.
...
The patch is automatically generated, it replaces the code that uses Py_CLEAR.
2015-12-27 12:38:28 +02:00
4a1e70fc31
Issue #20440 : Applied yet one patch for using Py_SETREF.
...
The patch is automatically generated, it replaces the code that uses Py_CLEAR.
2015-12-27 12:36:18 +02:00
f9253c96fd
Issue #22995 : Instances of extension types with a state that aren't
...
subclasses of list or dict and haven't implemented any pickle-related
methods (__reduce__, __reduce_ex__, __getnewargs__, __getnewargs_ex__,
or __getstate__), can no longer be pickled. Including memoryview.
2015-12-25 21:05:35 +02:00
f81be8aa3f
Issue #22995 : Instances of extension types with a state that aren't
...
subclasses of list or dict and haven't implemented any pickle-related
methods (__reduce__, __reduce_ex__, __getnewargs__, __getnewargs_ex__,
or __getstate__), can no longer be pickled. Including memoryview.
2015-12-25 21:04:29 +02:00
ef1585eb9a
Issue #25923 : Added more const qualifiers to signatures of static and private functions.
2015-12-25 20:01:53 +02:00
2d06e84455
Issue #25923 : Added the const qualifier to static constant arrays.
2015-12-25 19:53:18 +02:00
f006940351
Issue #20440 : Massive replacing unsafe attribute setting code with special
...
macro Py_SETREF.
2015-12-24 10:39:57 +02:00
5a57ade58e
Issue #20440 : Massive replacing unsafe attribute setting code with special
...
macro Py_SETREF.
2015-12-24 10:35:59 +02:00
3987e26e75
Issue #25914 : Fixed and simplified OrderedDict.__sizeof__.
2015-12-22 08:22:05 +02:00
0ce7a3a34c
Issue #25914 : Fixed and simplified OrderedDict.__sizeof__.
2015-12-22 08:16:18 +02:00
3182db356c
Issue #25766 : Special method __bytes__() now works in str subclasses.
2015-12-20 16:37:21 +02:00
5aac3ed799
Issue #25766 : Special method __bytes__() now works in str subclasses.
2015-12-20 16:36:34 +02:00
a9406e77fa
Issue #25421 : __sizeof__ methods of builtin types now use dynamic basic size.
...
This allows sys.getsize() to work correctly with their subclasses with
__slots__ defined.
2015-12-19 20:07:11 +02:00
5c4064e8bd
Issue #25421 : __sizeof__ methods of builtin types now use dynamic basic size.
...
This allows sys.getsize() to work correctly with their subclasses with
__slots__ defined.
2015-12-19 20:05:25 +02:00
42bb126f0a
Issue #25899 : Converted Objects/listsort.txt to UTF-8.
...
Original patch by Chris Angelico.
2015-12-18 13:23:33 +02:00
9b3a2eec1c
Issues #25890 , #25891 , #25892 : Removed unused variables in Windows code.
...
Reported by Alexander Riccio.
2015-12-18 10:03:13 +02:00
e4495877dd
Minor tweek. Counting down rather than up reduces register pressure.
2015-12-15 00:42:30 -08:00
86d322f020
Undo inadvertent line swap
2015-12-13 19:27:17 -08:00
5088f6005f
Hoist constant expressions (so->table and so->mask) out of the inner-loop.
2015-12-13 18:45:01 -08:00
7c088a9b5c
Issue #25709 : Fixed problem with in-place string concatenation and utf-8 cache.
2015-12-03 01:05:52 +02:00
6648bf5661
Issue #25709 : Fixed problem with in-place string concatenation and utf-8 cache.
2015-12-03 01:04:37 +02:00
7aa690860e
Issue #25709 : Fixed problem with in-place string concatenation and utf-8 cache.
2015-12-03 01:02:03 +02:00
6a50e79130
Issue #24097 : Fixed crash in object.__reduce__() if slot name is freed inside
...
__getattr__. Original patch by Antoine Pitrou.
2015-11-25 18:35:15 +02:00
670d78abc5
Issue #24097 : Fixed crash in object.__reduce__() if slot name is freed inside
...
__getattr__. Original patch by Antoine Pitrou.
2015-11-25 18:34:19 +02:00
d28bb624d1
Issue #24097 : Fixed crash in object.__reduce__() if slot name is freed inside
...
__getattr__. Original patch by Antoine Pitrou.
2015-11-25 18:33:29 +02:00
bb6e4a0b31
Issue #24731 : Fixed crash on converting objects with special methods
...
__bytes__, __trunc__, and __float__ returning instances of subclasses of
bytes, int, and float to subclasses of bytes, int, and float correspondingly.
2015-11-25 15:53:19 +02:00
f9afda57ad
Issue #24731 : Fixed crash on converting objects with special methods
...
__bytes__, __trunc__, and __float__ returning instances of subclasses of
bytes, int, and float to subclasses of bytes, int, and float correspondingly.
2015-11-25 15:52:04 +02:00
15095800a3
Issue #24731 : Fixed crash on converting objects with special methods
...
__bytes__, __trunc__, and __float__ returning instances of subclasses of
bytes, int, and float to subclasses of bytes, int, and float correspondingly.
2015-11-25 15:47:01 +02:00
b4efc963d6
Issue #25557 : Refactor _PyDict_LoadGlobal()
...
Don't fallback to PyDict_GetItemWithError() if the hash is unknown: compute the
hash instead. Add also comments to explain the optimization a little bit.
2015-11-20 09:24:02 +01:00
66f6238fca
Add assertion to verify the pre-condition in the comments.
2015-11-17 20:58:43 -08:00
6019c8ced0
Issue #25629 : Move set fill/used updates out of inner loop
2015-11-17 08:28:07 -08:00
d798dc1034
merge 3.5 ( #25630 )
2015-11-15 21:57:50 -08:00
a4d33b3428
make the PyUnicode_FSConverter cleanup set the decrefed argument to NULL ( closes #25630 )
2015-11-15 21:57:39 -08:00
413fdcea21
Issue #24821 : Refactor STRINGLIB(fastsearch_memchr_1char) and split it on
...
STRINGLIB(find_char) and STRINGLIB(rfind_char) that can be used independedly
without special preconditions.
2015-11-14 15:42:17 +02:00
1010921c71
Issue #25462 : The hash of the key now is calculated only once in most
...
operations in C implementation of OrderedDict.
2015-11-13 15:18:26 +02:00
19a70e7f5d
Issue #25462 : The hash of the key now is calculated only once in most
...
operations in C implementation of OrderedDict.
2015-11-13 14:48:36 +02:00
12ab296f82
Issue #22995 : Default implementation of __reduce__ and __reduce_ex__ now
...
rejects builtin types with not defined __new__.
Added tests for non-pickleable types.
2015-11-12 11:34:39 +02:00
609a2e17ad
Issue #22995 : Default implementation of __reduce__ and __reduce_ex__ now
...
rejects builtin types with not defined __new__.
Added tests for non-pickleable types.
2015-11-12 11:31:51 +02:00
d7a4415599
Issue #22995 : Default implementation of __reduce__ and __reduce_ex__ now
...
rejects builtin types with not defined __new__.
Added tests for non-pickleable types.
2015-11-12 11:23:04 +02:00
e2cf984384
Merge 3.5.
2015-11-10 18:35:47 +01:00
7c3f3859a5
Fix Visual Studio warning.
2015-11-10 18:35:19 +01:00
3407cc26cb
Merge 3.5.
2015-11-10 18:18:07 +01:00
0ce5b6e268
Iaaue #25598 : Fix memory_hex from #9951 for non-contiguous buffers.
2015-11-10 18:17:22 +01:00
9513ba3b8f
Issue #24802 : Merge null termination fixes from 3.5
2015-11-07 03:15:32 +00:00
61d6e4ae9d
Issue #24802 : Merge null termination fixes from 3.4 into 3.5
2015-11-07 02:56:11 +00:00
eeb896c411
Issue #24802 : Copy bytes-like objects to null-terminated buffers if necessary
...
This avoids possible buffer overreads when int(), float(), compile(), exec()
and eval() are passed bytes-like objects. Similar code is removed from the
complex() constructor, where it was not reachable.
Patch by John Leitch, Serhiy Storchaka and Martin Panter.
2015-11-07 02:32:21 +00:00
0c009bf3b5
Issue #25410 : Made testing that od_fast_nodes and dk_entries are in sync more
...
reliable.
2015-11-06 12:00:43 +02:00
97f46db904
Issue #25410 : Made testing that od_fast_nodes and dk_entries are in sync more
...
reliable.
2015-11-06 12:00:03 +02:00
3a811f272a
Issue #24726 : Revert setting the value on the dict if
...
_odict_add_new_node() fails.
2015-11-06 11:07:31 +02:00
d5f353ec8d
Issue #24726 : Revert setting the value on the dict if
...
_odict_add_new_node() fails.
2015-11-06 11:07:11 +02:00
6980be40bd
Issue #25558 : Refactoring OrderedDict iteration.
2015-11-06 10:40:05 +02:00
9c967611e3
Issue #25558 : Refactoring OrderedDict iteration.
2015-11-06 10:39:51 +02:00
e20310fa19
Issue #25556 : Add assertions to PyObject_GetItem() to ensure that an exception
...
is raised when it returns NULL.
Simplify also ceval.c: rely on the fact that PyObject_GetItem() raised an
exception when it returns NULL.
2015-11-05 13:56:58 +01:00
f83afa508f
Issue #25449 : Fixed a crash and leaking NULL in repr() of OrderedDict that
...
was mutated by direct calls of dict methods.
2015-11-04 22:33:33 +02:00
710cd34bdb
Issue #25449 : Fixed a crash and leaking NULL in repr() of OrderedDict that
...
was mutated by direct calls of dict methods.
2015-11-04 22:33:07 +02:00
161b6ff19c
Issue #25449 : Iterating OrderedDict with keys with unstable hash now raises
...
KeyError in C implementations as well as in Python implementation.
Added tests for OrderedDict subclasses.
2015-11-04 22:08:52 +02:00
b45b7b2137
Issue #25449 : Iterating OrderedDict with keys with unstable hash now raises
...
KeyError in C implementations as well as in Python implementation.
Added tests for OrderedDict subclasses.
2015-11-04 22:05:38 +02:00
4a7c03aab4
Issue #25523 : Merge a-to-an corrections from 3.5.
2015-11-02 14:44:29 +02:00
a84f6c3dd3
Issue #25523 : Merge a-to-an corrections from 3.4.
2015-11-02 14:39:05 +02:00
d65c9496da
Issue #25523 : Further a-to-an corrections.
2015-11-02 14:10:23 +02:00
e56a919100
Issue #25523 : Merge a-to-an corrections from 3.5
2015-11-02 04:27:17 +00:00
2eb819f7a8
Issue #25523 : Merge "a" to "an" fixes from 3.4 into 3.5
2015-11-02 04:04:57 +00:00
7462b64911
Issue #25523 : Correct "a" article to "an" article
...
This changes the main documentation, doc strings, source code comments, and a
couple error messages in the test suite. In some cases the word was removed
or edited some other way to fix the grammar.
2015-11-02 03:37:02 +00:00
9b69491901
Issue #25395 : Fixed crash when highly nested OrderedDict structures were
...
garbage collected.
2015-11-01 16:13:45 +02:00
14eefe353e
Issue #25395 : Fixed crash when highly nested OrderedDict structures were
...
garbage collected.
2015-11-01 16:12:34 +02:00
afbea7af02
Issue #25410 : C implementation of OrderedDict now uses type(self) instead of
...
self.__class__ in __repr__() and __reduce__() for simplicity and reliability.
2015-10-22 20:19:14 +03:00
4575beba4b
Issue #25410 : C implementation of OrderedDict now uses type(self) instead of
...
self.__class__ in __repr__() and __reduce__() for simplicity and reliability.
2015-10-22 20:18:24 +03:00
60a08929af
Issue #25410 : Fixed a memory leak in OrderedDict in the case when key's hash
...
calculation fails.
2015-10-20 18:22:36 +03:00
d17427b7bd
Issue #25410 : Fixed a memory leak in OrderedDict in the case when key's hash
...
calculation fails.
2015-10-20 18:21:48 +03:00
3e9f87782e
Issue #25410 : Cleaned up and fixed minor bugs in C implementation of OrderedDict.
2015-10-18 09:54:42 +03:00
8003bafd7f
Issue #25410 : Cleaned up and fixed minor bugs in C implementation of OrderedDict.
2015-10-18 09:53:17 +03:00
91108f049f
Issue #25210 : Change error message of do_richcompare()
...
Don't add parenthesis to type names. Add also quotes around the type names.
Before:
TypeError: unorderable types: int() < NoneType()
After:
TypeError: '<' not supported between instances of 'int' and 'NoneType'
2015-10-14 18:25:31 +02:00
dbb98c1443
Issue #25406 : Fixed a bug in C implementation of OrderedDict.move_to_end()
...
that caused segmentation fault or hang in iterating after moving several
items to the start of ordered dict.
2015-10-14 19:22:44 +03:00
992ec46acc
Issue #25406 : Fixed a bug in C implementation of OrderedDict.move_to_end()
...
that caused segmentation fault or hang in iterating after moving several
items to the start of ordered dict.
2015-10-14 19:21:24 +03:00
c3d2bc19e4
Use _PyBytesWriter in _PyBytes_FromIterator()
2015-10-14 14:15:49 +02:00
c5c3ba4bec
Add _PyBytesWriter_Resize() function
...
This function gives a control to the buffer size without using min_size.
2015-10-14 13:56:47 +02:00
3c50ce39bf
Factorize _PyBytes_FromList() and _PyBytes_FromTuple() code using a C macro
2015-10-14 13:50:40 +02:00
f2eafa323b
Split PyBytes_FromObject() into subfunctions
2015-10-14 13:44:29 +02:00
2ec8063cc9
Modify _PyBytes_DecodeEscapeRecode() to use _PyBytesAPI
...
* Don't overallocate by 400% when recode is needed: only overallocate on demand
using _PyBytesWriter.
* Use _PyLong_DigitValue to convert hexadecimal digit to int
* Create _PyBytes_DecodeEscapeRecode() subfunction
2015-10-14 13:32:13 +02:00
1285e5c805
Fix compiler warnings (uninitialized variables), false alarms in fact
2015-10-14 12:10:20 +02:00
f6358a7e4c
_PyBytesWriter_Alloc(): only use 10 bytes of the small buffer in debug mode to
...
enhance code to detect buffer under- and overflow.
2015-10-14 12:02:39 +02:00
f091033b14
Issue #25401 : Remove now unused hex_digit_to_int() function
2015-10-14 11:59:46 +02:00
2bf8993db9
Optimize bytes.fromhex() and bytearray.fromhex()
...
Issue #25401 : Optimize bytes.fromhex() and bytearray.fromhex(): they are now
between 2x and 3.5x faster. Changes:
* Use a fast-path working on a char* string for ASCII string
* Use a slow-path for non-ASCII string
* Replace slow hex_digit_to_int() function with a O(1) lookup in
_PyLong_DigitValue precomputed table
* Use _PyBytesWriter API to handle the buffer
* Add unit tests to check the error position in error messages
2015-10-14 11:25:33 +02:00
772b2b09f2
Optimize bytearray % args
...
Issue #25399 : Don't create temporary bytes objects: modify _PyBytes_Format() to
create work directly on bytearray objects.
* Rename _PyBytes_Format() to _PyBytes_FormatEx() just in case if something
outside CPython uses it
* _PyBytes_FormatEx() now uses (char*, Py_ssize_t) for the input string, so
bytearray_format() doesn't need tot create a temporary input bytes object
* Add use_bytearray parameter to _PyBytes_FormatEx() which is passed to
_PyBytesWriter, to create a bytearray buffer instead of a bytes buffer
Most formatting operations are now between 2.5 and 5 times faster.
2015-10-14 09:56:53 +02:00
661aaccf9d
Add use_bytearray attribute to _PyBytesWriter
...
Issue #25399 : Add a new use_bytearray attribute to _PyBytesWriter to use a
bytearray buffer, instead of using a bytes object.
2015-10-14 09:41:48 +02:00
199c9a6f4b
Fix long_format_binary()
...
Issue #25399 : Fix long_format_binary(), allocate bytes for the bytes writer.
2015-10-14 09:47:23 +02:00
03dab786b2
Rewrite PyBytes_FromFormatV() using _PyBytesWriter API
...
* Add much more unit tests on PyBytes_FromFormatV()
* Remove the first loop to compute the length of the output string
* Use _PyBytesWriter to handle the bytes buffer, use overallocation
* Cleanup the code to make simpler and easier to review
2015-10-14 00:21:35 +02:00
358af13526
Issue #25353 : Optimize unicode escape and raw unicode escape encoders to use
...
the new _PyBytesWriter API.
2015-10-12 22:36:57 +02:00
e9aa5950bb
Fix compilation error in _PyBytesWriter_WriteBytes() on Windows
2015-10-12 13:57:47 +02:00
6c2cdae9e6
Writer APIs: use empty string singletons
...
Modify _PyBytesWriter_Finish() and _PyUnicodeWriter_Finish() to return the
empty bytes/Unicode string if the string is empty.
2015-10-12 13:29:43 +02:00
c29e29bed1
Relax _PyBytesWriter API
...
Don't require _PyBytesWriter pointer to be a "char *". Same change for
_PyBytesWriter_WriteBytes() parameter.
For example, binascii uses "unsigned char*".
2015-10-12 13:12:54 +02:00
0d554d7ef1
Issue #24164 : Objects that need calling ``__new__`` with keyword arguments,
...
can now be pickled using pickle protocols older than protocol version 4.
2015-10-10 22:42:18 +03:00
0cdad1e2bc
Issue #25349 : Add fast path for b'%c' % int
...
Optimize also %% formater.
2015-10-09 22:50:36 +02:00
be75b8cf23
Issue #25349 : Optimize bytes % int
...
Optimize bytes.__mod__(args) for integere formats: %d (%i, %u), %o, %x and %X.
_PyBytesWriter is now used to format directly the integer into the writer
buffer, instead of using a temporary bytes object.
Formatting is between 30% and 50% faster on a microbenchmark.
2015-10-09 22:43:24 +02:00
6bd525b656
Optimize error handlers of ASCII and Latin1 encoders when the replacement
...
string is pure ASCII: use _PyBytesWriter_WriteBytes(), don't check individual
character.
Cleanup unicode_encode_ucs1():
* Rename repunicode to rep
* Clear rep object on error
* Factorize code between bytes and unicode path
2015-10-09 13:10:05 +02:00
ce179bf6ba
Add _PyBytesWriter_WriteBytes() to factorize the code
2015-10-09 12:57:22 +02:00
ad7715891e
_PyBytesWriter: simplify code to avoid "prealloc" parameters
...
Substract preallocate bytes from min_size before calling
_PyBytesWriter_Prepare().
2015-10-09 12:38:53 +02:00
53926a1ce2
_PyBytesWriter: rename size attribute to min_size
2015-10-09 12:37:03 +02:00
fa7762ec06
Issue #25349 : Optimize bytes % args using the new private _PyBytesWriter API
...
* Thanks to the _PyBytesWriter API, output smaller than 512 bytes are allocated
on the stack and so avoid calling _PyBytes_Resize(). Because of that, change
the default buffer size to fmtcnt instead of fmtcnt+100.
* Rely on _PyBytesWriter algorithm to overallocate the buffer instead of using
a custom code. For example, _PyBytesWriter uses a different overallocation
factor (25% or 50%) depending on the platform to get best performances.
* Disable overallocation for the last write.
* Replace C loops to fill characters with memset()
* Add also many comments to _PyBytes_Format()
* Remove unused FORMATBUFLEN constant
* Avoid the creation of a temporary bytes object when formatting a floating
point number (when no custom formatting option is used)
* Fix also reference leaks on error handling
* Use Py_MEMCPY() to copy bytes between two formatters (%)
2015-10-09 11:48:06 +02:00
b3653a3458
Issue #25318 : cleanup code _PyBytesWriter
...
Rename "stack buffer" to "small buffer".
Add also an assertion in _PyBytesWriter_GetPos().
2015-10-09 03:38:24 +02:00
3fa36ff5e4
Issue #25318 : Fix backslashreplace()
...
Fix code to estimate the needed space.
2015-10-09 03:37:11 +02:00
797485e101
Issue #25318 : Avoid sprintf() in backslashreplace()
...
Rewrite backslashreplace() to be closer to PyCodec_BackslashReplaceErrors().
Add also unit tests for non-BMP characters.
2015-10-09 03:17:30 +02:00
b13b97d3b8
Issue #25318 : Fix compilation error
...
Replace "#if Py_DEBUG" with "#ifdef Py_DEBUG".
2015-10-09 02:52:16 +02:00
0016507c16
Issue #25318 : Move _PyBytesWriter to bytesobject.c
...
Declare also the private API in bytesobject.h.
2015-10-09 01:53:21 +02:00
e7bf86cd7d
Optimize backslashreplace error handler
...
Issue #25318 : Optimize backslashreplace and xmlcharrefreplace error handlers in
UTF-8 encoder. Optimize also backslashreplace error handler for ASCII and
Latin1 encoders.
Use the new _PyBytesWriter API to optimize these error handlers for the
encoders. It avoids to create an exception and call the slow implementation of
the error handler.
2015-10-09 01:39:28 +02:00
fdfbf78114
Issue #25318 : Add _PyBytesWriter API
...
Add a new private API to optimize Unicode encoders. It uses a small buffer
allocated on the stack and supports overallocation.
Use _PyBytesWriter API for UCS1 (ASCII and Latin1) and UTF-8 encoders. Enable
overallocation for the UTF-8 encoder with error handlers.
unicode_encode_ucs1(): initialize collend to collstart+1 to not check the
current character twice, we already know that it is not ASCII.
2015-10-09 00:33:49 +02:00
585a6acfef
Merge typo fixes from 3.5
2015-10-07 11:13:55 +00:00
ec1aa5c2a1
More typos in 3.5 documentation and comments
2015-10-07 11:03:53 +00:00
3f930dcd87
Merge typo fixes from 3.4 into 3.5
2015-10-07 11:01:47 +00:00
9955a373a8
Various minor typos in documentation and comments
2015-10-07 10:26:23 +00:00
cdae2cb88a
merge 3.5 ( closes #24806 )
2015-10-06 19:42:46 -07:00
59dc696821
merge 3.4 ( #24806 )
2015-10-06 19:42:02 -07:00
bd6c41a185
prevent unacceptable bases from becoming bases through multiple inheritance ( #24806 )
2015-10-06 19:36:54 -07:00
74e8fac3c8
Issue #25301 : Fix compatibility with ISO C90
2015-10-05 13:49:26 +02:00
1d65d9192d
Issue #25301 : The UTF-8 decoder is now up to 15 times as fast for error
...
handlers: ``ignore``, ``replace`` and ``surrogateescape``.
2015-10-05 13:43:50 +02:00
eb36fdaad8
Fix _PyUnicodeWriter_PrepareKind()
...
Initialize kind to 0 (PyUnicode_WCHAR_KIND) to ensure that
_PyUnicodeWriter_PrepareKind() handles correctly read-only buffer: copy the
buffer.
2015-10-03 01:55:51 +02:00
29e68edbf4
Issue #24848 : Fixed bugs in UTF-7 decoding of misformed data:
...
1. Non-ASCII bytes were accepted after shift sequence.
2. A low surrogate could be emitted in case of error in high surrogate.
3. In some circumstances the '\xfd' character was produced instead of the
replacement character '\ufffd' (due to a bug in _PyUnicodeWriter).
2015-10-02 13:14:03 +03:00
58c8f2bb6d
Issue #24848 : Fixed bugs in UTF-7 decoding of misformed data:
...
1. Non-ASCII bytes were accepted after shift sequence.
2. A low surrogate could be emitted in case of error in high surrogate.
3. In some circumstances the '\xfd' character was produced instead of the
replacement character '\ufffd' (due to a bug in _PyUnicodeWriter).
2015-10-02 13:13:14 +03:00
28b21e50c8
Issue #24848 : Fixed bugs in UTF-7 decoding of misformed data:
...
1. Non-ASCII bytes were accepted after shift sequence.
2. A low surrogate could be emitted in case of error in high surrogate.
2015-10-02 13:07:28 +03:00
5dbe245ef2
Issue #24483 : C implementation of functools.lru_cache() now calculates key's
...
hash only once.
2015-10-02 12:47:59 +03:00
b9d98d532c
Issue #24483 : C implementation of functools.lru_cache() now calculates key's
...
hash only once.
2015-10-02 12:47:11 +03:00
3222da26fe
Make _PyUnicode_TranslateCharmap() symbol private
...
unicodeobject.h exposes PyUnicode_TranslateCharmap() and PyUnicode_Translate().
2015-10-01 22:07:32 +02:00
01ada3996b
Issue #25267 : The UTF-8 encoder is now up to 75 times as fast for error
...
handlers: ``ignore``, ``replace``, ``surrogateescape``, ``surrogatepass``.
Patch co-written with Serhiy Storchaka.
2015-10-01 21:54:51 +02:00
d69dd8bd5e
(Merge 3.5) Issue #25182 : Fix compilation on Windows
2015-09-30 15:03:50 +02:00
ae86da9b20
(Merge 3.4) Issue #25182 : Fix compilation on Windows
2015-09-30 15:03:31 +02:00
89719e1daf
Issue #25182 : Fix compilation on Windows
...
Restore also errno value before calling PyErr_SetFromErrno().
2015-09-30 15:01:34 +02:00
85c386dee4
Issue #25182 : The stdprinter (used as sys.stderr before the io module is
...
imported at startup) now uses the backslashreplace error handler.
2015-09-30 15:51:01 +03:00
008fc77e1e
Issue #25182 : The stdprinter (used as sys.stderr before the io module is
...
imported at startup) now uses the backslashreplace error handler.
2015-09-30 15:50:32 +03:00
a59018c7ab
Issue #25182 : The stdprinter (used as sys.stderr before the io module is
...
imported at startup) now uses the backslashreplace error handler.
2015-09-30 15:46:53 +03:00
c3713e9706
Optimize ascii/latin1+surrogateescape encoders
...
Issue #25227 : Optimize ASCII and latin1 encoders with the ``surrogateescape``
error handler: the encoders are now up to 3 times as fast.
Initial patch written by Serhiy Storchaka.
2015-09-29 12:32:13 +02:00
0030cd52da
Issue #25227 : Cleanup unicode_encode_ucs1() error handler
...
* Change limit type from unsigned int to Py_UCS4, to use the same type than the
"ch" variable (an Unicode character).
* Reuse ch variable for _Py_ERROR_XMLCHARREFREPLACE
* Add some newlines for readability
2015-09-24 14:45:00 +02:00
54385b206d
Issue #24870 : revert unwanted change
...
Sorry, I pushed the patch on the UTF-8 decoder by mistake :-(
2015-09-22 10:46:52 +02:00
5ebae87628
Issue #25207 , #14626 : Fix my commit.
...
It doesn't work to use #define XXX defined(YYY)" and then "#ifdef XXX"
to check YYY.
2015-09-22 01:29:33 +02:00
6174474bea
_PyUnicodeWriter_PrepareInternal(): make the assertion more strict
2015-09-22 01:01:17 +02:00
ca9381ea01
Issue #24870 : Add _PyUnicodeWriter_PrepareKind() macro
...
Add a macro which ensures that the writer has at least the requested kind.
2015-09-22 00:58:32 +02:00
5014920cb7
Issue #24870 : Reuse the new _Py_error_handler enum
...
Factorize code with the new get_error_handler() function.
Add some empty lines for readability.
2015-09-22 00:26:54 +02:00
f96418de05
Issue #24870 : Optimize the ASCII decoder for error handlers: surrogateescape,
...
ignore and replace. Initial patch written by Naoki Inada.
The decoder is now up to 60 times as fast for these error handlers.
Add also unit tests for the ASCII decoder.
2015-09-21 23:06:27 +02:00
026977717e
Merge 3.5
2015-09-19 13:39:16 +02:00
5783fd2c58
Issue #24999 : In longobject.c, use two shifts instead of ">> 2*PyLong_SHIFT" to
...
avoid undefined behaviour when LONG_MAX type is smaller than 60 bits.
This change should fix a warning with the ICC compiler.
2015-09-19 13:39:03 +02:00
058258652a
Merge 3.5 (pytime, odict)
2015-09-18 13:55:15 +02:00
4a0d1e7c36
odictobject.c: fix compiler warning
...
PyObject_Length() returns a P_ssize_t, not an int. Use a Py_ssize_t to avoid
overflow.
2015-09-18 13:44:11 +02:00
56f6e76c68
Issue #15989 : Fixed some scarcely probable integer overflows.
...
It is very unlikely that they can occur in real code for now.
2015-09-06 21:25:30 +03:00
ba5f59089a
Issue #24912 : Prevent __class__ assignment to immutable built-in objects. (Merge 3.5 -> 3.6)
2015-09-05 15:20:57 -07:00
37fdcbc4c3
Issue #24912 : Prevent __class__ assignment to immutable built-in objects. (Merge 3.5.0 -> 3.5)
2015-09-05 15:20:08 -07:00
7d293ee97d
Issue #24912 : Prevent __class__ assignment to immutable built-in objects.
2015-09-04 20:54:07 -07:00
fa9dfd4f82
Merge 3.5 (odict)
2015-09-03 17:50:30 +02:00
ca30b02abe
Issue #24992 : Fix error handling and a race condition (related to garbage
...
collection) in collections.OrderedDict constructor.
Patch reviewed by Serhiy Storchaka.
2015-09-03 17:50:04 +02:00
99bb14bf0c
type_call() now detect bugs in type new and init
...
* Call _Py_CheckFunctionResult() to check for bugs in type
constructors (tp_new)
* Add assertions to ensure an exception was raised if tp_init failed
or that no exception was raised if tp_init succeed
Refactor also the function to have less indentation.
2015-09-03 12:16:49 +02:00
ab2aa6dc91
Fixed an incorrect comment.
2015-08-26 14:10:32 -04:00
5f35725fb8
Merge #15944 .
2015-08-08 13:38:59 +02:00
0c51595a78
Issue #15944 : memoryview: Allow arbitrary formats when casting to bytes.
...
Original patch by Martin Panter.
2015-08-08 13:38:10 +02:00
5060bc51ca
Merge from 3.5 (issue #24667 ).
2015-08-07 17:47:35 -06:00
8c7f9558eb
Issue #24667 : Resize odict in all cases that the underlying dict resizes.
2015-08-07 17:45:12 -06:00
577d2069da
Fix typo in comment.
2015-08-07 18:50:24 -04:00
4148195c45
Move the active entry multiplication to later in the hash calculation
2015-08-07 00:43:39 -07:00
b501a27ad8
Restore frozenset hash caching removed in cf707dd190a9
2015-08-06 22:15:22 -07:00
070bd62cfa
Closes #21279 : Merge with 3.5
2015-08-06 00:05:13 -05:00
d987a81d29
Issue #21279 : Merge with 3.4
2015-08-06 00:04:23 -05:00
79b98df023
Issue #21279 : Flesh out str.translate docs
...
Initial patch by Kinga Farkas, Martin Panter, and John Posner.
2015-08-05 23:54:15 -05:00
a286a51ae1
Fix comment typo
2015-08-01 11:07:11 -07:00
36c0500990
Tweak the comments
2015-08-01 10:57:42 -07:00
fbffdef47d
Issue #24762 : Speed-up frozenset_hash() and greatly beef-up the comments.
2015-08-01 09:53:00 -07:00
daffc916aa
Issue #24681 : Move the most likely test first in set_add_entry().
2015-07-31 07:58:56 -07:00
70559b5c20
Issue #24681 : Move the store of so->table to the code block where it is used.
2015-07-23 07:42:23 -04:00
c8fe04484e
Issue #23573 : Restored optimization of bytes.rfind() and bytearray.rfind()
...
for single-byte argument on Linux.
2015-07-20 22:58:29 +03:00
d92d4efe3d
Issue #23573 : Restored optimization of bytes.rfind() and bytearray.rfind()
...
for single-byte argument on Linux.
2015-07-20 22:58:02 +03:00
ff9e18a863
Issue #24583 : Consolidate previous set object updates into a single function
...
with a single entry point, named exit points at the bottom, more self-evident
refcount adjustments, and a comment describing why the pre-increment was
necessary at all.
2015-07-20 07:34:05 -04:00
482c05cbb5
Issue #24583 : Fix refcount leak.
2015-07-20 01:23:32 -04:00
061091a7c5
Issue #24583 : Fix crash when set is mutated while being updated.
2015-07-15 23:54:02 -07:00
d702044bcd
merge
2015-07-15 23:52:29 -07:00
a3626bc5bd
Issue #24583 : Fix crash when set is mutated while being updated.
2015-07-15 23:50:14 -07:00
5d2385ff6f
Neaten-up a little bit.
2015-07-08 11:52:27 -07:00
9632a7d735
Issue 24581: Revert c9782a9ac031 pending a stronger test for mutation during iteration.
2015-07-07 15:29:24 -07:00
11ce8e6c37
Minor bit of factoring-out common code.
2015-07-06 19:08:49 -07:00
3dbc11cadd
Tighten-up code in the set iterator to use an entry pointer rather than indexing.
2015-07-06 19:03:01 -07:00
ef6bd7d963
Tighten-up code in set_next() to use an entry pointer rather than indexing.
2015-07-06 08:43:37 -07:00
b48d6a63ff
Bring related functions add/contains/discard together in the code.
2015-07-05 16:27:44 -07:00
73799b181e
Change add/contains/discard calls to pass the key and hash instead of an entry struct.
2015-07-05 16:06:10 -07:00
dc28d5a198
Clean-up call patterns for add/contains/discard to better match the caller's needs.
2015-07-05 10:03:20 -07:00
191a8f0a34
merge heads
2015-07-04 22:53:45 -05:00
d34c246baf
merge 3.5 ( #24407 )
2015-07-04 19:59:50 -05:00
e54d5321cc
merge 3.4 ( #24407 )
2015-07-04 19:59:24 -05:00
2a48a6eb33
merge 3.3 ( #24407 )
2015-07-04 19:58:11 -05:00
a82f77fb00
protect against mutation of the dict during insertion ( closes #24407 )
2015-07-04 19:55:16 -05:00
ac2ef65c32
Make the unicode equality test an external function rather than in-lining it.
...
The real benefit of the unicode specialized function comes from
bypassing the overhead of PyObject_RichCompareBool() and not
from being in-lined (especially since there was almost no shared
data between the caller and callee). Also, the in-lining was
having a negative effect on code generation for the callee.
2015-07-04 16:04:44 -07:00
e186c7674c
Make sure the dummy percentage calculation won't overflow.
2015-07-04 11:28:35 -07:00
c2480dc0c4
Minor cleanup.
2015-07-04 08:46:31 -07:00
b322326f48
Minor nit: Make the style of checking error return values more consistent.
2015-07-03 23:37:16 -07:00
4897300276
Minor factoring: move redundant resize scaling logic into the resize function.
2015-07-03 20:00:03 -07:00
Serhiy Storchaka
Raymond Hettinger
Victor Stinner
Benjamin Peterson
Yury Selivanov
Martin Panter
doko@ubuntu.com
Barry Warsaw
Stefan Krah
Guido van Rossum
Guido van Rossum
Eric V. Smith
Eric Snow
Zachary Ware