Commit Graph

6178 Commits

Author SHA1 Message Date
Raymond Hettinger e1af6964b4 Remove unnecessary variables.
* so->used never gets changed during a resize
* so->filled only changes when dummies are present and being eliminated
2017-02-02 08:24:48 -08:00
Serhiy Storchaka 196a7bc1ea Issue #29421: Make int.to_bytes() and int.from_bytes() slightly faster
(10-20% for small integers).
2017-02-02 16:54:45 +02:00
Serhiy Storchaka 495e8808d7 Issue #20185: Converted the int class to Argument Clinic.
Based on patch by Vajrasky Kok.
2017-02-01 23:12:20 +02:00
INADA Naoki 35a96a25ee Issue #29383: reduce temporary interned unicode
add_methods(), add_members(), and add_getset() used PyDict_SetItemString()
to register descriptor to the type's dict.
So descr_new() and PyDict_SetItemString() creates interned unicode from same
C string.

This patch takes interned unicode from descriptor, and use PyDict_SetItem()
instead of PyDict_SetItemString().

python_startup_no_site:
  default: Median +- std dev: 12.7 ms +- 0.1 ms
  patched: Median +- std dev: 12.5 ms +- 0.1 ms
2017-01-28 16:35:44 +09:00
Victor Stinner bda5a2be37 Issue #29358: Add postcondition checks on types 2017-01-25 23:33:27 +01:00
Serhiy Storchaka 6e08baf12d Issue #27867: Function PySlice_GetIndicesEx() is deprecated and replaced with
a macro if Py_LIMITED_API is not set or set to the value between 0x03050400
and 0x03060000 (not including) or 0x03060100 or higher.  Added functions
PySlice_Unpack() and PySlice_AdjustIndices().
2017-01-25 13:27:44 +02:00
Serhiy Storchaka 512c57cb72 Issue #27867: Function PySlice_GetIndicesEx() is replaced with a macro if
Py_LIMITED_API is not set or set to the value between 0x03050400
and 0x03060000 (not including) or 0x03060100 or higher.
2017-01-25 13:25:52 +02:00
Serhiy Storchaka b2a5be0763 Issue #27867: Function PySlice_GetIndicesEx() is replaced with a macro if
Py_LIMITED_API is not set or set to the value between 0x03050400
and 0x03060000 (not including) or 0x03060100 or higher.
2017-01-25 13:23:05 +02:00
Serhiy Storchaka 78d9e58f20 Issues #29311, #29289: Fixed and improved docstrings for dict and OrderedDict
methods.
2017-01-25 00:30:04 +02:00
Serhiy Storchaka 067cbd02bf Issue #29337: Fixed possible BytesWarning when compare the code objects.
Warnings could be emitted at compile time.
2017-01-24 20:54:07 +02:00
Serhiy Storchaka 4102d25b7e Issue #29337: Fixed possible BytesWarning when compare the code objects.
Warnings could be emitted at compile time.
2017-01-24 20:52:43 +02:00
Serhiy Storchaka 713640c4c9 Issue #29337: Fixed possible BytesWarning when compare the code objects.
Warnings could be emitted at compile time.
2017-01-24 20:49:26 +02:00
Victor Stinner c3858bd7c6 Issue #29360: _PyStack_AsDict() doesn't check kwnames
Remove two assertions which can fail on legit code. Keyword arguments are
checked later with better tests and raise a regular (TypeError) exception.
2017-01-24 15:05:30 +01:00
Martin Panter 91a8866dc1 Fix grammar in doc string, RST markup 2017-01-24 00:30:06 +00:00
Serhiy Storchaka d1302c0154 Issue #28999: Use Py_RETURN_NONE, Py_RETURN_TRUE and Py_RETURN_FALSE wherever
possible but Coccinelle couldn't find opportunity.
2017-01-23 10:23:58 +02:00
Serhiy Storchaka 228b12edcc Issue #28999: Use Py_RETURN_NONE, Py_RETURN_TRUE and Py_RETURN_FALSE wherever
possible.  Patch is writen with Coccinelle.
2017-01-23 09:47:21 +02:00
Serhiy Storchaka 2a404b63d4 Issue #28769: The result of PyUnicode_AsUTF8AndSize() and PyUnicode_AsUTF8()
is now of type "const char *" rather of "char *".
2017-01-22 23:07:07 +02:00
Serhiy Storchaka 7cf8bebb07 Issue #29331: Simplified argument parsing in sorted() and list.sort(). 2017-01-21 23:05:00 +02:00
Serhiy Storchaka a70eaf21e3 Issue #29289: Argument Clinic generates reasonable name for the parameter "default". 2017-01-19 19:38:13 +02:00
Serhiy Storchaka 48088ee9ba Issue #29311: Argument Clinic generates reasonable name for the parameter "default". 2017-01-19 19:00:30 +02:00
Serhiy Storchaka 41baebd8b9 Issue #20186: Converted builtins enumerate() and reversed() to Argument Clinic.
Patch by Tal Einat.
2017-01-19 18:48:17 +02:00
Victor Stinner 91f0d4aa2f Add a note explaining why dict_update() doesn't use METH_FASTCALL
Issue #29312.
2017-01-19 12:45:06 +01:00
Victor Stinner 7dc6a5f50a dict.get() and dict.setdefault() now use AC
Issue #29311: dict.get() and dict.setdefault() methods now use Argument Clinic
to parse arguments. Their calling convention changes from METH_VARARGS to
METH_FASTCALL which avoids the creation of a temporary tuple.

The signature of docstrings is also enhanced. For example,
    get(...)
becomes:
    get(self, key, default=None, /)
2017-01-19 12:37:13 +01:00
Victor Stinner 06d34393c2 _PyStack_AsDict() now checks kwnames != NULL
Issue #29259.
2017-01-18 18:06:32 +01:00
Victor Stinner 0a2e46835d Cleanup _PyMethodDef_RawFastCallDict()
Issue #29259: use a different case for METH_VARARGS and
METH_VARARGS|METH_KEYWORDS to avoid testing again flags to decide if keywords
should be checked or not.
2017-01-18 14:16:57 +01:00
Victor Stinner a8cb515a29 Rephrase !PyErr_Occurred() comment: may=>can
Issue #29259.
2017-01-18 14:12:51 +01:00
Victor Stinner 98ccba8344 _PyObject_FastCallKeywords() now checks !PyErr_Occurred()
Issue #29259. All other functions calling functions start with the similar
assertion.
2017-01-18 14:06:38 +01:00
Victor Stinner c89ef828cf PyCFunction_Call() now calls _PyCFunction_FastCallDict()
Issue #29259. We had 3 versions of similar code:

* PyCFunction_Call()
* _PyCFunction_FastCallDict()
* _PyCFunction_FastCallKeywords()

PyCFunction_Call() now calls _PyCFunction_FastCallDict() to factorize the code.
2017-01-18 14:04:37 +01:00
Victor Stinner 250e4b0063 Fix _PyMethodDef_RawFastCallDict() argument parsing
Issue #29259:

* Move also the !PyErr_Occurred() assertion to the top, similar to
  other functions.
* Fix also comment/error messages: the function was renamed to
  _PyMethodDef_RawFastCallDict()
2017-01-18 14:01:12 +01:00
Victor Stinner 2a1b676d1f _PyObject_FastCallKeywords() now checks the result
Issue ##27830, Issue #29259.
2017-01-18 11:27:22 +01:00
Victor Stinner c52572319c Optimize methoddescr_call(): avoid temporary PyCFunction
Issue #29259, #29263. methoddescr_call() creates a PyCFunction object, call it
and the destroy it. Add a new _PyMethodDef_RawFastCallDict() method to avoid
the temporary PyCFunction object.
2017-01-18 10:38:09 +01:00
Victor Stinner 35ecebe165 Remove unused func parameter of _PyStack_UnpackDict()
Issue #29259.
2017-01-18 10:31:46 +01:00
Victor Stinner b05cbac052 Convert some OrderedDict methods to Argument Clinic
Issue #29289. Convert methods:

* fromkeys() class method
* setdefault()
* popitem()
* move_to_end()
2017-01-17 03:46:13 +01:00
Victor Stinner 0c4a828cad Run Argument Clinic: METH_VARARGS=>METH_FASTCALL
Issue #29286. Run Argument Clinic to get the new faster METH_FASTCALL calling
convention for functions using "boring" positional arguments.

Manually fix _elementtree: _elementtree_XMLParser_doctype() must remain
consistent with the clinic code.
2017-01-17 02:21:47 +01:00
Victor Stinner 259f0e4437 Run Argument Clinic: METH_VARARGS=>METH_FASTCALL
Issue #29286. Run Argument Clinic to get the new faster METH_FASTCALL calling
convention for functions using only positional arguments.
2017-01-17 01:35:17 +01:00
Victor Stinner 998c20962c _PyStack_UnpackDict() now returns -1 on error
Issue #29286. Change _PyStack_UnpackDict() prototype to be able to notify of
failure when args is NULL.
2017-01-17 01:57:29 +01:00
Victor Stinner 3e1fad6913 Rename _PyArg_ParseStack to _PyArg_ParseStackAndKeywords
Issue #29286.
2017-01-17 01:29:01 +01:00
Victor Stinner d526cfe546 type_prepare() now uses fast call (METH_FASTCALL) 2017-01-16 23:52:00 +01:00
Victor Stinner 69de71b255 Add _PyStack_AsTupleSlice() helper 2017-01-16 23:50:53 +01:00
Victor Stinner 7fc252adfb Optimize _PyCFunction_FastCallKeywords()
Issue #29259: Write fast path in _PyCFunction_FastCallKeywords() for
METH_FASTCALL, avoid the creation of a temporary dictionary for keyword
arguments.

Cleanup also _PyCFunction_FastCallDict():

* Don't dereference func before checking that it's not NULL
* Move code to raise the "no keyword argument" exception into a new
  no_keyword_error label.

Update python-gdb.py for the change.
2017-01-16 17:18:53 +01:00
INADA Naoki 15f94596b6 Issue #20180: forgot to update AC output. 2017-01-16 21:49:13 +09:00
INADA Naoki 3ae2056512 Issue #20180: convert unicode methods to AC. 2017-01-16 20:41:20 +09:00
Martin Panter 446a498a63 Merge doc fixes from 3.6 2017-01-14 09:56:00 +00:00
Martin Panter 4659ddc433 Merge doc fixes from 3.5 2017-01-14 09:54:57 +00:00
Martin Panter 536d70ed33 Fix grammar, typos and markup in documentation and code comments
* Indent versionchanged at method level, not class level
* Mark up ``--help`` to avoid generating an en dash
* Use forward slash in Unix command line with a dollar sign ($) prompt
2017-01-14 08:23:08 +00:00
Martin Panter 94b39ceb73 Issue #1621: Overflow should not be possible in listextend() 2017-01-14 06:30:37 +00:00
Serhiy Storchaka 617c7753ce Issue #28969: Fixed race condition in C implementation of functools.lru_cache.
KeyError could be raised when cached function with full cache was
simultaneously called from differen threads with the same uncached arguments.
2017-01-12 19:42:20 +02:00
Serhiy Storchaka 42e1ea9a10 Issue #28969: Fixed race condition in C implementation of functools.lru_cache.
KeyError could be raised when cached function with full cache was
simultaneously called from differen threads with the same uncached arguments.
2017-01-12 19:12:21 +02:00
Serhiy Storchaka 67796521dd Issue #28969: Fixed race condition in C implementation of functools.lru_cache.
KeyError could be raised when cached function with full cache was
simultaneously called from differen threads with the same uncached arguments.
2017-01-12 18:34:33 +02:00
Victor Stinner b915bc354e Disable _PyStack_AsTuple() inlining
Issue #29234: Inlining _PyStack_AsTuple() into callers increases their stack
consumption, Disable inlining to optimize the stack consumption.

Add _Py_NO_INLINE: use __attribute__((noinline)) of GCC and Clang.

It reduces the stack consumption, bytes per call, before => after:

test_python_call: 1040 => 976 (-64 B)
test_python_getitem: 976 => 912 (-64 B)
test_python_iterator: 1120 => 1056 (-64 B)

=> total: 3136 => 2944 (- 192 B)
2017-01-11 01:07:03 +01:00
Victor Stinner 434723f94c call_method() now uses _PyObject_FastCall()
Issue #29233: Replace the inefficient _PyObject_VaCallFunctionObjArgs() with
_PyObject_FastCall() in call_method() and call_maybe().

Only a few functions call call_method() and call it with a fixed number of
arguments. Avoid the complex and expensive _PyObject_VaCallFunctionObjArgs()
function, replace it with an array allocated on the stack with the exact number
of argumlents.

It reduces the stack consumption, bytes per call, before => after:

test_python_call: 1168 => 1152 (-16 B)
test_python_getitem: 1344 => 1008 (-336 B)
test_python_iterator: 1568 => 1232 (-336 B)

Remove the _PyObject_VaCallFunctionObjArgs() function which became useless.
Rename it to object_vacall() and make it private.
2017-01-11 00:07:40 +01:00
Xiang Zhang 7a4da324dc Issue #29145: Merge 3.6. 2017-01-10 10:56:38 +08:00
Xiang Zhang 95403d74d7 Issue #29145: Merge 3.5. 2017-01-10 10:54:19 +08:00
Xiang Zhang b0541f4cdf Issue #29145: Fix overflow checks in str.replace() and str.join().
Based on patch by Martin Panter.
2017-01-10 10:52:00 +08:00
INADA Naoki 7ed28a8914 Merge 3.6 2017-01-06 17:44:43 +09:00
INADA Naoki a634e23209 Issue #29159: Fix regression in bytes(x) when x.__index__() raises Exception. 2017-01-06 17:32:01 +09:00
Victor Stinner 6f7c0ae46d Issue #28839: Optimize function_call()
function_call() now simply calls _PyFunction_FastCallDict().

_PyFunction_FastCallDict() is more efficient: it contains fast paths for the
common case (optimized code object and no keyword argument).
2017-01-03 01:58:17 +01:00
Stefan Krah 089a062da4 Merge 3.6. 2016-12-30 12:24:59 +01:00
Stefan Krah a50006c977 Merge 3.5. 2016-12-30 12:24:23 +01:00
Stefan Krah 195319e6bb Issue #29111: Fix memoryview signature. 2016-12-30 12:23:35 +01:00
Serhiy Storchaka c9ad8b7a23 Issue #29073: bytearray formatting no longer truncates on first null byte. 2016-12-28 09:54:22 +02:00
Antoine Pitrou c06ae208eb Issue #28427: old keys should not remove new values from
WeakValueDictionary when collecting from another thread.
2016-12-27 14:34:54 +01:00
Antoine Pitrou d741ed492f Issue #28427: old keys should not remove new values from
WeakValueDictionary when collecting from another thread.
2016-12-27 14:23:43 +01:00
Antoine Pitrou e10ca3a0fe Issue #28427: old keys should not remove new values from
WeakValueDictionary when collecting from another thread.
2016-12-27 14:19:20 +01:00
INADA Naoki 5a625d0aa6 Issue #29049: Call _PyObject_GC_TRACK() lazily when calling Python function.
Calling function is up to 5% faster.
2016-12-24 20:19:08 +09:00
Xiang Zhang 62497d52d9 Issue #29044: Merge 3.6. 2016-12-22 15:31:55 +08:00
Xiang Zhang 437a5d2c25 Issue #29044: Merge 3.5. 2016-12-22 15:31:22 +08:00
Xiang Zhang ea1cf87030 Issue #29044: Fix a use-after-free in string '%c' formatter. 2016-12-22 15:30:47 +08:00
Xiang Zhang b211068f5c Issue #28822: Adjust indices handling of PyUnicode_FindChar(). 2016-12-20 22:52:33 +08:00
INADA Naoki 6165d55f13 Issue #28147: Fix a memory leak in split-table dictionaries
setattr() must not convert combined table into split table.
2016-12-20 09:54:24 +09:00
Serhiy Storchaka dd1da7f74a Issue #28927: bytes.fromhex() and bytearray.fromhex() now ignore all ASCII
whitespace, not only spaces.  Patch by Robert Xiao.
2016-12-19 18:51:37 +02:00
Serhiy Storchaka 9d16b616d5 Issue #29000: Fixed bytes formatting of octals with zero padding in alternate
form.
2016-12-17 22:15:10 +02:00
Serhiy Storchaka 025f8953f1 Issue #29000: Fixed bytes formatting of octals with zero padding in alternate
form.
2016-12-17 22:13:05 +02:00
Serhiy Storchaka b1a1619bf0 Issue #29000: Fixed bytes formatting of octals with zero padding in alternate
form.
2016-12-17 21:48:03 +02:00
Serhiy Storchaka 5bb8b9134b Issue #18896: Python function can now have more than 255 parameters.
collections.namedtuple() now supports tuples with more than 255 elements.
2016-12-16 19:19:02 +02:00
Serhiy Storchaka 5ab81d787f Issue #28959: Added private macro PyDict_GET_SIZE for retrieving the size of dict. 2016-12-16 16:18:57 +02:00
Xavier de Gaye 31eaf49ed9 Merge 3.6. 2016-12-15 21:01:52 +01:00
Xavier de Gaye 76febd0792 Issue #26919: On Android, operating system data is now always encoded/decoded
to/from UTF-8, instead of the locale encoding to avoid inconsistencies with
os.fsencode() and os.fsdecode() which are already using UTF-8.
2016-12-15 20:59:58 +01:00
Victor Stinner 5cc70c9935 Merge 3.6 2016-12-15 17:23:24 +01:00
Victor Stinner 3d3f264849 Fix a memory leak in split-table dictionaries
Issue #28147: Fix a memory leak in split-table dictionaries: setattr() must not
convert combined table into split table.

Patch written by INADA Naoki.
2016-12-15 17:21:23 +01:00
Victor Stinner 944dbc6431 Use _PyDict_NewPresized() in _PyStack_AsDict()
Issue #27810.
2016-12-15 16:59:40 +01:00
Victor Stinner bc08ab4598 Add _PY_FASTCALL_SMALL_STACK constant
Issue #28870: Add a new _PY_FASTCALL_SMALL_STACK constant, size of "small
stacks" allocated on the C stack to pass positional arguments to
_PyObject_FastCall().

_PyObject_Call_Prepend() now uses a small stack of 5 arguments (40 bytes)
instead of 8 (64 bytes), since it is modified to use _PY_FASTCALL_SMALL_STACK.
2016-12-15 12:40:53 +01:00
Victor Stinner d1e35dd9ee Fix _PyObject_CallFunctionVa(), use the small stack
Issue #28915. Oops, I disabled the small stack to test both code paths. It's
now fixed.
2016-12-15 12:36:50 +01:00
Serhiy Storchaka bdfc5ff17e Merge from 3.6. 2016-12-14 19:56:53 +02:00
Serhiy Storchaka 386072ebe0 Merge from 3.6. 2016-12-14 19:54:38 +02:00
Serhiy Storchaka 49010ee323 Revert changeset 1f31bf3f76f5 (issue5322) except tests. 2016-12-14 19:52:17 +02:00
Yury Selivanov f2392133eb Issue #26110: Add LOAD_METHOD/CALL_METHOD opcodes.
Special thanks to INADA Naoki for pushing the patch through
the last mile, Serhiy Storchaka for reviewing the code, and to
Victor Stinner for suggesting the idea (originally implemented
in the PyPy project).
2016-12-13 19:03:51 -05:00
Martin Panter 9bcaa37e1c Issue #28820: Merge typo fixes from 3.6 2016-12-10 05:39:12 +00:00
Martin Panter 186b204997 Fix typos in comment and documentation 2016-12-10 05:32:55 +00:00
Victor Stinner 59a73276e9 Backed out changeset 99c34e47348b
The change broke test_gdb.
2016-12-09 18:51:13 +01:00
Victor Stinner 0ca246c5e7 Inline PyEval_EvalFrameEx() in callers
The PEP 523 modified PyEval_EvalFrameEx(): it's now an indirection to
interp->eval_frame().

Inline the call in performance critical code. Leave PyEval_EvalFrame()
unchanged, this function is only kept for backward compatibility.
2016-12-09 17:12:17 +01:00
Victor Stinner 14e6d09547 Remove useless variable initialization
Don't initialize variables which are not used before they are assigned.
2016-12-09 17:08:59 +01:00
Victor Stinner 5abaa2b139 Use PyObject_CallFunctionObjArgs()
Issue #28915: Replace PyObject_CallFunction() with
PyObject_CallFunctionObjArgs() when the format string was only made of "O"
formats, PyObject* arguments.

PyObject_CallFunctionObjArgs() avoids the creation of a temporary tuple and
doesn't have to parse a format string.
2016-12-09 16:22:32 +01:00
Victor Stinner 887b4307b6 Use _PyObject_FastCallVa() in type slots
Issue #28915: Replace Py_VaBuildValue()+PyObject_Call() with
_PyObject_FastCallVa() to avoid the creation of temporary tuple.
2016-12-09 00:41:46 +01:00
Victor Stinner 7f39c0ccfc Add _PyObject_VaCallFunctionObjArgs() private function
Issue #28915: Similar to _PyObject_CallFunctionObjArgs() but use va_list to
pass arguments.
2016-12-09 00:40:33 +01:00
Victor Stinner 7e42541d08 Use _PyObject_CallMethodIdObjArgs()
Issue #28915: Replace _PyObject_CallMethodId() with
_PyObject_CallMethodIdObjArgs() when the format string only use the format 'O'
for objects, like "(O)".

_PyObject_CallMethodIdObjArgs() avoids the code to parse a format string and
avoids the creation of a temporary tuple.
2016-12-09 00:36:19 +01:00
Victor Stinner ef7def94c7 _PyObject_CallFunctionVa() uses fast call
Issue #28915: Use _Py_VaBuildStack() to build a C array of PyObject* and then
use _PyObject_FastCall().

The function has a special case if the stack only contains one parameter and
the parameter is a tuple: "unpack" the tuple of arguments in this case.
2016-12-09 00:31:47 +01:00
Victor Stinner e83aab12b0 Add _PyObject_CallFunctionVa() helper
Issue #28915: Add _PyObject_CallFunctionVa() helper to factorize code of
functions:

* PyObject_CallFunction()
* _PyObject_CallFunction_SizeT()
* callmethod()
2016-12-09 00:22:56 +01:00
Victor Stinner 3bb711998d Add _PyObject_FastCallVa() helper
Issue #28915: Add _PyObject_FastCallVa() helper to factorize code of functions:

* PyObject_CallFunctionObjArgs()
* PyObject_CallMethodObjArgs()
* _PyObject_CallMethodIdObjArgs()

Inline objargs_mkstack() into _PyObject_FastCallVa(), remove
objargs_mkstack().
2016-12-09 00:21:55 +01:00
INADA Naoki ba6097734d Issue #28818: Simplify lookdict functions 2016-12-07 20:41:42 +09:00
Serhiy Storchaka 1b58d8aafa Issue #5322: Fixed setting __new__ to a PyCFunction inside Python code.
Original patch by Andreas Stührk.
2016-12-07 11:27:55 +02:00
Serhiy Storchaka 0c78634d78 Issue #5322: Fixed setting __new__ to a PyCFunction inside Python code.
Original patch by Andreas Stührk.
2016-12-07 11:26:49 +02:00
Victor Stinner f2d568ff95 _PyObject_FastCallKeywords() now calls directly tp_call
_PyObject_FastCallKeywords() doesn't call _PyObject_FastCallDict() anymore:
call directly tp_call.
2016-12-07 00:37:38 +01:00
Victor Stinner 2f35ca3e84 Fix typo in a comment of abstract.c 2016-12-06 18:49:15 +01:00
Victor Stinner f17c3de263 Use _PyObject_CallNoArg()
Replace:
    PyObject_CallFunctionObjArgs(callable, NULL)
with:
    _PyObject_CallNoArg(callable)
2016-12-06 18:46:19 +01:00
Victor Stinner a5ed5f000a Use _PyObject_CallNoArg()
Replace:
    PyObject_CallObject(callable, NULL)
with:
    _PyObject_CallNoArg(callable)
2016-12-06 18:45:50 +01:00
Victor Stinner 2d0eb65f45 Uniformize argument names of "call" functions
Issue #28838: Rename parameters of the "calls" functions of the Python C API.

* Rename 'callable_object' and 'func' to 'callable': any Python callable object
  is accepted, not only Python functions
* Rename 'method' and 'nameid' to 'name' (method name)
* Rename 'o' to 'obj'
* Move, fix and update documentation of PyObject_CallXXX() functions
  in abstract.h
* Update also the documentaton of the C API (update parameter names)
2016-12-06 16:27:24 +01:00
Serhiy Storchaka fb3134f4d4 Issue #28808: PyUnicode_CompareWithASCIIString() now never raises exceptions. 2016-12-06 00:20:26 +02:00
Serhiy Storchaka 9a953dbb34 Issue #28808: PyUnicode_CompareWithASCIIString() now never raises exceptions. 2016-12-06 00:17:45 +02:00
Serhiy Storchaka 419967b832 Issue #28808: PyUnicode_CompareWithASCIIString() now never raises exceptions. 2016-12-06 00:13:34 +02:00
Victor Stinner 9083eb6aac Use directly _PyObject_GenericSetAttrWithDict()
Modify type_setattro() to call directly _PyObject_GenericSetAttrWithDict()
instead of PyObject_GenericSetAttr().

PyObject_GenericSetAttr() is a thin wrapper to
_PyObject_GenericSetAttrWithDict().
2016-12-05 18:23:27 +01:00
Victor Stinner 7bfb42d5b7 Issue #28858: Remove _PyObject_CallArg1() macro
Replace
   _PyObject_CallArg1(func, arg)
with
   PyObject_CallFunctionObjArgs(func, arg, NULL)

Using the _PyObject_CallArg1() macro increases the usage of the C stack, which
was unexpected and unwanted. PyObject_CallFunctionObjArgs() doesn't have this
issue.
2016-12-05 17:04:32 +01:00
Nick Coghlan d77e5b7211 Merge #23722 from 3.6 2016-12-05 16:59:22 +10:00
Nick Coghlan 19d246745d Issue #23722: improve __classcell__ compatibility
Handling zero-argument super() in __init_subclass__ and
__set_name__ involved moving __class__ initialisation to
type.__new__. This requires cooperation from custom
metaclasses to ensure that the new __classcell__ entry
is passed along appropriately.

The initial implementation of that change resulted in abruptly
broken zero-argument super() support in metaclasses that didn't
adhere to the new requirements (such as Django's metaclass for
Model definitions).

The updated approach adopted here instead emits a deprecation
warning for those cases, and makes them work the same way they
did in Python 3.5.

This patch also improves the related class machinery documentation
to cover these details and to include more reader-friendly
cross-references and index entries.
2016-12-05 16:47:55 +10:00
Victor Stinner de4ae3d486 Backed out changeset b9c9691c72c5
Issue #28858: The change b9c9691c72c5 introduced a regression. It seems like
_PyObject_CallArg1() uses more stack memory than
PyObject_CallFunctionObjArgs().
2016-12-04 22:59:09 +01:00
Serhiy Storchaka 5adfac2c1b Issue #5322: Fixed setting __new__ to a PyCFunction inside Python code.
Original patch by Andreas Stührk.
2016-12-02 08:42:43 +02:00
Victor Stinner 4778eab1f2 Replace PyObject_CallFunction() with fastcall
Replace
    PyObject_CallFunction(func, "O", arg)
and
    PyObject_CallFunction(func, "O", arg, NULL)
with
    _PyObject_CallArg1(func, arg)

Replace
    PyObject_CallFunction(func, NULL)
with
    _PyObject_CallNoArg(func)

_PyObject_CallNoArg() and _PyObject_CallArg1() are simpler and don't allocate
memory on the C stack.
2016-12-01 14:51:04 +01:00
Victor Stinner 27580c1fb5 Replace PyObject_CallFunctionObjArgs() with fastcall
* PyObject_CallFunctionObjArgs(func, NULL) => _PyObject_CallNoArg(func)
* PyObject_CallFunctionObjArgs(func, arg, NULL) => _PyObject_CallArg1(func, arg)

PyObject_CallFunctionObjArgs() allocates 40 bytes on the C stack and requires
extra work to "parse" C arguments to build a C array of PyObject*.

_PyObject_CallNoArg() and _PyObject_CallArg1() are simpler and don't allocate
memory on the C stack.

This change is part of the fastcall project. The change on listsort() is
related to the issue #23507.
2016-12-01 14:43:22 +01:00
Victor Stinner 8be1c39eb3 Backed out changeset 7efddbf1aa70 2016-11-30 12:10:54 +01:00
Victor Stinner ebea9988e0 Uniformize argument names of "call" functions
* Callable object: callable, o, callable_object => func
* Object for method calls: o => obj
* Method name: name or nameid => method

Cleanup also the C code:

* Don't initialize variables to NULL if they are not used before their first
  assignement
* Add braces for readability
2016-11-29 18:47:56 +01:00
Serhiy Storchaka baa7223bcd Issue #28797: Modifying the class __dict__ inside the __set_name__ method of
a descriptor that is used inside that class no longer prevents calling the
 __set_name__ method of other descriptors.
2016-11-29 09:56:07 +02:00
Serhiy Storchaka 9ec07721f4 Issue #28797: Modifying the class __dict__ inside the __set_name__ method of
a descriptor that is used inside that class no longer prevents calling the
 __set_name__ method of other descriptors.
2016-11-29 09:54:17 +02:00
Victor Stinner ed6de7345e Merge 3.6 2016-11-24 22:33:49 +01:00
Victor Stinner f7d199ff32 Fix _PyGen_yf()
Issue #28782: Fix a bug in the implementation ``yield from`` when checking
if the next instruction is YIELD_FROM. Regression introduced by WORDCODE
(issue #26647).

Reviewed by Serhiy Storchaka and Yury Selivanov.
2016-11-24 22:33:01 +01:00
Serhiy Storchaka 99250d5c63 Issue #28774: Simplified encoding a str result of an error handler in ASCII
and Latin1 encoders.
2016-11-23 15:13:00 +02:00
Xiang Zhang d04d8474df Issue #28774: Fix start/end pos in unicode_encode_ucs1().
Fix error position of the unicode error in ASCII and Latin1
encoders when a string returned by the error handler contains multiple
non-encodable characters (non-ASCII for the ASCII codec, characters out
of the U+0000-U+00FF range for Latin1).
2016-11-23 19:34:01 +08:00
INADA Naoki 2c5a830f2a Issue #28731: Optimize _PyDict_NewPresized() to create correct size dict.
Improve speed of dict literal with constant keys up to 30%.
2016-12-07 18:34:44 +09:00
Serhiy Storchaka 007d7ff73f Issue #28761: The fields name and doc of structures PyMemberDef, PyGetSetDef,
PyStructSequence_Field, PyStructSequence_Desc, and wrapperbase are now of
type "const char *" rather of "char *".
2016-11-22 07:58:08 +02:00
INADA Naoki 92c50eee52 Issue #28731: Optimize _PyDict_NewPresized() to create correct size dict
Improve speed of dict literal with constant keys up to 30%.
2016-11-22 00:57:02 +09:00
Serhiy Storchaka 50911476f5 Issue #28760: Clean up and fix comments in PyUnicode_AsUnicodeEscapeString().
Patch by Xiang Zhang.
2016-11-21 11:47:16 +02:00
Serhiy Storchaka ac0720eaa4 Issue #28760: Clean up and fix comments in PyUnicode_AsUnicodeEscapeString().
Patch by Xiang Zhang.
2016-11-21 11:46:51 +02:00
Serhiy Storchaka b57d9eac41 Issue #28748: Private variable _Py_PackageContext is now of type "const char *"
rather of "char *".
2016-11-21 10:25:54 +02:00
Serhiy Storchaka 460bd0d284 Issue #19569: Compiler warnings are now emitted if use most of deprecated
functions.
2016-11-20 12:16:46 +02:00
Serhiy Storchaka 85b0f5beb1 Added the const qualifier to char* variables that refer to readonly internal
UTF-8 represenatation of Unicode objects.
2016-11-20 10:16:47 +02:00
Serhiy Storchaka a98c4a984b Replaced outdated macros _PyUnicode_AsString and _PyUnicode_AsStringAndSize
with PyUnicode_AsUTF8 and PyUnicode_AsUTF8AndSize.
2016-11-20 09:13:40 +02:00
Serhiy Storchaka 06515833fe Replaced outdated macros _PyUnicode_AsString and _PyUnicode_AsStringAndSize
with PyUnicode_AsUTF8 and PyUnicode_AsUTF8AndSize.
2016-11-20 09:13:07 +02:00
Yury Selivanov bb2356500c Merge 3.6 (issue #28721) 2016-11-16 18:16:32 -05:00
Yury Selivanov 41782e4970 Issue #28721: Fix asynchronous generators aclose() and athrow() 2016-11-16 18:16:17 -05:00
Serhiy Storchaka 27b74244fb Issue #28701: _PyUnicode_EqualToASCIIId and _PyUnicode_EqualToASCIIString now
require ASCII right argument and assert this condition in debug build.
2016-11-16 20:03:03 +02:00
Serhiy Storchaka a83a6a3275 Issue #28701: _PyUnicode_EqualToASCIIId and _PyUnicode_EqualToASCIIString now
require ASCII right argument and assert this condition in debug build.
2016-11-16 20:02:44 +02:00
Serhiy Storchaka e6d6131f78 Fixed an off-by-one error in _PyUnicode_EqualToASCIIString (issue #28701). 2016-11-16 16:13:13 +02:00
Serhiy Storchaka df66b9c425 Fixed an off-by-one error in _PyUnicode_EqualToASCIIString (issue #28701). 2016-11-16 16:12:56 +02:00
Serhiy Storchaka 292dd1b2ad Fixed an off-by-one error in _PyUnicode_EqualToASCIIString (issue #28701). 2016-11-16 16:12:34 +02:00
Serhiy Storchaka 503db266a5 Issue #21449: Removed private function _PyUnicode_CompareWithId. 2016-11-16 15:56:50 +02:00
Serhiy Storchaka dddec81b2d Issue #21449: Removed private function _PyUnicode_CompareWithId. 2016-11-16 15:56:27 +02:00
Serhiy Storchaka 29a5447360 Issue #28701: Replace _PyUnicode_CompareWithId with _PyUnicode_EqualToASCIIId.
The latter function is more readable, faster and doesn't raise exceptions.

Based on patch by Xiang Zhang.
2016-11-16 15:41:31 +02:00
Serhiy Storchaka fab6acd9f5 Issue #28701: Replace _PyUnicode_CompareWithId with _PyUnicode_EqualToASCIIId.
The latter function is more readable, faster and doesn't raise exceptions.

Based on patch by Xiang Zhang.
2016-11-16 15:41:11 +02:00
Serhiy Storchaka f5894dd646 Issue #28701: Replace _PyUnicode_CompareWithId with _PyUnicode_EqualToASCIIId.
The latter function is more readable, faster and doesn't raise exceptions.

Based on patch by Xiang Zhang.
2016-11-16 15:40:39 +02:00
Serhiy Storchaka 1a73bf365e Issue #28701: Replace PyUnicode_CompareWithASCIIString with _PyUnicode_EqualToASCIIString.
The latter function is more readable, faster and doesn't raise exceptions.
2016-11-16 10:19:57 +02:00
Serhiy Storchaka 3b73ea1278 Issue #28701: Replace PyUnicode_CompareWithASCIIString with _PyUnicode_EqualToASCIIString.
The latter function is more readable, faster and doesn't raise exceptions.
2016-11-16 10:19:20 +02:00
Serhiy Storchaka f4934ea77d Issue #28701: Replace PyUnicode_CompareWithASCIIString with _PyUnicode_EqualToASCIIString.
The latter function is more readable, faster and doesn't raise exceptions.
2016-11-16 10:17:58 +02:00
Victor Stinner c7a8f67411 Issue #28618: Mark dict lookup functions as hot
It's common to see these functions in the top 3 of "perf report".
2016-11-15 15:13:40 +01:00
Serhiy Storchaka 616034eb73 Issue #28648: Fixed crash in Py_DecodeLocale() in debug build on Mac OS X
when decode astral characters.
2016-11-12 14:37:11 +02:00
Serhiy Storchaka babe4f8e5e Issue #28648: Fixed crash in Py_DecodeLocale() in debug build on Mac OS X
when decode astral characters.
2016-11-12 14:36:02 +02:00
Serhiy Storchaka 6b4b6e956e Issue #28648: Fixed crash in Py_DecodeLocale() in debug build on Mac OS X
when decode astral characters.
2016-11-12 14:35:46 +02:00
Serhiy Storchaka 84293aff9f Issue #28648: Fixed crash in Py_DecodeLocale() in debug build on Mac OS X
when decode astral characters.
2016-11-12 14:29:48 +02:00
Serhiy Storchaka b626643734 Issue #28648: Fixed crash in Py_DecodeLocale() in debug build on Mac OS X
when decode astral characters.
2016-11-12 14:28:06 +02:00
Victor Stinner c6944e7edc Issue #28618: Make hot functions using __attribute__((hot))
When Python is not compiled with PGO, the performance of Python on call_simple
and call_method microbenchmarks depend highly on the code placement. In the
worst case, the performance slowdown can be up to 70%.

The GCC __attribute__((hot)) attribute helps to keep hot code close to reduce
the risk of such major slowdown. This attribute is ignored when Python is
compiled with PGO.

The following functions are considered as hot according to statistics collected
by perf record/perf report:

* _PyEval_EvalFrameDefault()
* call_function()
* _PyFunction_FastCall()
* PyFrame_New()
* frame_dealloc()
* PyErr_Occurred()
2016-11-11 02:13:35 +01:00
Victor Stinner 0cae609847 Use PyThreadState_GET() in performance critical code
It seems like _PyThreadState_UncheckedGet() is not inlined as expected, even
when using gcc -O3.
2016-11-11 01:43:56 +01:00
Yury Selivanov f787c90764 Merge 3.6 (issue #27942) 2016-11-09 09:43:37 -05:00
Yury Selivanov 6c10650617 Merge 3.5 (issue #27942) 2016-11-09 09:43:08 -05:00
Yury Selivanov d2fd3599ab ssue #27942: Fix memory leak in codeobject.c 2016-11-09 09:42:14 -05:00
Yury Selivanov 9a2ffbd47c Merge 3.6 (issue #28003) 2016-11-08 19:46:41 -05:00
Yury Selivanov 29310c47a7 Issue #28003: Make WrappedVal, ASend and AThrow GC types 2016-11-08 19:46:22 -05:00
Yury Selivanov ff0273fb20 Merge 3.6 2016-11-08 19:20:08 -05:00
Yury Selivanov 49ffdf6bb2 Merge 3.5 2016-11-08 19:19:52 -05:00
Yury Selivanov 33499b7eed genobject: Remove unnecessary tp_free slots from aiter_wrapper and coro_wrapper 2016-11-08 19:19:28 -05:00
Serhiy Storchaka b2e64f903d Issue #28621: Sped up converting int to float by reusing faster bits counting
implementation.  Patch by Adrian Wielgosik.
2016-11-08 20:34:22 +02:00
Steve Dower 257a4c1503 Closes #27781: Removes special cases for the experimental aspect of PEP 529 2016-11-06 19:35:24 -08:00
Steve Dower 78057b4159 Closes #27781: Removes special cases for the experimental aspect of PEP 529 2016-11-06 19:35:08 -08:00
Benjamin Peterson 6e9d6b9536 merge 3.6 2016-11-06 13:01:23 -08:00
Benjamin Peterson 3bd7900155 merge 3.5 2016-11-06 13:01:15 -08:00
Benjamin Peterson db87c99444 make sure dict view types are initialized 2016-11-06 13:01:07 -08:00
Serhiy Storchaka 85bcf37e15 Issue #23996: Added _PyGen_SetStopIterationValue for safe raising
StopIteration with value. More safely handle non-normalized exceptions
in -_PyGen_FetchStopIterationValue.
2016-11-06 18:47:35 +02:00
Serhiy Storchaka 60e49aa756 Issue #23996: Added _PyGen_SetStopIterationValue for safe raising
StopIteration with value. More safely handle non-normalized exceptions
in -_PyGen_FetchStopIterationValue.
2016-11-06 18:47:03 +02:00
Serhiy Storchaka 24411f8a8d Issue #23996: Added _PyGen_SetStopIterationValue for safe raising
StopIteration with value. More safely handle non-normalized exceptions
in -_PyGen_FetchStopIterationValue.
2016-11-06 18:44:42 +02:00
Serhiy Storchaka f0b311bd73 Issue #28123: _PyDict_GetItem_KnownHash() now can raise an exception as
PyDict_GetItemWithError().  Patch by Xiang Zhang.
2016-11-06 13:18:24 +02:00
INADA Naoki 93f26f794d Issue #28583: PyDict_SetDefault didn't combine split table when needed.
Patch by Xiang Zhang.
2016-11-02 18:45:16 +09:00
Eric V. Smith 5646648678 Issue 28128: Print out better error/warning messages for invalid string escapes. Backport to 3.6. 2016-10-31 14:46:26 -04:00
Serhiy Storchaka 7f0514ad54 Backed out changeset 6b88dfc7b25d 2016-10-31 20:14:05 +02:00
Serhiy Storchaka 7984bff52a Issue #28385: An error message when non-empty format spec is passed to
object.__format__ now contains the name of actual type.
2016-10-30 19:37:46 +02:00
Serhiy Storchaka d1af5effc2 Issue #28385: An error message when non-empty format spec is passed to
object.__format__ now contains the name of actual type.
2016-10-30 19:33:54 +02:00
Serhiy Storchaka 998c9cdd42 Issue #28561: Clean up UTF-8 encoder: remove dead code, update comments, etc.
Patch by Xiang Zhang.
2016-10-30 18:25:27 +02:00
Serhiy Storchaka b7d14a09c2 Merge from 3.5. 2016-10-30 17:25:45 +02:00
Serhiy Storchaka 0438683939 Backed out changeset 9f7505019767 (issue #27275). 2016-10-30 17:17:24 +02:00
Serhiy Storchaka e26e20db95 Issue #28199: Microoptimized dict resizing. Based on patch by Naoki Inada. 2016-10-29 10:50:00 +03:00
Serhiy Storchaka 04230c4087 Issue #28123: _PyDict_GetItem_KnownHash() now can raise an exception as
PyDict_GetItemWithError().  Patch by Xiang Zhang.
2016-11-06 13:19:38 +02:00
INADA Naoki ca2d8be4ba Issue #28580: Optimize iterating split table values.
Patch by Xiang Zhang.
2016-11-04 16:59:10 +09:00
INADA Naoki a3498c7569 Issue #28583: PyDict_SetDefault didn't combine split table when needed.
Patch by Xiang Zhang.
2016-11-02 18:47:24 +09:00
Serhiy Storchaka 5e325d9c41 Merge from 3.6. 2016-10-31 20:15:48 +02:00
Eric V. Smith 42454af094 Issue 28128: Print out better error/warning messages for invalid string escapes. 2016-10-31 09:22:08 -04:00
Serhiy Storchaka 79879e002f Issue #28385: An error message when non-empty format spec is passed to
object.__format__ now contains the name of actual type.
2016-10-30 19:38:05 +02:00
Serhiy Storchaka b982ae8317 Issue #28561: Clean up UTF-8 encoder: remove dead code, update comments, etc.
Patch by Xiang Zhang.
2016-10-30 18:25:46 +02:00
Serhiy Storchaka a182d7bc14 Merge from 3.6. 2016-10-30 17:27:40 +02:00
Serhiy Storchaka 041794908b Issue #28199: Microoptimized dict resizing. Based on patch by Naoki Inada. 2016-10-29 10:50:00 +03:00
Serhiy Storchaka d76d8bfee1 Issue #28199: Microoptimized dict resizing. Based on patch by Naoki Inada. 2016-10-29 10:49:43 +03:00
Yury Selivanov f0bbee6228 Merge 3.6 (issue #28544) 2016-10-28 19:01:46 -04:00
Yury Selivanov 684ef2c888 Issue #28544: Pass `PyObject*` to _PyDict_Pop, not `PyDictObject*` 2016-10-28 19:01:21 -04:00
Serhiy Storchaka 2edcd1cba4 Issue #28426: Deprecated undocumented functions PyUnicode_AsEncodedObject(),
PyUnicode_AsDecodedObject(), PyUnicode_AsDecodedUnicode() and
PyUnicode_AsEncodedUnicode().
2016-10-27 21:08:00 +03:00
Serhiy Storchaka 0093907f0e Issue #28426: Deprecated undocumented functions PyUnicode_AsEncodedObject(),
PyUnicode_AsDecodedObject(), PyUnicode_AsDecodedUnicode() and
PyUnicode_AsEncodedUnicode().
2016-10-27 21:05:49 +03:00