Commit Graph

5751 Commits

Author SHA1 Message Date
Victor Stinner ae8b69c410 Issue #27810: Add _PyCFunction_FastCallKeywords()
Use _PyCFunction_FastCallKeywords() in ceval.c: it allows to remove a lot of
code from ceval.c which was only used to call C functions.
2016-09-09 14:07:44 -07:00
Benjamin Peterson 621b430a14 remove all usage of Py_LOCAL 2016-09-09 13:54:34 -07:00
Victor Stinner d873572095 Add _PyObject_FastCallKeywords()
Issue #27830: Add _PyObject_FastCallKeywords(): avoid the creation of a
temporary dictionary for keyword arguments.

Other changes:

* Cleanup call_function() and fast_function() (ex: rename nk to nkwargs)
* Remove now useless do_call(), replaced with _PyObject_FastCallKeywords()
2016-09-09 12:36:44 -07:00
Benjamin Peterson 6cb526e54d remove unconvincing use of Py_LOCAL 2016-09-09 12:42:51 -07:00
Eric Snow 06aed90a1f Issue #27576: Fix call order in OrderedDict.__init__(). 2016-09-09 11:59:08 -07:00
Yury Selivanov eb6364557f Issue #28003: Implement PEP 525 -- Asynchronous Generators. 2016-09-08 22:01:51 -07:00
Berker Peksag 71c01d43a7 Issue #28033: Fix typo in dictobject.c
Patch by Wesley Emeneker.
2016-09-09 03:57:23 +03:00
Eric Snow 4f29e75289 Issue #24254: Drop cls.__definition_order__. 2016-09-08 15:11:11 -07:00
Christian Heimes 07a2a1b7e5 Additional safe-guard against dereferencing NULL in reduce_newobj
_PyObject_GetNewArguments() can leave args == NULL but the __newobj_ex__
branch expects args to be not-NULL.

CID 1353201
2016-09-09 00:21:22 +02:00
Christian Heimes 884332b45a Add NULL check for gen->gi_code in gen_send_ex()
_PyGen_Finalize() checks that gen->gi_code is not NULL before it
accesses the flags of the code object. This means that the flag
could be NULL.

It passes down the generatore to gen_close() and gen_send_ex().
gen_send_ex() did not check for gen->gi_code != NULL.

CID 1297900
2016-09-09 00:20:13 +02:00
Martin Panter 0be894b2f6 Issue #27895: Spelling fixes (Contributed by Ville Skyttä). 2016-09-07 12:03:06 +00:00
Benjamin Peterson b88db8745b supress coroutine warning when an exception is pending (#27968) 2016-09-07 08:46:59 -07:00
Benjamin Peterson 33d2a492d0 promote some shifts to unsigned, so as not to invoke undefined behavior 2016-09-06 20:40:04 -07:00
Benjamin Peterson 5a7d923e75 make sure to not call memcpy with a NULL second argument 2016-09-06 17:58:25 -07:00
Benjamin Peterson bdddb11b0e clear out f_gen during generator finalization (closes #27812)
Patch from Armin Rigo.
2016-09-05 10:39:57 -07:00
Benjamin Peterson 2f40ed4b94 do not allow _PyGen_Finalize to fail (closes #27811)
Patch from Armin Rigo.
2016-09-05 10:14:54 -07:00
Benjamin Peterson 3c569295d6 do not worry about 64-bit dict sizes on 32-bit platforms 2016-09-08 13:16:41 -07:00
Victor Stinner 3b6a6b4215 Add a new private version to the builtin dict type
Issue #26058: Add a new private version to the builtin dict type, incremented
at each dictionary creation and at each dictionary change.

Implementation of the PEP 509.
2016-09-08 12:51:24 -07:00
R David Murray 110b6fecbb #27364: Deprecate invalid escape strings in str/byutes.
Patch by Emanuel Barry, reviewed by Serhiy Storchaka and Martin Panter.
2016-09-08 15:34:08 -04:00
Benjamin Peterson 186122ead2 access dk_indices through a union 2016-09-08 12:20:12 -07:00
Victor Stinner a4348cc1be Add documentation to the dict implementation
Issue #27350.
2016-09-08 12:01:25 -07:00
Victor Stinner 58f7c5a955 Reindeint DK_xxx macros
Issue #27350.
2016-09-08 11:37:36 -07:00
Victor Stinner 208857ec33 dk_get_index/dk_set_index uses a type indices variable
Issue #27350.
2016-09-08 11:35:46 -07:00
Benjamin Peterson 82cce4c5ef use static inline instead of Py_LOCAL_INLINE 2016-09-08 11:56:06 -07:00
Victor Stinner dee6e252cc Split lookdict_unicode_nodummy() assertion to debug
Issue #27350.
2016-09-08 11:16:07 -07:00
Benjamin Peterson ee178e6d6e fix spelling 2016-09-08 11:08:30 -07:00
Victor Stinner 71211e3386 Add assertions to dk_set_index()
Issue #27350.
2016-09-08 10:52:46 -07:00
Steve Dower cc16be85c0 Issue #27781: Change file system encoding on Windows to UTF-8 (PEP 529) 2016-09-08 10:35:16 -07:00
Benjamin Peterson 003f059b3a link to canonical blogspot 2016-09-08 10:14:31 -07:00
Benjamin Peterson 7322225a57 use native inline instead of Py_LOCAL_INLINE 2016-09-08 09:58:47 -07:00
Benjamin Peterson f0acae2d3c rearrange to make gcc happy 2016-09-08 09:50:08 -07:00
Victor Stinner 98ee9d5b73 Add Py_MEMBER_SIZE macro
Issue #27350: use Py_MEMBER_SIZE() macro to get the size of
PyDictKeyEntry.dk_indices, rather than hardcoding 8.
2016-09-08 09:33:56 -07:00
Victor Stinner 742da040db Implement compact dict
Issue #27350: `dict` implementation is changed like PyPy. It is more compact
and preserves insertion order.

_PyDict_Dummy() function has been removed.

Disable test_gdb: python-gdb.py is not updated yet to the new structure of
compact dictionaries (issue #28023).

Patch written by INADA Naoki.
2016-09-07 17:40:12 -07:00
Benjamin Peterson 47ff0734b8 more PY_LONG_LONG to long long 2016-09-08 09:15:54 -07:00
Eric Snow c943265ba5 Issue #15767: Add ModuleNotFoundError. 2016-09-07 15:42:32 -07:00
Benjamin Peterson 2e7c5e9c11 replace some Py_LOCAL_INLINE with the inline keyword 2016-09-07 15:33:32 -07:00
Brett Cannon d0600ed524 Make PyCodeObject.co_extra even more private to force users through the proper API. 2016-09-07 14:30:39 -07:00
Benjamin Peterson d39206a78c use c++ style comments 2016-09-07 14:12:36 -07:00
Brett Cannon 3788b85628 Change error return value to be more consistent with the rest of Python 2016-09-07 12:51:08 -07:00
Benjamin Peterson 1bf494b0ec use a the bool type for a boolean variable 2016-09-07 11:28:35 -07:00
Brett Cannon 5c4de2863b Add the co_extra field and accompanying APIs to code objects.
This completes PEP 523.
2016-09-07 11:16:41 -07:00
Benjamin Peterson a9296e7f3b require C99 bool 2016-09-07 11:06:17 -07:00
Benjamin Peterson 2f8bfef158 replace PY_SIZE_MAX with SIZE_MAX 2016-09-07 09:26:18 -07:00
Benjamin Peterson 9d2617bec0 merge 3.5 (#27968) 2016-09-07 08:47:18 -07:00
Benjamin Peterson 4b9abf3a27 merge 3.5 2016-09-06 20:42:17 -07:00
Brett Cannon a571120410 Issue #27182: Add support for path-like objects to PyUnicode_FSDecoder(). 2016-09-06 19:36:01 -07:00
Benjamin Peterson 05448a62ae merge 3.5 2016-09-06 17:58:44 -07:00
Victor Stinner 62ec3317d2 Optimize unicode_escape and raw_unicode_escape
Issue #16334. Patch written by Serhiy Storchaka.
2016-09-06 17:04:34 -07:00
Victor Stinner 2740e46089 _PyUnicodeWriter: assert that max character <= MAX_UNICODE 2016-09-06 16:58:36 -07:00
Brett Cannon ec6ce879c7 Issue #26027: Support path-like objects in PyUnicode-FSConverter().
This is to add support for os.exec*() and os.spawn*() functions. Part
of PEP 519.
2016-09-06 15:50:29 -07:00
Benjamin Peterson ca47063998 replace Py_(u)intptr_t with the c99 standard types 2016-09-06 13:47:26 -07:00
Benjamin Peterson 9b3d77052f replace Python aliases for standard integer types with the standard integer types (#17884) 2016-09-06 13:24:00 -07:00
Serhiy Storchaka ea525a2d1a Issue #27078: Added BUILD_STRING opcode. Optimized f-strings evaluation. 2016-09-06 22:07:53 +03:00
Benjamin Peterson d953f8e10b remove some silly defined() tests 2016-09-06 10:51:19 -07:00
Benjamin Peterson af580dff4a replace PY_LONG_LONG with long long 2016-09-06 10:46:49 -07:00
Victor Stinner 3466bde1cc Avoid calling functions with an empty string as format string
Directly pass NULL rather than an empty string.
2016-09-05 18:16:01 -07:00
Victor Stinner ad8c83ad6b Avoid inefficient way to call functions without argument
Don't pass "()" format to PyObject_CallXXX() to call a function without
argument: pass NULL as the format string instead. It avoids to have to parse a
string to produce 0 argument.
2016-09-05 17:53:15 -07:00
Benjamin Peterson ed4aa83ff7 require a long long data type (closes #27961) 2016-09-05 17:44:18 -07:00
Victor Stinner 942889aae2 Issue #27938: Add a fast-path for us-ascii encoding
Other changes:

* Rewrite _Py_normalize_encoding() as a C implementation of
  encodings.normalize_encoding(). For example, " utf-8 " is now normalized to
  "utf_8". So the fast path is now used for more name variants of the same
  encoding.
* Avoid strcpy() when encoding is NULL: call directly the UTF-8 codec
2016-09-05 15:40:10 -07:00
Eric Snow 92a6c170e6 Issue #24254: Preserve class attribute definition order. 2016-09-05 14:50:11 -07:00
Benjamin Peterson 55f445282d remove memory indirections in dict_traverse (closes #27956) 2016-09-05 12:12:59 -07:00
Victor Stinner 2250c4cb81 Issue #27830: Remove unused _PyStack_AsDict()
I forgot to remove this function, I made a mistake in my revert.
2016-09-05 11:43:18 -07:00
Benjamin Peterson 7b24b27302 merge 3.5 (#27812) 2016-09-05 10:40:34 -07:00
Benjamin Peterson 98020812e6 merge 3.5 (closes #27811) 2016-09-05 10:16:31 -07:00
Mark Dickinson 7c4e409d07 Issue #11734: Add support for IEEE 754 half-precision floats to the struct module. Original patch by Eli Stevens. 2016-09-03 17:21:29 +01:00
Victor Stinner 1a05d6c04d PEP 7 style for if/else in C
Add also a newline for readability in normalize_encoding().
2016-09-02 12:12:23 +02:00
Raymond Hettinger 15f44ab043 Issue #27895: Spelling fixes (Contributed by Ville Skyttä). 2016-08-30 10:47:49 -07:00
Mark Dickinson 82a95277b8 Issue #27870: A left shift of zero by a large integer no longer attempts to allocate large amounts of memory. 2016-08-29 19:27:06 +01:00
Mark Dickinson 4e1de16f88 Issue #25402: in int-to-decimal-string conversion, reduce intermediate storage requirements and relax restriction on converting large integers. Patch by Serhiy Storchaka. 2016-08-29 17:26:43 +01:00
Mark Dickinson 583c6e860c Issue #27214: Fix potential bug and remove useless optimization in long_invert. Thanks Oren Milman. 2016-08-29 16:40:29 +01:00
Martin Panter 1b6c6da85d Issue #27506: Support bytes/bytearray.translate() delete as keyword argument
Patch by Xiang Zhang.
2016-08-27 08:35:02 +00:00
Raymond Hettinger f74c33ad5c Merge 2016-08-25 21:12:16 -07:00
Raymond Hettinger 7ea386e56e Issue 19504: Change "customise" to "customize" American spelling. 2016-08-25 21:11:50 -07:00
Victor Stinner e90bdb19f2 Issue #27830: Revert, remove _PyFunction_FastCallKeywords() 2016-08-25 23:26:50 +02:00
Victor Stinner 3f1057a4b6 method_call() and slot_tp_new() now uses fast call
Issue #27841: Add _PyObject_Call_Prepend() helper function to prepend an
argument to existing arguments to call a function. This helper uses fast calls.

Modify method_call() and slot_tp_new() to use _PyObject_Call_Prepend().
2016-08-25 01:04:14 +02:00
Victor Stinner f7507dd3e8 Issue #27830: Fix _PyObject_FastCallKeywords()
Pass stack, not unrelated and uninitialized args!
2016-08-25 01:00:31 +02:00
Victor Stinner 53cb489ed0 _PyObject_FastCallDict(): avoid _Py_CheckFunctionResult()
_PyObject_FastCallDict() only requires _Py_CheckFunctionResult() for the
slow-path. Other cases already check for the result.
2016-08-25 00:39:34 +02:00
Victor Stinner 577e1f8cb4 Add _PyObject_FastCallKeywords()
Issue #27830: Similar to _PyObject_FastCallDict(), but keyword arguments are
also passed in the same C array than positional arguments, rather than being
passed as a Python dict.
2016-08-25 00:29:32 +02:00
Victor Stinner 74319ae219 Use Py_ssize_t type for number of arguments
Issue #27848: use Py_ssize_t rather than C int for the number of function
positional and keyword arguments.
2016-08-25 00:04:09 +02:00
Victor Stinner dd69564c38 PyObject_CallMethodObjArgs() now uses fast call
Issue #27809:

* PyObject_CallMethodObjArgs(), _PyObject_CallMethodIdObjArgs() and
  PyObject_CallFunctionObjArgs() now use fast call to avoid the creation of a
  temporary tuple
* Rename objargs_mktuple() to objargs_mkstack()
* objargs_mkstack() now stores objects in a C array using borrowed references,
  instead of storing arguments into a tuple

objargs_mkstack() uses a small buffer allocated on the C stack for 5 arguments
or less, or allocates a buffer in the heap memory.

Note: this change is different than the change 0e4f26083bbb, I fixed the test
to decide if the small stack can be used or not. sizeof(PyObject**) was also
replaced with sizeof(stack[0]) since the sizeof() was wrong (but gave the same
result).
2016-08-24 01:14:54 +02:00
Victor Stinner d48af09043 Backed out changeset 0e4f26083bbb (PyObject_CallMethodObjArgs) 2016-08-24 00:59:40 +02:00
Victor Stinner e62b852605 PyObject_CallMethodObjArgs() now uses fast call
Issue #27809:

* PyObject_CallMethodObjArgs(), _PyObject_CallMethodIdObjArgs() and
  PyObject_CallFunctionObjArgs() now use fast call to avoid the creation of a
  temporary tuple
* Rename objargs_mktuple() to objargs_mkstack()
* objargs_mkstack() now stores objects in a C array using borrowed references,
  instead of storing arguments into a tuple

objargs_mkstack() uses a small buffer allocated on the C stack for 5 arguments
or less, or allocates a buffer in the heap memory.
2016-08-24 00:01:56 +02:00
Victor Stinner 463b86a881 Issue #27809: Use _PyObject_FastCallDict()
Modify:

* init_subclass()
* builtin___build_class__()

Fix also a bug in init_subclass(): check for super() failure.
2016-08-22 23:33:13 +02:00
Victor Stinner 2990fa11bc Issue #27809: Use _PyObject_FastCallDict()
Modify:

* builtin_sorted()
* classmethoddescr_call()
* methoddescr_call()
* wrapperdescr_call()
2016-08-22 23:21:55 +02:00
Victor Stinner b900939186 _PyFunction_FastCallDict() supports keyword args
Issue #27809:

* Rename _PyFunction_FastCall() to _PyFunction_FastCallDict()
* Rename _PyCFunction_FastCall() to _PyCFunction_FastCallDict()
*  _PyFunction_FastCallDict() now supports keyword arguments
2016-08-22 23:15:44 +02:00
Victor Stinner 559bb6a713 Rename _PyObject_FastCall() to _PyObject_FastCallDict()
Issue #27809:

* Rename _PyObject_FastCall() function to _PyObject_FastCallDict()
* Add _PyObject_FastCall(), _PyObject_CallNoArg() and _PyObject_CallArg1()
  macros calling _PyObject_FastCallDict()
2016-08-22 22:48:54 +02:00
Mark Dickinson b820d7f631 Issue #27792: force int return type for modulo operations involving bools. 2016-08-22 12:24:46 +01:00
Serhiy Storchaka 54cd1969ea Issue #26984: int() now always returns an instance of exact int. 2016-08-21 20:03:08 +03:00
Mark Dickinson 1dc3c898a8 Untabify Objects/longobject.c. 2016-08-21 10:33:36 +01:00
Mark Dickinson dc590a4cc3 Issue #25604: Fix minor bug in integer true division, which could
have caused off-by-one-ulp results on certain platforms.
2016-08-21 10:23:23 +01:00
Mark Dickinson 5d13238f6e Issue #27662: don't use PY_SIZE_MAX for overflow checking in List_New. Patch by Xiang Zhang. 2016-08-21 08:55:15 +01:00
Nick Coghlan cbcd221de4 Merge #27782 fix from 3.5 2016-08-21 17:43:58 +10:00
Nick Coghlan 8682f578c1 Issue #27782: Fix m_methods handling in multiphase init
Multi-phase extension module import now correctly allows the
``m_methods`` field to be used to add module level functions
to instances of non-module types returned from ``Py_create_mod``.

Patch by Xiang Zhang.
2016-08-21 17:41:56 +10:00
Serhiy Storchaka 3766572a48 Issue #27692: Removed unnecessary NULL checks in exceptions.c.
Patch by Xiang Zhang.
2016-08-20 21:22:03 +03:00
Victor Stinner 253021dd94 Issue #27366: Fix init_subclass()
Handle PyTuple_New(0) failure.
2016-08-20 02:37:41 +02:00
Victor Stinner 9def0901e2 PyFile_WriteObject() now uses fast call
Issue #27128: PyFile_WriteObject() now calls _PyObject_FastCall() to avoid the
creation of a temporary tuple.
2016-08-20 00:44:04 +02:00
Victor Stinner 99ee9c70a7 calliter_iternext() now uses fast call
Issue #27128: calliter_iternext() now calls _PyObject_FastCall() to avoid a
temporary empty tuple.

Cleanup also the code to reduce the indentation level.
2016-08-19 18:47:10 +02:00
Victor Stinner 6911267615 slot_tp_iter() now uses fast call
Issue #27128: slot_tp_iter() now calls _PyObject_FastCall() to avoid a
temporary empty tuple.
2016-08-19 18:41:02 +02:00
Victor Stinner 20a3007a8d slot_nb_bool() now uses fast call
Issue #27128: slot_nb_bool() now calls _PyObject_FastCall() to avoid a
temporary empty tuple to call the slot function.
2016-08-19 18:28:25 +02:00
Victor Stinner a12eec48b6 Issue #27128: Cleanup slot_nb_bool()
Use an error label to reduce the level of indentation.
2016-08-19 18:26:05 +02:00