Mailman 3 January 2017 - Python-checkins

BAD Benchmark Results for Python Default 2017-01-26
by lp_benchmark_robot＠intel.com 26 Jan '17

26 Jan '17

Results for project Python default, build date 2017-01-26 11:03:42 +0000 commit: 8feb6a5ce4c6 previous commit: ffc0840762e4 revision date: 2017-01-25 22:33:27 +0000 environment: Haswell-EP cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2x18 cores, stepping 2, LLC 45 MB mem: 128 GB os: CentOS 7.1 kernel: Linux 3.10.0-229.4.2.el7.x86_64 Baseline results were generated using release v3.4.3, with hash b4cbecbc0781 from 2015-02-25 12:15:33+00:00 ---------------------------------------------------------------------------------- benchmark relative change since change since current rev run std_dev* last run baseline with PGO ---------------------------------------------------------------------------------- :-) django_v2 0.24% -0.16% 9.91% 15.70% :-) pybench 0.16% -0.96% 2.61% 5.30% :-| regex_v8 3.60% 0.55% 0.89% 0.68% :-) nbody 0.11% 0.17% 3.49% 0.53% :-) json_dump_v2 0.32% -1.38% 9.38% 10.55% :-| normal_startup 0.93% 0.45% -0.02% 6.34% ---------------------------------------------------------------------------------- * Relative Standard Deviation (Standard Deviation/Average) If this is not displayed properly please visit our results page here: http://languagesperformance.intel.com/bad-benchmark-results-for-python-defa… Note: Benchmark results are measured in seconds. Subject Label Legend: Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration. NEUTRAL: performance did not change by more than 1% for any workload GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1% BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1% UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly. Intel technologies' features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

1 0

BAD Benchmark Results for Python 2.7 2017-01-26
by lp_benchmark_robot＠intel.com 26 Jan '17

26 Jan '17

Results for project Python 2.7, build date 2017-01-26 11:47:48 +0000 commit: 743ff7902879 previous commit: fb2885f9b4dd revision date: 2017-01-25 11:28:45 +0000 environment: Haswell-EP cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2x18 cores, stepping 2, LLC 45 MB mem: 128 GB os: CentOS 7.1 kernel: Linux 3.10.0-229.4.2.el7.x86_64 Baseline results were generated using release v2.7.10, with hash 15c95b7d81dc from 2015-05-23 16:02:14+00:00 ---------------------------------------------------------------------------------- benchmark relative change since change since current rev run std_dev* last run baseline with PGO ---------------------------------------------------------------------------------- :-) django_v2 0.21% -1.36% 4.59% 6.86% :-) pybench 0.15% -0.17% 7.63% 3.00% :-| regex_v8 0.79% -0.11% -0.24% 9.78% :-) nbody 0.06% -2.81% 11.28% 3.16% :-| json_dump_v2 0.26% -1.02% 0.50% 9.06% :-( normal_startup 1.88% 0.04% -2.47% 1.97% :-| ssbench 0.13% -0.14% 0.06% 2.39% ---------------------------------------------------------------------------------- * Relative Standard Deviation (Standard Deviation/Average) If this is not displayed properly please visit our results page here: http://languagesperformance.intel.com/bad-benchmark-results-for-python-2-7-… Note: Benchmark results for ssbench are measured in requests/second while all other are measured in seconds. Subject Label Legend: Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration. NEUTRAL: performance did not change by more than 1% for any workload GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1% BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1% UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly. Intel technologies' features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

1 0

Daily reference leaks (8feb6a5ce4c6): sum=4
by solipsis＠pitrou.net 26 Jan '17

26 Jan '17

results for 8feb6a5ce4c6 on branch "default" -------------------------------------------- test_collections leaked [7, 0, -7] memory blocks, sum=0 test_functools leaked [0, 3, 1] memory blocks, sum=4 Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/psf-users/antoine/refleaks/reflogYDAzU6', '--timeout', '7200']

1 0

cpython (3.6): remove comment about updating the parser module; we do not need to do that
by benjamin.peterson 25 Jan '17

25 Jan '17

https://hg.python.org/cpython/rev/3275d4b584a2 changeset: 106318:3275d4b584a2 branch: 3.6 parent: 106313:b4457fe7fdb8 user: Benjamin Peterson <benjamin(a)python.org> date: Wed Jan 25 23:35:40 2017 -0800 summary: remove comment about updating the parser module; we do not need to do that anymore (#26526) files: Grammar/Grammar | 7 ------- 1 files changed, 0 insertions(+), 7 deletions(-) diff --git a/Grammar/Grammar b/Grammar/Grammar --- a/Grammar/Grammar +++ b/Grammar/Grammar @@ -1,12 +1,5 @@ # Grammar for Python -# Note: Changing the grammar specified in this file will most likely -# require corresponding changes in the parser module -# (../Modules/parsermodule.c). If you can't make the changes to -# that module yourself, please co-ordinate the required changes -# with someone who can; ask around on python-dev for help. Fred -# Drake <fdrake(a)acm.org> will probably be listening there. - # NOTE WELL: You should also follow all the steps listed at # https://docs.python.org/devguide/grammar.html -- Repository URL: https://hg.python.org/cpython

1 0

cpython (merge 3.6 -> default): merge 3.6
by benjamin.peterson 25 Jan '17

25 Jan '17

https://hg.python.org/cpython/rev/34fb29694c76 changeset: 106319:34fb29694c76 parent: 106317:8feb6a5ce4c6 parent: 106318:3275d4b584a2 user: Benjamin Peterson <benjamin(a)python.org> date: Wed Jan 25 23:35:46 2017 -0800 summary: merge 3.6 files: Grammar/Grammar | 7 ------- 1 files changed, 0 insertions(+), 7 deletions(-) diff --git a/Grammar/Grammar b/Grammar/Grammar --- a/Grammar/Grammar +++ b/Grammar/Grammar @@ -1,12 +1,5 @@ # Grammar for Python -# Note: Changing the grammar specified in this file will most likely -# require corresponding changes in the parser module -# (../Modules/parsermodule.c). If you can't make the changes to -# that module yourself, please co-ordinate the required changes -# with someone who can; ask around on python-dev for help. Fred -# Drake <fdrake(a)acm.org> will probably be listening there. - # NOTE WELL: You should also follow all the steps listed at # https://docs.python.org/devguide/grammar.html -- Repository URL: https://hg.python.org/cpython

1 0

cpython: Issue #29358: Add postcondition checks on types
by victor.stinner 25 Jan '17

25 Jan '17

https://hg.python.org/cpython/rev/8feb6a5ce4c6 changeset: 106317:8feb6a5ce4c6 user: Victor Stinner <victor.stinner(a)gmail.com> date: Wed Jan 25 23:33:27 2017 +0100 summary: Issue #29358: Add postcondition checks on types files: Objects/typeobject.c | 27 ++++++++++++++++++++++++--- 1 files changed, 24 insertions(+), 3 deletions(-) diff --git a/Objects/typeobject.c b/Objects/typeobject.c --- a/Objects/typeobject.c +++ b/Objects/typeobject.c @@ -121,6 +121,22 @@ return NULL; } +#ifdef Py_DEBUG +static int +_PyType_CheckConsistency(PyTypeObject *type) +{ + if (!(type->tp_flags & Py_TPFLAGS_READY)) { + /* don't check types before PyType_Ready() */ + return 1; + } + + assert(!(type->tp_flags & Py_TPFLAGS_READYING)); + assert(type->tp_mro != NULL && PyTuple_Check(type->tp_mro)); + assert(type->tp_dict != NULL); + return 1; +} +#endif + static const char * _PyType_DocWithoutSignature(const char *name, const char *internal_doc) { @@ -719,6 +735,7 @@ Py_DECREF(old_bases); Py_DECREF(old_base); + assert(_PyType_CheckConsistency(type)); return res; undo: @@ -752,6 +769,7 @@ Py_DECREF(old_base); } + assert(_PyType_CheckConsistency(type)); return -1; } @@ -3034,6 +3052,7 @@ static int type_setattro(PyTypeObject *type, PyObject *name, PyObject *value) { + int res; if (!(type->tp_flags & Py_TPFLAGS_HEAPTYPE)) { PyErr_Format( PyExc_TypeError, @@ -3043,7 +3062,9 @@ } if (_PyObject_GenericSetAttrWithDict((PyObject *)type, name, value, NULL) < 0) return -1; - return update_slot(type, name); + res = update_slot(type, name); + assert(_PyType_CheckConsistency(type)); + return res; } extern void @@ -4851,7 +4872,7 @@ Py_ssize_t i, n; if (type->tp_flags & Py_TPFLAGS_READY) { - assert(type->tp_dict != NULL); + assert(_PyType_CheckConsistency(type)); return 0; } assert((type->tp_flags & Py_TPFLAGS_READYING) == 0); @@ -5045,9 +5066,9 @@ } /* All done -- set the ready flag */ - assert(type->tp_dict != NULL); type->tp_flags = (type->tp_flags & ~Py_TPFLAGS_READYING) | Py_TPFLAGS_READY; + assert(_PyType_CheckConsistency(type)); return 0; error: -- Repository URL: https://hg.python.org/cpython

1 0

BAD Benchmark Results for Python Default 2017-01-25
by lp_benchmark_robot＠intel.com 25 Jan '17

25 Jan '17

Results for project Python default, build date 2017-01-25 11:03:49 +0000 commit: ffc0840762e4 previous commit: cebc9c7ad195 revision date: 2017-01-24 22:30:04 +0000 environment: Haswell-EP cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2x18 cores, stepping 2, LLC 45 MB mem: 128 GB os: CentOS 7.1 kernel: Linux 3.10.0-229.4.2.el7.x86_64 Baseline results were generated using release v3.4.3, with hash b4cbecbc0781 from 2015-02-25 12:15:33+00:00 ---------------------------------------------------------------------------------- benchmark relative change since change since current rev run std_dev* last run baseline with PGO ---------------------------------------------------------------------------------- :-) django_v2 0.17% -0.25% 10.06% 15.19% :-) pybench 0.15% -0.01% 3.53% 4.46% :-| regex_v8 3.83% -1.26% 0.34% 3.90% :-) nbody 0.11% -0.04% 3.33% -0.25% :-) json_dump_v2 0.27% -1.06% 10.61% 10.59% :-| normal_startup 1.08% -0.53% -0.35% 6.89% ---------------------------------------------------------------------------------- * Relative Standard Deviation (Standard Deviation/Average) If this is not displayed properly please visit our results page here: http://languagesperformance.intel.com/bad-benchmark-results-for-python-defa… Note: Benchmark results are measured in seconds. Subject Label Legend: Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration. NEUTRAL: performance did not change by more than 1% for any workload GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1% BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1% UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly. Intel technologies' features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

1 0

GOOD Benchmark Results for Python 2.7 2017-01-25
by lp_benchmark_robot＠intel.com 25 Jan '17

25 Jan '17

Results for project Python 2.7, build date 2017-01-25 11:47:43 +0000 commit: fb2885f9b4dd previous commit: 8e3d412f8e89 revision date: 2017-01-24 00:22:45 +0000 environment: Haswell-EP cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2x18 cores, stepping 2, LLC 45 MB mem: 128 GB os: CentOS 7.1 kernel: Linux 3.10.0-229.4.2.el7.x86_64 Baseline results were generated using release v2.7.10, with hash 15c95b7d81dc from 2015-05-23 16:02:14+00:00 ---------------------------------------------------------------------------------- benchmark relative change since change since current rev run std_dev* last run baseline with PGO ---------------------------------------------------------------------------------- :-) django_v2 0.12% 1.38% 5.87% 2.55% :-) pybench 0.20% -0.02% 7.79% 3.17% :-| regex_v8 0.56% 0.02% -0.13% 10.28% :-) nbody 0.07% 0.08% 13.71% 2.65% :-| json_dump_v2 0.31% 0.34% 1.50% 6.16% :-( normal_startup 1.97% 0.52% -2.51% 2.18% :-| ssbench 0.14% -0.04% 0.19% 2.14% ---------------------------------------------------------------------------------- * Relative Standard Deviation (Standard Deviation/Average) If this is not displayed properly please visit our results page here: http://languagesperformance.intel.com/good-benchmark-results-for-python-2-7… Note: Benchmark results for ssbench are measured in requests/second while all other are measured in seconds. Subject Label Legend: Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration. NEUTRAL: performance did not change by more than 1% for any workload GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1% BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1% UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly. Intel technologies' features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

1 0

cpython: Issue #29369: Use Py_IDENTIFIER in Python-ast.c
by inada.naoki 25 Jan '17

25 Jan '17

https://hg.python.org/cpython/rev/a507882736b2 changeset: 106316:a507882736b2 parent: 106314:6093ce8eed6c user: INADA Naoki <songofacandy(a)gmail.com> date: Wed Jan 25 22:33:43 2017 +0900 summary: Issue #29369: Use Py_IDENTIFIER in Python-ast.c files: Parser/asdl_c.py | 21 +++++++++++++-------- Python/Python-ast.c | 18 ++++++++++++------ 2 files changed, 25 insertions(+), 14 deletions(-) diff --git a/Parser/asdl_c.py b/Parser/asdl_c.py --- a/Parser/asdl_c.py +++ b/Parser/asdl_c.py @@ -622,6 +622,9 @@ def visitModule(self, mod): self.emit(""" +_Py_IDENTIFIER(_fields); +_Py_IDENTIFIER(_attributes); + typedef struct { PyObject_HEAD PyObject *dict; @@ -650,7 +653,6 @@ static int ast_type_init(PyObject *self, PyObject *args, PyObject *kw) { - _Py_IDENTIFIER(_fields); Py_ssize_t i, numfields = 0; int res = -1; PyObject *key, *value, *fields; @@ -775,6 +777,8 @@ static PyTypeObject* make_type(char *type, PyTypeObject* base, char**fields, int num_fields) { + _Py_IDENTIFIER(__module__); + _Py_IDENTIFIER(_ast); PyObject *fnames, *result; int i; fnames = PyTuple_New(num_fields); @@ -787,8 +791,11 @@ } PyTuple_SET_ITEM(fnames, i, field); } - result = PyObject_CallFunction((PyObject*)&PyType_Type, "s(O){sOss}", - type, base, "_fields", fnames, "__module__", "_ast"); + result = PyObject_CallFunction((PyObject*)&PyType_Type, "s(O){OOOO}", + type, base, + _PyUnicode_FromId(&PyId__fields), fnames, + _PyUnicode_FromId(&PyId___module__), + _PyUnicode_FromId(&PyId__ast)); Py_DECREF(fnames); return (PyTypeObject*)result; } @@ -796,7 +803,6 @@ static int add_attributes(PyTypeObject* type, char**attrs, int num_fields) { int i, result; - _Py_IDENTIFIER(_attributes); PyObject *s, *l = PyTuple_New(num_fields); if (!l) return 0; @@ -942,8 +948,8 @@ d = AST_type.tp_dict; empty_tuple = PyTuple_New(0); if (!empty_tuple || - PyDict_SetItemString(d, "_fields", empty_tuple) < 0 || - PyDict_SetItemString(d, "_attributes", empty_tuple) < 0) { + _PyDict_SetItemId(d, &PyId__fields, empty_tuple) < 0 || + _PyDict_SetItemId(d, &PyId__attributes, empty_tuple) < 0) { Py_XDECREF(empty_tuple); return -1; } @@ -1091,8 +1097,7 @@ self.emit("%s o = (%s)_o;" % (ctype, ctype), 1) self.emit("PyObject *result = NULL, *value = NULL;", 1) self.emit('if (!o) {', 1) - self.emit("Py_INCREF(Py_None);", 2) - self.emit('return Py_None;', 2) + self.emit("Py_RETURN_NONE;", 2) self.emit("}", 1) self.emit('', 0) diff --git a/Python/Python-ast.c b/Python/Python-ast.c --- a/Python/Python-ast.c +++ b/Python/Python-ast.c @@ -504,6 +504,9 @@ }; +_Py_IDENTIFIER(_fields); +_Py_IDENTIFIER(_attributes); + typedef struct { PyObject_HEAD PyObject *dict; @@ -532,7 +535,6 @@ static int ast_type_init(PyObject *self, PyObject *args, PyObject *kw) { - _Py_IDENTIFIER(_fields); Py_ssize_t i, numfields = 0; int res = -1; PyObject *key, *value, *fields; @@ -657,6 +659,8 @@ static PyTypeObject* make_type(char *type, PyTypeObject* base, char**fields, int num_fields) { + _Py_IDENTIFIER(__module__); + _Py_IDENTIFIER(_ast); PyObject *fnames, *result; int i; fnames = PyTuple_New(num_fields); @@ -669,8 +673,11 @@ } PyTuple_SET_ITEM(fnames, i, field); } - result = PyObject_CallFunction((PyObject*)&PyType_Type, "s(O){sOss}", - type, base, "_fields", fnames, "__module__", "_ast"); + result = PyObject_CallFunction((PyObject*)&PyType_Type, "s(O){OOOO}", + type, base, + _PyUnicode_FromId(&PyId__fields), fnames, + _PyUnicode_FromId(&PyId___module__), + _PyUnicode_FromId(&PyId__ast)); Py_DECREF(fnames); return (PyTypeObject*)result; } @@ -678,7 +685,6 @@ static int add_attributes(PyTypeObject* type, char**attrs, int num_fields) { int i, result; - _Py_IDENTIFIER(_attributes); PyObject *s, *l = PyTuple_New(num_fields); if (!l) return 0; @@ -824,8 +830,8 @@ d = AST_type.tp_dict; empty_tuple = PyTuple_New(0); if (!empty_tuple || - PyDict_SetItemString(d, "_fields", empty_tuple) < 0 || - PyDict_SetItemString(d, "_attributes", empty_tuple) < 0) { + _PyDict_SetItemId(d, &PyId__fields, empty_tuple) < 0 || + _PyDict_SetItemId(d, &PyId__attributes, empty_tuple) < 0) { Py_XDECREF(empty_tuple); return -1; } -- Repository URL: https://hg.python.org/cpython

1 0

cpython (2.7): Issue #27867: Replaced function PySlice_GetIndicesEx() with a macro.
by serhiy.storchaka 25 Jan '17

25 Jan '17

https://hg.python.org/cpython/rev/d5590f357d74 changeset: 106311:d5590f357d74 branch: 2.7 parent: 106292:fb2885f9b4dd user: Serhiy Storchaka <storchaka(a)gmail.com> date: Wed Jan 25 13:22:06 2017 +0200 summary: Issue #27867: Replaced function PySlice_GetIndicesEx() with a macro. files: Include/sliceobject.h | 10 +++ Misc/NEWS | 5 + Objects/sliceobject.c | 78 +++++++++++++++++++++--------- 3 files changed, 70 insertions(+), 23 deletions(-) diff --git a/Include/sliceobject.h b/Include/sliceobject.h --- a/Include/sliceobject.h +++ b/Include/sliceobject.h @@ -38,6 +38,16 @@ Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t *step, Py_ssize_t *slicelength); +#define PySlice_GetIndicesEx(slice, length, start, stop, step, slicelen) ( \ + _PySlice_Unpack((PyObject *)(slice), (start), (stop), (step)) < 0 ? -1 : \ + ((*slicelen = _PySlice_AdjustIndices((length), (start), (stop), *(step))), \ + 0)) +PyAPI_FUNC(int) _PySlice_Unpack(PyObject *slice, + Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t *step); +PyAPI_FUNC(Py_ssize_t) _PySlice_AdjustIndices(Py_ssize_t length, + Py_ssize_t *start, Py_ssize_t *stop, + Py_ssize_t step); + #ifdef __cplusplus } #endif diff --git a/Misc/NEWS b/Misc/NEWS --- a/Misc/NEWS +++ b/Misc/NEWS @@ -68,6 +68,11 @@ - Issue #28925: cPickle now correctly propagates errors when unpickle instances of old-style classes. +C API +----- + +- Issue #27867: Function PySlice_GetIndicesEx() is replaced with a macro. + Build ----- diff --git a/Objects/sliceobject.c b/Objects/sliceobject.c --- a/Objects/sliceobject.c +++ b/Objects/sliceobject.c @@ -131,13 +131,12 @@ } int -PySlice_GetIndicesEx(PySliceObject *r, Py_ssize_t length, - Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t *step, Py_ssize_t *slicelength) +_PySlice_Unpack(PyObject *_r, + Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t *step) { + PySliceObject *r = (PySliceObject *)_r; /* this is harder to get right than you might think */ - Py_ssize_t defstart, defstop; - if (r->step == Py_None) { *step = 1; } @@ -150,42 +149,75 @@ } } - defstart = *step < 0 ? length-1 : 0; - defstop = *step < 0 ? -1 : length; - if (r->start == Py_None) { - *start = defstart; + *start = *step < 0 ? PY_SSIZE_T_MAX-1 : 0;; } else { if (!_PyEval_SliceIndex(r->start, start)) return -1; - if (*start < 0) *start += length; - if (*start < 0) *start = (*step < 0) ? -1 : 0; - if (*start >= length) - *start = (*step < 0) ? length - 1 : length; } if (r->stop == Py_None) { - *stop = defstop; + *stop = *step < 0 ? -PY_SSIZE_T_MAX : PY_SSIZE_T_MAX; } else { if (!_PyEval_SliceIndex(r->stop, stop)) return -1; - if (*stop < 0) *stop += length; - if (*stop < 0) *stop = (*step < 0) ? -1 : 0; - if (*stop >= length) - *stop = (*step < 0) ? length - 1 : length; } - if ((*step < 0 && *stop >= *start) - || (*step > 0 && *start >= *stop)) { - *slicelength = 0; + return 0; +} + +Py_ssize_t +_PySlice_AdjustIndices(Py_ssize_t length, + Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t step) +{ + /* this is harder to get right than you might think */ + + assert(step != 0); + assert(step >= -PY_SSIZE_T_MAX); + + if (*start < 0) { + *start += length; + if (*start < 0) { + *start = (step < 0) ? -1 : 0; + } } - else if (*step < 0) { - *slicelength = (*stop-*start+1)/(*step)+1; + else if (*start >= length) { + *start = (step < 0) ? length - 1 : length; + } + + if (*stop < 0) { + *stop += length; + if (*stop < 0) { + *stop = (step < 0) ? -1 : 0; + } + } + else if (*stop >= length) { + *stop = (step < 0) ? length - 1 : length; + } + + if (step < 0) { + if (*stop < *start) { + return (*start - *stop - 1) / (-step) + 1; + } } else { - *slicelength = (*stop-*start-1)/(*step)+1; + if (*start < *stop) { + return (*stop - *start - 1) / step + 1; + } } + return 0; +} +#undef PySlice_GetIndicesEx + +int +PySlice_GetIndicesEx(PySliceObject *r, Py_ssize_t length, + Py_ssize_t *start, Py_ssize_t *stop, Py_ssize_t *step, + Py_ssize_t *slicelength) +{ + if (_PySlice_Unpack((PyObject *)r, start, stop, step) < 0) + return -1; + *slicelength = _PySlice_AdjustIndices(length, start, stop, *step); return 0; } -- Repository URL: https://hg.python.org/cpython

1 0