[issue37761] Inaccurate explanation of ArgumentParser.add_argument()'s name-or-flags in JA

Change by Inada Naoki : -- resolution: -> third party stage: -> resolved status: open -> closed ___ Python tracker <https://bugs.python.org/issue37761> ___ ___

[issue37587] JSON loads performance improvement for long strings

Inada Naoki added the comment: I tried without PGO and confirmed performance improved on GCC 7.2.0. No change on other compiler versions. $ ./python -m pyperf timeit -s "import json; x = json.dumps({'k': '1' * 2 ** 20})" "json.loads(x)" old: 9211e2

[issue37587] JSON loads performance improvement for long strings

Change by Inada Naoki : -- pull_requests: +14873 pull_request: https://github.com/python/cpython/pull/15134 ___ Python tracker <https://bugs.python.org/issue37

[issue37587] JSON loads performance improvement for long strings

Inada Naoki added the comment: And I confirmed performance improvement by my patch (GH-15134) on all of 4 compilers. $ ./python -m pyperf timeit -s "import json; x = json.dumps({'k': '1' * 2 ** 20})" "json.loads(x)" old: 9211e2 new: 8a758f opt2: 2

[issue34488] improve performance of BytesIO.writelines() by avoiding creation of unused PyLongs

Change by Inada Naoki : -- components: +IO -Extension Modules versions: +Python 3.9 -Python 3.8 ___ Python tracker <https://bugs.python.org/issue34488> ___ ___

[issue34488] improve performance of BytesIO.writelines() by avoiding creation of unused PyLongs

Inada Naoki added the comment: New changeset 3e41f3cabb661824a1a197116f7f5ead64eb6ced by Inada Naoki (Sergey Fedoseev) in branch 'master': bpo-34488: optimize BytesIO.writelines() (GH-8904) https://github.com/python/cpython/commit/3e41f3cabb661824a1a197116f7f5e

[issue34488] improve performance of BytesIO.writelines() by avoiding creation of unused PyLongs

Change by Inada Naoki : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue37781] Use "z" for PY_FORMAT_SIZE_T

New submission from Inada Naoki : MSVC 2015 supports "z" for size_t format. I'm not sure about 2013. AIX support it too. Now "z" is portable enough. https://mail.python.org/archives/list/python-...@python.org/thread/CAXKWESUIWJNJFLLXXWTQDUWTN3F7KOU/#QVSBYYI

[issue37781] Use "z" for PY_FORMAT_SIZE_T

Change by Inada Naoki : -- versions: +Python 3.9 ___ Python tracker <https://bugs.python.org/issue37781> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue37781] Use "z" for PY_FORMAT_SIZE_T

Change by Inada Naoki : -- keywords: +patch pull_requests: +14890 stage: -> patch review pull_request: https://github.com/python/cpython/pull/15156 ___ Python tracker <https://bugs.python.org/issu

[issue37774] Micro-optimize vectorcall using PY_LIKELY

Inada Naoki added the comment: I'm conservative about adding public APIs. But in this case, LIKELY/UNLIKELY macros are battle-tested in the world already. So I think it's safe to make it public. But I'll wait to merge for a week for other core

[issue37587] JSON loads performance improvement for long strings

Inada Naoki added the comment: New changeset 2a570af12ac5e4ac5575a68f8739b31c24d01367 by Inada Naoki in branch 'master': bpo-37587: optimize json.loads (GH-15134) https://github.com/python/cpython/commit/2a570af12ac5e4ac5575a68f8739b3

[issue11165] Document PyEval_Call* functions

Change by Inada Naoki : -- resolution: -> fixed stage: needs patch -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue30671] dict: simplify and improve lookup function

INADA Naoki added the comment: You may misunderstood current lookup loop. In your collisions_count.py, `yield i` must be added right before `while True:` -- ___ Python tracker <http://bugs.python.org/issue30

[issue30671] dict: simplify and improve lookup function

INADA Naoki added the comment: Firstly, do you understand the lookup without perturb? It's documented here: https://github.com/python/cpython/blob/258bfc462b1e58689b43f662a10e44ece3a10bef/Objects/dictobject.c#L161-L182 >>> n = 0; L = [] >>> for i in range(16): ... L.app

[issue30671] dict: simplify and improve lookup function

INADA Naoki added the comment: It seems not "simplify". Please try implement it and show benchmark result if you think it's really worth enough. -- ___ Python tracker <http://bugs.pyt

[issue30686] make `make install` faster

New submission from INADA Naoki: Currently, `make install` doesn't use -jN option of compileall.py My idea is: * Use os.cpu_count() for -j0. * Use the option in Makefile -- components: Build messages: 296216 nosy: inada.naoki priority: normal severity: normal status: open title:

[issue30565] PEP 538: silence locale coercion and compatibility warnings by default?

INADA Naoki added the comment: If this warnings are disabled by default, who enable it? How about just remove them? I'm OK to remove them all. Since it's not ideal, nothing go worse than Python 3.6. Additionally, if PEP 540 is accepted, we can use UTF-8 for stdio and filesystem enc

[issue30686] make `make install` faster

Changes by INADA Naoki : -- pull_requests: +2315 ___ Python tracker <http://bugs.python.org/issue30686> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29304] dict: simplify lookup function

Changes by INADA Naoki : -- pull_requests: +2323 ___ Python tracker <http://bugs.python.org/issue29304> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: pr-2237 only changes lookdict_index() function. The function is used from only dict_popitem(). So it's not performance critical part. And microbench doesn't show significant performance changes: $ ./python.default -m perf timeit -q -l 2000 -s '

[issue15216] Add encoding & errors parameters to TextIOWrapper.reconfigure()

Changes by INADA Naoki : -- pull_requests: +2387 ___ Python tracker <http://bugs.python.org/issue15216> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: New changeset 073ae487b3ff9001c69d530c7555ddaa530dee16 by INADA Naoki in branch 'master': bpo-29304: simplify lookdict_index() function. (GH-2273) https://github.com/python/cpython/commit/073ae487b3ff9001c69d530c7555dd

[issue29304] dict: simplify lookup functions

Changes by INADA Naoki : -- pull_requests: +2402 ___ Python tracker <http://bugs.python.org/issue29304> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: Antonie: Thank you for you comment. Actually speaking, size of instructions are reduced on amd64@gcc. lookdict_unicode_nodummy and lookdict_split doesn't have any reason to have duplicated code anymore. I'll do dedupe for them first. lo

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: Tim Peters: Thanks for your suggestion. But I want to focus on very simple code cleanup without any change to proving algorithm nor memory access pattern in this issue. I'll reply about my thought about reducing collision on t

[issue29304] dict: simplify lookup functions

Changes by INADA Naoki : -- pull_requests: +2455 ___ Python tracker <http://bugs.python.org/issue29304> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: Finally, I removed `freeslot` and all lookdict*() functions are much simpler in GH-2407. Macro bench doesn't show any significant performance change. I'll try to write micro benchmarks. -- ___ Python trac

[issue19325] _osx_support imports many modules

INADA Naoki added the comment: FYI, issue29585 removes _osx_support dependency from site.py -- nosy: +inada.naoki ___ Python tracker <http://bugs.python.org/issue19

[issue29585] site.py imports relatively large `sysconfig` module.

INADA Naoki added the comment: New changeset a8f8d5b4bd30dbe0828550469d98f12d2ebb2ef4 by INADA Naoki in branch 'master': bpo-29585: optimize site.py startup time (GH-136) https://github.com/python/cpython/commit/a8f8d5b4bd30dbe0828550469d98f1

[issue29585] site.py imports relatively large `sysconfig` module.

Changes by INADA Naoki : -- pull_requests: +2530 ___ Python tracker <http://bugs.python.org/issue29585> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29585] site.py imports relatively large `sysconfig` module.

Changes by INADA Naoki : -- pull_requests: +2542 ___ Python tracker <http://bugs.python.org/issue29585> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29585] site.py imports relatively large `sysconfig` module.

INADA Naoki added the comment: New changeset 6b42eb17649bed9615b6e6cecaefdb2f46990b2c by INADA Naoki in branch 'master': bpo-29585: Fix sysconfig.get_config_var("PYTHONFRAMEWORK") (GH-2483) https://github.com/python/cpython/commit/6b42eb17649bed9615b6

[issue29585] site.py imports relatively large `sysconfig` module.

Changes by INADA Naoki : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker <http://bugs.python.org/issue29585> ___

[issue30815] StringIO uses inefficient PyUnicode_AsUCS4

New submission from INADA Naoki: >From PEP393, PyUnicode_AsUCS4 is inefficient. And C implementation of io.StringIO() uses it. That's why Python 3 is slower than Python 2 on logging_format and logging_simple benchmarks. https://mail.python.org/pipermail/speed/2017-February/000503.htm

[issue30815] StringIO uses inefficient PyUnicode_AsUCS4

INADA Naoki added the comment: I'm sorry, it's my mistake. I used vmprof on mac and I thought as_ucs4 is bottleneck. But vmprof on Linux (and perf) shows totally different result. Maybe, current vmprof doesn't work well for native code on macOS. -- resolution: -&g

[issue30815] StringIO uses inefficient PyUnicode_AsUCS4

INADA Naoki added the comment: FYI, https://github.com/python/performance/pull/27 will fix performance regression. Python 3 performance is similar to Python 2 after s/warn/warning/ -- ___ Python tracker <http://bugs.python.org/issue30

[issue30804] bolen-dmg-3.x build-installer.py failed

Changes by INADA Naoki : -- pull_requests: +2586 ___ Python tracker <http://bugs.python.org/issue30804> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue30804] bolen-dmg-3.x build-installer.py failed

INADA Naoki added the comment: Sorry, It seems I failed to merge local patch. -- ___ Python tracker <http://bugs.python.org/issue30804> ___ ___ Python-bugs-list m

[issue28812] Deadlock between GIL and pystate head_mutex.

INADA Naoki added the comment: #30395 is about forking, but this is about multi threading. -- ___ Python tracker <http://bugs.python.org/issue28812> ___ ___ Pytho

[issue28638] Optimize namedtuple creation

INADA Naoki added the comment: I like your idea. Would you make pull request? -- resolution: rejected -> status: closed -> open title: Creating namedtuple is too slow to be used in common stdlib (e.g. functools) -> Optimize namedtuple

[issue28638] Optimize namedtuple creation

INADA Naoki added the comment: I respect Raymond's rejection. But I want to write down why I like Jelle's approach. Currently, functools is the only module which is very popular. But leaving this means every new namedtuple makes startup time about 0.6ms slower. This is also p

[issue28638] Optimize namedtuple creation

INADA Naoki added the comment: I want to focus on pure Python implementation in this issue. While "40x faster" is more 10x faster than "4x faster", C implementation can boost only CPython and makes maintenance more harder. And sometimes "more 10x faster" is not

[issue29225] distutils.command.install_lib.get_outputs() wrong with extensions built inplace

INADA Naoki added the comment: I thought `inplace` option is for debugging/testing without install, not for installing or packaging. As far as document [1], `inplace=1` in `setup.cfg` is noticed as: "which is probably a bad idea for this option, since always building extensions in-

[issue28638] Optimize namedtuple creation

INADA Naoki added the comment: I didn't say "let's not do it". I just want to focus on pure Python implementation at this issue, because this thread is too long already. Feel free to open new issue about C implementation. Even if C implementation is added later, pure Pyt

[issue29225] distutils.command.install_lib.get_outputs() wrong with extensions built inplace

INADA Naoki added the comment: Sorry, I'm not packaging expert. Please wait for expert for right direction. -- ___ Python tracker <http://bugs.python.org/is

[issue29469] AST-level Constant folding

Changes by INADA Naoki : -- pull_requests: +2909 ___ Python tracker <http://bugs.python.org/issue29469> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31043] Tests running twice.

INADA Naoki added the comment: New changeset ede9084476f88f5a86c7eaaac33cdd938e4cce93 by INADA Naoki (Utkarsh Upadhyay) in branch 'master': bpo-31043: fixed test_datetime run twice. (GH-2891) https://github.com/python/cpython/commit/ede9084476f88f5a86c7eaaac33cdd938e4cce93 -

[issue31043] Tests running twice.

Changes by INADA Naoki : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker <http://bugs.python.org/issue31043> ___

[issue30188] test_nntplib: random EOFError in setUpClass()

Changes by INADA Naoki : -- pull_requests: +2944 ___ Python tracker <http://bugs.python.org/issue30188> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue30188] test_nntplib: random EOFError in setUpClass()

INADA Naoki added the comment: New changeset 067931dd95ddfa34297be9c602a796d9f8d9bea3 by INADA Naoki in branch 'master': bpo-30188: fix TypeError in test_nntplib (GH-2892) https://github.com/python/cpython/commit/067931dd95ddfa34297be9c602a796d9f8d9bea3 -- nosy: +i

[issue29585] site.py imports relatively large `sysconfig` module.

INADA Naoki added the comment: https://docs.python.org/3.6/library/site.html#site.USER_SITE > ~/Library/Python/X.Y/lib/python/site-packages for Mac framework builds So it seems I broke sysconfig.get_path('purelib', 'posix_user'). -- ___

[issue29585] site.py imports relatively large `sysconfig` module.

INADA Naoki added the comment: https://github.com/python/cpython/pull/136/files -if sys.platform == 'darwin': -from sysconfig import get_config_var -if get_config_var('PYTHONFRAMEWORK'): -USER_SITE = get_path('pure

[issue29585] site.py imports relatively large `sysconfig` module.

Changes by INADA Naoki : -- pull_requests: +2980 ___ Python tracker <http://bugs.python.org/issue29585> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29585] site.py imports relatively large `sysconfig` module.

INADA Naoki added the comment: New changeset ba9ddb7eea39a651ba7f1ab3eb012e4129c03620 by INADA Naoki in branch 'master': bpo-29585: fix test fail on macOS Framework build (GH-2928) https://github.com/python/cpython/commit/ba9ddb7eea39a651ba7f1ab3eb012e

[issue29585] site.py imports relatively large `sysconfig` module.

Changes by INADA Naoki : -- resolution: -> fixed stage: needs patch -> resolved status: open -> closed ___ Python tracker <http://bugs.python.or

[issue29519] weakref spewing exceptions during finalization when combined with multiprocessing

Changes by INADA Naoki : -- pull_requests: +3006 ___ Python tracker <http://bugs.python.org/issue29519> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue29519] weakref spewing exceptions during finalization when combined with multiprocessing

INADA Naoki added the comment: [gsutil](https://github.com/GoogleCloudPlatform/gsutil) shows massive "Exception ... ignored" message from this regression. https://twitter.com/minimum2scp/status/890021408482316289 I think we should backport this fix to Python 2.7.14. -

[issue31085] Add option for namedtuple to name its result type automatically

INADA Naoki added the comment: When subclassing, current __repr__ uses `self.__class__.__name__`. So you get meaningful name already. When automatic generation, I recommend you to use some wrapper to cache same namedtuple, since creating namedtuple on the fly is costly job. I'm a

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

New submission from INADA Naoki: like GH-2966, most types with Py_TPFLAGS_HAVE_GC should call PyObject_GC_UnTrack() at top of the tp_dealloc. For example, I found lru_cache doesn't call it and I can produce segmentation fault. I'll check other types too. -- components:

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: # collection module dequeiter_dealloc doesn't call Untrack(), but it's safe because it only frees deque and deque_dealloc calls Untrack() dequeiter_dealloc(dequeiterobject *dio) { Py_XDECREF(dio->deque); defdict_dealloc

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: > dequeiter_dealloc doesn't call Untrack(), but it's safe because it only frees > deque > and deque_dealloc calls Untrack() It may be not true, while I don't have exploit yet. -- _

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: # _json module scanner_dealloc() encoder_dealloc() # _struct module unpackiter_dealloc # _ssl module context_dealloc() # Objects/ setiter_dealloc dictiter_dealloc dictview_dealloc # Parser/ ast_dealloc

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: > should the base method which calls tp_dealloc do this? Maybe can kill all > birds with one stone. It may be possible, but unclear. Object finalization process is very complicated. I agree that UnTrack object as soon as refcnt=0, and Track onl

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: Docs/extending/newtypes.rst and Docs/include/noddy3.c should be updated too. But I'm not good English writer. I need help. -- ___ Python tracker <http://bugs.python.org/is

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

Changes by INADA Naoki : -- pull_requests: +3019 ___ Python tracker <http://bugs.python.org/issue31095> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31061] asyncio segfault when using threadpool and "_asyncio" native module

INADA Naoki added the comment: New changeset de34cbe9cdaaf7b85fed86f99c2fd071e1a7b1d2 by INADA Naoki (Alexander Mohr) in branch 'master': bpo-31061: fix crash in asyncio speedup module (GH-2966) https://github.com/python/cpython/commit/de34cbe9cdaaf7b85fed86f99c2fd0

[issue31061] asyncio segfault when using threadpool and "_asyncio" native module

Changes by INADA Naoki : -- pull_requests: +3023 ___ Python tracker <http://bugs.python.org/issue31061> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31061] asyncio segfault when using threadpool and "_asyncio" native module

INADA Naoki added the comment: New changeset f142e85d22ba135d5205280240f3a2fe1df2649f by INADA Naoki in branch '3.6': bpo-31061: fix crash in asyncio speedup module (GH-2984) https://github.com/python/cpython/commit/f142e85d22ba135d5205280240f3a2

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: # microbench https://gist.github.com/methane/db16224a99d12ad4ed170cf3bd45b819 Slower (2): - fromkeys_str: 3.85 us +- 0.07 us -> 4.09 us +- 0.09 us: 1.06x slower (+6%) - fromkeys_int: 4.24 us +- 0.13 us -> 4.48 us +- 0.16 us: 1.06x slower (+6%) Fas

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: New changeset 778928b0c7aa438c282727535814d73df850693a by INADA Naoki in branch 'master': bpo-29304: Simplify dict lookup functions (GH-2407) https://github.com/python/cpython/commit/778928b0c7aa438c282727535814d7

[issue29304] dict: simplify lookup functions

Changes by INADA Naoki : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker <http://bugs.python.org/issue29304> ___

[issue29304] dict: simplify lookup functions

INADA Naoki added the comment: On i386 docker image, pyperformance: ./python -m perf compare_to before.json after.json -G --min-speed=2 Slower (4): - regex_dna: 279 ms +- 1 ms -> 288 ms +- 3 ms: 1.03x slower (+3%) - deltablue: 17.9 ms +- 0.3 ms -> 18.3 ms +- 1.0 ms: 1.03x slow

[issue31166] null pointer deref and segfault in _PyObject_Alloc (obmalloc.c:1258)

INADA Naoki added the comment: I think it's false positive of ASAN. We have dynamically sized block. https://github.com/python/cpython/blob/3b0f620c1a2a21272a9e2aeca6ca1d1ac10f8162/Objects/dict-common.h#L49-L69 dictobject.c:547 calls memcpy to fill the block and head pointer is defin

[issue31166] null pointer deref and segfault in _PyObject_Alloc (obmalloc.c:1258)

INADA Naoki added the comment: As Benjamin commented, this is caused by mutating internal dict. PyType_Lookup() use "method cache", based on "tp_version_tag" in the type object. When you modify internal dict directly, namespace is changed without invalidating tp_version_ta

[issue31166] null pointer deref and segfault in _PyObject_Alloc (obmalloc.c:1258)

INADA Naoki added the comment: But we should check dicts of all parents. It will has significant penalty, especially for classes having long mro (inheriting metaclass from typing module cause long mro). -- ___ Python tracker <http://bugs.python.

[issue31179] Speed-up dict.copy() up to 5.5 times.

INADA Naoki added the comment: I like idea. One worrying point is how to deal with dirty dict. How about do it only when ma_used == keys->dk_nentries? Slightly off topic. Copy on write can be implemented via dk_refcnt. Functions just passing `**kwargs` has temporal copy of dict. And CoW w

[issue31149] Add Japanese to the language switcher

INADA Naoki added the comment: My preference is "Japanese". After selecting drop down list, we can type "jap" to select Japanese. On the other hand, we should use IME to input "日本語". It's not so easy as input alphabets. -- ___

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: > Maybe it would help to have a short comment, maybe with a link to this issue, > on each PyObject_GC_UnTrack(). Is this looks good to you? /* UnTrack is needed before calling any callbacks (bpo-31095) */ PyObject_GC_UnTrac

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: BTW, should this backported to Python 3.5? -- nosy: +larry ___ Python tracker <http://bugs.python.org/issue31095> ___ ___ Python-bug

[issue31265] Remove doubly-linked list from C OrderedDict

New submission from INADA Naoki: Since dict preserves insertion order, doubly linked list in OrderedDict can be removed. There is small performance improvement for odict creation: $ curl https://api.github.com/orgs/python/repos > repos.json $ ./py-patched -m perf timeit --compare-to `pwd`

[issue31265] Remove doubly-linked list from C OrderedDict

Changes by INADA Naoki : -- pull_requests: +3232 ___ Python tracker <http://bugs.python.org/issue31265> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: There are some failing tests remaining, and I want to discuss about some of them here. Traceback (most recent call last): File "/home/inada-n/work/python/nodebug/Lib/test/test_ordered_dict.py", line 261, in test_pop self.assertEqual(m.pop('a&

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: I think typical usage is: get, set (incl, creating), and iterate. * get: there is no difference * set: inserting new key is little faster since no updating linked list * iterate: in typical case, new odict is faster because current odict iterator do lookup for

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: od.move_to_end() is slower too: $ ./py-patched -m perf timeit --compare-to `pwd`/py-default -s 'from collections import OrderedDict as odict; od = odict.fromkeys("abcdefghijklmnopqrstuvwxyz")' -- 'od.move_to_end("a"

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: Another idea is making "dict preserves insertion order" as language spec. There are still some difference between dict and odict: .move_to_end() and __eq__. But most use cases of odict is just keep insertion order. For example, parsing config file, js

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: When no change happens: $ ./py-patched -m perf timeit --compare-to `pwd`/py-default -s 'from collections import Ordedict; od = odict.fromkeys("abcdefghijklmnopqrstuvwxyz")' -- &#

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: New changeset a6296d34a478b4f697ea9db798146195075d496c by INADA Naoki in branch 'master': bpo-31095: fix potential crash during GC (GH-2974) https://github.com/python/cpython/commit/a6296d34a478b4f697ea9db7981461

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

Changes by INADA Naoki : -- pull_requests: +3234 ___ Python tracker <http://bugs.python.org/issue31095> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

Changes by INADA Naoki : -- pull_requests: +3235 ___ Python tracker <http://bugs.python.org/issue31095> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

Changes by INADA Naoki : -- pull_requests: +3236 ___ Python tracker <http://bugs.python.org/issue31095> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: I opened backport PR for 3.6, 2.7 and 3.5. -- versions: +Python 2.7, Python 3.5 ___ Python tracker <http://bugs.python.org/issue31

[issue31265] Remove doubly-linked list from C OrderedDict

Changes by INADA Naoki : -- nosy: +eric.snow ___ Python tracker <http://bugs.python.org/issue31265> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: New changeset 2eea952b1b9ebbc2d94fd3faca1536c6b4963725 by INADA Naoki in branch '3.6': bpo-31095: fix potential crash during GC (GH-3195) https://github.com/python/cpython/commit/2eea952b1b9ebbc2d94fd3faca1536

[issue31095] Checking all tp_dealloc with Py_TPFLAGS_HAVE_GC

INADA Naoki added the comment: New changeset 4cde4bdcc86cb08ee3847500a172cc24eba37ffe by INADA Naoki in branch '2.7': bpo-31095: Fix potential crash during GC (GH-3197) https://github.com/python/cpython/commit/4cde4bdcc86cb08ee3847500a172cc

[issue31265] Remove doubly-linked list from C OrderedDict

INADA Naoki added the comment: FYI, my approach is not original, but copied from PyPy. https://bitbucket.org/pypy/pypy/src/3e52029e9a5a677f7de62ef49f36090465cfbf4c/rpython/rtyper/lltypesystem/rordereddict.py?at=default&fileviewer=file-view-default#rordereddict.py-1437:

[issue31333] Implement ABCMeta in C

INADA Naoki added the comment: > Hm, indeed, but I see that module 'abc' is in 'sys.modules', probably it is > then only used by 'collections.abc'/'_collections_abc'. This means that the > influence of 'ABCMeta' on interpreter start-up

[issue31336] Speed up _PyType_Lookup() for class creation

INADA Naoki added the comment: There is special internal API in dictobject.c _PyDict_LoadGlobal(PyDictObject *globals, PyDictObject *builtins, PyObject *key) Maybe, we can have special API like that _PyDict_LookupMro(PyObject *mro, PyObject *key); BTW, method cache in _PyType_Lookup is not

[issue31333] Implement ABCMeta in C

INADA Naoki added the comment: "python_startup_nosite" benchmark imports io module, and "python_startup" benchmark imports os too. So no need to additional test for them. You can see it with `python -v -S -c 'pass&

[issue31336] Speed up _PyType_Lookup() for class creation

INADA Naoki added the comment: Confirmed: $ ./python-patched -m perf timeit --compare-to `pwd`/python -- 'class C: pass' python: . 11.9 us +- 0.1 us python-patched: . 10.3 us +- 0.1 us Mean +- std dev: [python] 11.9 us +- 0.1 us -> [python-

<    7   8   9   10   11   12   13   14   15   16   >