[issue36984] typing docs "versionadded" is inaccurate for many attributes

New submission from Anthony Sottile : expanding on https://bugs.python.org/issue36983 the docs are also a bit out of date in places I'm not sure how to document something that appeared in two versions, but I'll leave that part to review. Using data generated / collected in https

[issue36983] typing.__all__ has drifted from actual contents

Anthony Sottile added the comment: using the same heuristic as the test uses, here's what's been missing from __all__ historically (note: I _manually_ skipped `Final` for 3.5.0-3) ===(3, 5, 0)=== +FrozenSet +SupportsBytes

[issue36983] typing.__all__ has drifted from actual contents

Change by Anthony Sottile : -- keywords: +patch pull_requests: +13365 stage: -> patch review ___ Python tracker <https://bugs.python.org/issue36983> ___ _

[issue36983] typing.__all__ has drifted from actual contents

New submission from Anthony Sottile : notably it is missing ForwardRef, OrderedDict, ChainMap in python3.8 it is missing others in other versions I'm going to attempt to write a test which should classify things that should belong there -- components: Library (Lib) messages: 342975

[issue35894] Apparent regression in 3.8-dev: 'TypeError: required field "type_ignores" missing from Module'

Anthony Sottile added the comment: there's other optional fields in the ast, type ignores don't seem essential to the `Module`, could those be made optional as well? -- nosy: +Anthony Sottile ___ Python tracker <https://bugs.python.org/issue35

[issue36919] Exception from 'compile' reports a newline char not present in input

Anthony Sottile added the comment: still not able to answer the why, but at least I can answer the what here: https://github.com/python/cpython/blob/f665b96e92a6a6943e312e2c606f348db95939ab/Parser/tokenizer.c#L984-L987 cpython adds a newline during tokenization if the file does not end

[issue2180] tokenize: mishandles line joining

Change by Anthony Sottile : -- keywords: +patch pull_requests: +13312 stage: needs patch -> patch review ___ Python tracker <https://bugs.python.org/iss

[issue36917] ast.NodeVisitor no longer calls visit_Str

Anthony Sottile added the comment: > You can not use the same implementation of the visitor for Num, Str, > NameConstant and Ellipsis, because all these classes use different attribute > for saving the value ah yes, this is true -- maybe the better change would be to just add `

[issue36917] ast.NodeVisitor no longer calls visit_Str

Anthony Sottile added the comment: The simplest case is just the addition of an `isinstance` check: https://github.com/asottile/dead/blob/85f5edbb84b5e118beab4be3346a630e41418a02/dead.py#L165-L170 class V(ast.NodeVisitor): def visit_Str(self, node): ... def visit_Constant

[issue36917] ast.NodeVisitor no longer calls visit_Str

Anthony Sottile added the comment: spent some more time thinking about this and I think we should strongly consider reverting. simplification in the core interpreter should not be weighed lightly against complexity and breaking changes for users. the change is also unfortunate because

[issue36917] ast.NodeVisitor no longer calls visit_Str

Anthony Sottile added the comment: There would still be a breakage for that if someone was defining py36+ `visit_Constant` (which would clobber the `ast.NodeVisitor.visit_Constant` if we were to add it) -- ___ Python tracker <ht

[issue32892] Remove specific constant AST types in favor of ast.Constant

Anthony Sottile added the comment: hitting this in https://bugs.python.org/issue36917? Is the simplification here really worth the breaking change to consumers? I now have to write something that's essentially this to work around this which feels more like the complexity has just been

[issue36917] ast.NodeVisitor no longer calls visit_Str

Anthony Sottile added the comment: wrong bpo, this is the right one: https://bugs.python.org/issue32892 -- ___ Python tracker <https://bugs.python.org/issue36

[issue36917] ast.NodeVisitor no longer calls visit_Str

New submission from Anthony Sottile : More fallout from the Constant change in https://bugs.python.org/issue32892 minimal reproduction: import ast class V(ast.NodeVisitor): def visit_Str(self, node): print(node.s) def main(): V().visit(ast.parse('x = "hi"')) i

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: If you can write a test similar to the AnotherLeak.test_callback test case, and commit that first. It will show in the CI/CD log as failed and verify the issue (incase it comes up in PR review) Then add another commit with the patch itself and we should see

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: Thanks Eryk that saved a lot of debugging. dgelessus - if you want to write a patch for CPython am happy to take you through this and get it over the line. Else: am Happy to write a test against the gc counter and a patch

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: In the stack it's calling none_dealloc() which should never happen. Assume this is being triggered by ctypes PyCFuncPtr_call. The stacktrace I'm getting comes after the double decref so it's not showing the root cause. Someone who knows ctypes better might

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: Full trace for reference: (lldb) bt all * thread #1, queue = 'com.apple.main-thread', stop reason = signal SIGABRT * frame #0: 0x7fff5984f2c6 libsystem_kernel.dylib`__pthread_kill + 10 frame #1: 0x7fff59904bf1 libsystem_pthread.dylib`pthread_kill

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: >From lldb (lldb) run ~/cpython/test_gc_ctypes.py Process 20059 launched: '/Users/anthonyshaw/CLionProjects/cpython/python.exe' (x86_64) Fatal Python error: deallocating None Current thread 0x0001005c85c0 (most recent call first): File &quo

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: The documentation says: >> Note Make sure you keep references to CFUNCTYPE() objects as long as they are used from C code. ctypes doesn’t, and if you don’t, they may be garbage collected, crashing your program when a callback is made. Also

[issue36880] Returning None from a callback with restype py_object decrements None's refcount too much

anthony shaw added the comment: Thanks, I'll check this out -- assignee: -> anthonypjshaw nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issu

[issue36882] 在ctypes里调用C库返回c_char_p类型时的问题

anthony shaw added the comment: "When the C library is called in ctypes to return the c_char_p type, the problem that only the string before \0 can be obtained when the string contains \\0" This function is specifically for null-terminated strings (\0), please could you be mor

[issue36883] 在ctypes里调用C库返回c_char_p类型时的问题

anthony shaw added the comment: Closing as duplicate of 36882 -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue36883> ___ ___ Python-bug

[issue36883] 在ctypes里调用C库返回c_char_p类型时的问题

Change by anthony shaw : -- stage: -> resolved status: open -> closed ___ Python tracker <https://bugs.python.org/issue36883> ___ ___ Python-bugs-list

[issue25541] Wrong usage of sockaddr_un struct for abstract namespace unix sockets

anthony shaw added the comment: The existing tests in place add the null-termination bytes in the test string: def testLinuxAbstractNamespace(self): address = b"\x00python-test-hello\x00\xff" with socket.socket(socket.AF_UNIX, socket.SOCK_STREAM) as s1:

[issue25541] Wrong usage of sockaddr_un struct for abstract namespace unix sockets

Change by anthony shaw : -- versions: +Python 3.8 -Python 2.7, Python 3.4, Python 3.7 ___ Python tracker <https://bugs.python.org/issue25541> ___ ___ Python-bug

[issue25541] Wrong usage of sockaddr_un struct for abstract namespace unix sockets

anthony shaw added the comment: thanks, your code example zero-pads the socket address, and looking at the socketmodule.c code it does some padding under certain circumstances. https://github.com/python/cpython/blob/master/Modules/socketmodule.c#L1318-L1330 The Unix man page specify the same

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

Change by anthony shaw : -- resolution: -> rejected stage: patch review -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue35668] Improve test coverage for idlelib

Change by anthony shaw : -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue35668> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue36500] Add "regen-*" equivalent projects for Windows builds

Change by anthony shaw : -- assignee: -> anthonypjshaw nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue36500> ___ ___ Python-bugs-lis

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

Change by anthony shaw : -- assignee: -> anthonypjshaw nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue36551> ___ ___ Python-bugs-lis

[issue36552] Replace OverflowError with ValueError when calculating length of range objects > PY_SIZE_MAX

Change by anthony shaw : -- assignee: -> anthonypjshaw nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue36552> ___ ___ Python-bugs-lis

[issue36862] Add Visual Studio 2019 support to PCBuild/env.bat

Change by anthony shaw : -- nosy: +steve.dower stage: patch review -> ___ Python tracker <https://bugs.python.org/issue36862> ___ ___ Python-bugs-list mai

[issue36862] Add Visual Studio 2019 support to PCBuild/env.bat

Change by anthony shaw : -- keywords: +patch pull_requests: +13126 stage: -> patch review ___ Python tracker <https://bugs.python.org/issue36862> ___ ___ Py

[issue36862] Add Visual Studio 2019 support to PCBuild/env.bat

New submission from anthony shaw : The PCBuild tool should support VS2019 but default to 2017 as per documentation. PR to follow. -- assignee: anthonypjshaw components: Build messages: 341963 nosy: anthonypjshaw priority: normal severity: normal status: open title: Add Visual Studio

[issue24263] unittest cannot load module whose name starts with Unicode

anthony shaw added the comment: thanks, will wait for a review from Serhiy, Rbcollins or ezio -- ___ Python tracker <https://bugs.python.org/issue24263> ___ ___

[issue12188] PEP 7 (or guide) add C style policies and explanation

anthony shaw added the comment: Hi, This discussion came to a stop, but it doesn't seem conclusive. PEP discussions are now in GitHub on https://github.com/python/peps/issues so I'm going to close this BPO issue. There is no additional section in PEP 7 for this level of detail

[issue36844] abiflag `m` is no longer showing when compiled with --enable-shared

Anthony Sottile added the comment: seems intentional => closing -- stage: -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue36844] abiflag `m` is no longer showing when compiled with --enable-shared

Anthony Sottile added the comment: +vstinner Hmmm, actually the relevant commit appears to be 6c44fde3e03079e0c69f823dafbe04af50b5bd0d and intentional (I ran a git bisect to find this!) https://github.com/python/cpython/pull/12931 Does PEP 3149 require an update with this change, it still

[issue36844] abiflag `m` is no longer showing when compiled with --enable-shared

New submission from Anthony Sottile : This appears to be a regression between 3.8a3 and 3.8a4 -- though it may be intentional and I'm missing something? I noticed this while packaging 3.8 for deadsnakes https://github.com/deadsnakes/python3.8 I've created a minimal reproduction: $ git

[issue36814] posix_spawn explicit file_actions=None throws error

anthony shaw added the comment: After patch: Python 3.8.0a3+ (heads/31968-dirty:c664b342a4, May 6 2019, 18:06:21) [Clang 10.0.1 (clang-1001.0.46.4)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>>

[issue36814] posix_spawn explicit file_actions=None throws error

anthony shaw added the comment: Issue is in parse_file_actions parse_file_actions(PyObject *file_actions, posix_spawn_file_actions_t *file_actionsp, PyObject *temp_buffer) { PyObject *seq; PyObject *file_action = NULL; PyObject *tag_obj

[issue36814] posix_spawn explicit file_actions=None throws error

anthony shaw added the comment: Raised a fix in GH-13144 -- stage: patch review -> ___ Python tracker <https://bugs.python.org/issue36814> ___ ___ Python-

[issue36814] posix_spawn explicit file_actions=None throws error

Change by anthony shaw : -- keywords: +patch pull_requests: +13057 stage: -> patch review ___ Python tracker <https://bugs.python.org/issue36814> ___ ___ Py

[issue36814] posix_spawn explicit file_actions=None throws error

anthony shaw added the comment: Verified on master Python 3.8.0a3+ (heads/bpo-28367:373c7aa098, May 6 2019, 17:34:39) [Clang 10.0.1 (clang-1001.0.46.4)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>>

[issue36819] Crash during encoding using UTF-16/32 and custom error handler

anthony shaw added the comment: Easily reproduced on master, thanks (lldb) run encode_crash.py Process 14743 launched: '/Users/anthonyshaw/repo/cpython/python.exe' (x86_64) Objects/unicodeobject.c:448: _PyUnicode_CheckConsistency: Assertion "((PyObject*)(op))->ob_type))->tp_fl

[issue36821] Termios module largely untested

anthony shaw added the comment: This could be a good issue for the PyCon sprints, otherwise I'm happy to implement it -- ___ Python tracker <https://bugs.python.org/issue36

[issue36821] Termios module largely untested

New submission from anthony shaw : I noticed that the termios.c module is largely untested. There is some coverage via test_pty, test_ioctl and test_getpass, but there is nothing to cover regression and the behaviours in the module functions. Tests are required for: - termios.tcgetattr

[issue28367] Add more standard baud rate constants to "termios"

anthony shaw added the comment: Converted the original patch as a PR GH-13142 this seems like a good idea and an easy change -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue28

[issue28367] Add more standard baud rate constants to "termios"

Change by anthony shaw : -- pull_requests: +13055 ___ Python tracker <https://bugs.python.org/issue28367> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue27534] IDLE: Reduce number and time for user process imports

anthony shaw added the comment: It would be great for this issue to be revisited, there has been some further interest from users. -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue27

[issue25251] Unknown MS Compiler version 1900

anthony shaw added the comment: Closing as 3rd party feature for setuptools -- nosy: +anthonypjshaw resolution: -> third party stage: -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue28775] Option to set startup directory in IDLE

anthony shaw added the comment: Hi Nofar are you still interested on working on this request? -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue28

[issue31968] exec(): method's default arguments from dict-inherited globals

anthony shaw added the comment: Added a PR for the documentation clarification. -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue31

[issue31968] exec(): method's default arguments from dict-inherited globals

Change by anthony shaw : -- pull_requests: +13053 ___ Python tracker <https://bugs.python.org/issue31968> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue24939] Remove unicode_format.h from stringlib

anthony shaw added the comment: The code is mostly: FieldNameIterator * related functions FormatterIterator * related functions MarkupIterator * related functions There are a few other utility methods in there as well -- ___ Python tracker

[issue34368] ftplib __init__ function can't handle 120 or 4xy reply when connect to the server

Change by anthony shaw : -- nosy: +giampaolo.rodola ___ Python tracker <https://bugs.python.org/issue34368> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue32971] Docs on unittest.TestCase.assertRaises() should clarify context manager details

Change by anthony shaw : -- stage: patch review -> resolved status: open -> closed ___ Python tracker <https://bugs.python.org/issue32971> ___ ___ Pyth

[issue32971] Docs on unittest.TestCase.assertRaises() should clarify context manager details

Change by anthony shaw : -- pull_requests: -9605 ___ Python tracker <https://bugs.python.org/issue32971> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue25541] Wrong usage of sockaddr_un struct for abstract namespace unix sockets

anthony shaw added the comment: hi, which version of Python were you using to do this? Please could you provide the full code snippet to reproduce the issue. The following example binds to the correct namespace from socket import * sock = socket(AF_UNIX, SOCK_STREAM) sock.bind(&qu

[issue33971] os.mknod is subject to "umask"

anthony shaw added the comment: NB: This issue would be good for PyCon sprints -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue33

[issue33971] os.mknod is subject to "umask"

Change by anthony shaw : -- assignee: docs@python -> Mariatta nosy: +Mariatta ___ Python tracker <https://bugs.python.org/issue33971> ___ ___ Python-bugs-lis

[issue24939] Remove unicode_format.h from stringlib

anthony shaw added the comment: > The things that are in this file but are unrelated to unicodeobject.c are the support routines for implementing string.Formatter. I'm not sure which functions that relates to, if you could let me know I'd be happy to add those to the

[issue24939] Remove unicode_format.h from stringlib

Change by anthony shaw : -- keywords: +patch pull_requests: +13050 stage: -> patch review ___ Python tracker <https://bugs.python.org/issue24939> ___ ___ Py

[issue24939] Remove unicode_format.h from stringlib

anthony shaw added the comment: Eric, there have been further changes to Objects/stringlib/unicode_format.h since this original note, I've raised a PR with the intent of your note from 2015. There also hasn't been any change to the situation, unicode_format.h is only used

[issue25597] unittest.mock does not wrap dunder methods (__getitem__ etc)

anthony shaw added the comment: The assertions in the attached test still fail on master (3.8a3), so this still applies. Michael, are you able to look at this, the code hasn't changed since the original PEP417 implementation, which doesn't specify if this behaviour should be supported

[issue13582] IDLE and pythonw.exe stderr problem

anthony shaw added the comment: Revisiting this issue as it's still open. The original patch was for 3.3a0, there have been many changes to both IDLE and the Windows build since, including the recent IDLE entry point in the new Windows installer. Steve, Terry, please could you review

[issue26278] BaseTransport.close() does not trigger connection_lost()

anthony shaw added the comment: This issue was never responded to, are you still having this issue? Which version of CPython are you using and can you please provide steps to reproduce the problem. -- nosy: +anthonypjshaw ___ Python tracker

[issue28742] argparse.ArgumentDefaultsHelpFormatter sometimes provides inaccurate documentation of defaults, so they should be overrideable

anthony shaw added the comment: Hi Peter, this proposal would be easier to review as a GitHub Pull Request. -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue28

[issue34946] inspect.getcallargs is marked as deprecated in documentation, but doesn't issue a DeprecationWarning when used

anthony shaw added the comment: Submitted PR to add the deprecation warning, since this initial discussion, some other functions also raise DeprecationWarning's in code. -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.

[issue34946] inspect.getcallargs is marked as deprecated in documentation, but doesn't issue a DeprecationWarning when used

Change by anthony shaw : -- keywords: +patch pull_requests: +13040 stage: -> patch review ___ Python tracker <https://bugs.python.org/issue34946> ___ ___ Py

[issue31603] Please add argument to override stdin/out/err in the input builtin

anthony shaw added the comment: The discussion on python-ideas has some unresolved questions. Wren, did you get a definitive answer on this idea? -- nosy: +anthonypjshaw ___ Python tracker <https://bugs.python.org/issue31

[issue5243] Missing dependency in distutils build

Change by anthony shaw : -- nosy: +dstufft, ncoghlan ___ Python tracker <https://bugs.python.org/issue5243> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue28540] math.degrees(sys.float_info.max) should throw an OverflowError exception

anthony shaw added the comment: Francisco, I recommend converting the patch into a GitHub pull request to make it easier to code review. Mark, Raymond, please could you re-review this patch. -- nosy: +anthonypjshaw, rhettinger ___ Python tracker

[issue26303] Shared execution context between doctests in a module

Change by anthony shaw : -- assignee: docs@python -> Mariatta ___ Python tracker <https://bugs.python.org/issue26303> ___ ___ Python-bugs-list mailing list Un

[issue26303] Shared execution context between doctests in a module

anthony shaw added the comment: this issue would be good for pycon sprints -- nosy: +Mariatta, anthonypjshaw ___ Python tracker <https://bugs.python.org/issue26

[issue5243] Missing dependency in distutils build

anthony shaw added the comment: This issue has been open for some time, looking at the install_lib in master, there have been no changes to call 'build_clib' so the issue documented here would still apply. -- nosy: +anthonypjshaw ___ Python

[issue31813] python -m ensurepip hangs

anthony shaw added the comment: I believe the issue is caused by the issue referenced in the last comment. Closing issue -- nosy: +anthonypjshaw resolution: -> not a bug stage: -> resolved status: open -> closed ___ Python tracke

[issue34820] binascii.c:1578:1: error: the control flow of function ‘binascii_crc32’ does not match its profile data (counter ‘arcs’)

anthony shaw added the comment: There is not enough information to triage this issue and since the last request there has been no reply. Closing issue. -- nosy: +anthonypjshaw stage: -> resolved status: open -> closed ___ Python tracker

[issue24263] unittest cannot load module whose name starts with Unicode

anthony shaw added the comment: The original PR refers to a branch that no longer exists, but the behaviour documented still applies to master. There were some changes to the test loader, but none that fixed this issue. -- nosy: +anthonypjshaw

[issue15987] Provide a way to compare AST nodes for equality recursively

anthony shaw added the comment: Closing issue, PR branch has since been removed and targets Python 3.4 -- nosy: +anthonypjshaw resolution: -> out of date stage: test needed -> resolved status: open -> closed ___ Python tracke

[issue15987] Provide a way to compare AST nodes for equality recursively

anthony shaw added the comment: This discussion is inconclusive and targets an old version of CPython, can this issue be closed? -- nosy: +anthony shaw ___ Python tracker <https://bugs.python.org/issue15

[issue31841] Several methods of collections.UserString do not return instances of UserString or its subclasses

anthony shaw added the comment: This issue has been open for some time and the PR references an upstream branch that no longer exists. Can this request be closed, or has the conversation yet to be resolved? -- nosy: +anthony shaw ___ Python

[issue32430] Simplify Modules/Setup{,.dist,.local}

anthony shaw added the comment: PR GH-5062 is still open in GitHub, is this PR still required? -- nosy: +anthony shaw ___ Python tracker <https://bugs.python.org/issue32

[issue31372] Add SSLSocket.get_verify_result()

anthony shaw added the comment: Please review the PR again, some of the code no longer applies as it targets master and the PR is quite old. -- nosy: +anthony shaw ___ Python tracker <https://bugs.python.org/issue31

[issue34214] Pylint recusion stack overflow abort

anthony shaw added the comment: I agree with Jerome, that handling recursion errors is a part of a 3rd party package and not a bug with CPython. -- nosy: +anthony shaw ___ Python tracker <https://bugs.python.org/issue34

[issue35534] SIGSEGV in stackdepth_walk

anthony shaw added the comment: stackdepth_walk no longer exists in Python 3, this is a 2.7 specific issue. opcode_stack_effect in compile.c will have the expected stack effect. If there is a mismatch between the Opcodes compiled with the distribution and the ones in the compile.c

[issue36782] Add tests for the datetime C API

Change by anthony shaw : -- nosy: +edison.abahurire ___ Python tracker <https://bugs.python.org/issue36782> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue36782] Add tests for the datetime C API

anthony shaw added the comment: Reserving this issue (in mentored sprint) -- nosy: +anthony shaw ___ Python tracker <https://bugs.python.org/issue36782> ___ ___

[issue34850] Emit a syntax warning for "is" with a literal

Anthony Sottile added the comment: Should this also produce warnings for `list` / `dict` / `set` literals? ``` $ python3.8 Python 3.8.0a3 (default, Mar 27 2019, 03:46:44) [GCC 7.3.0] on linux Type "help", "copyright", "credits" or "license" for m

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: >I am -1 for this optimization because it affects only one particular case >(neither other kinds of comprehensions, nor generator expressions, nor list >comprehensions with conditions) and even in this case it is small. It is possible to add a lot

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: Have just optimized some of the code and pushed another change as 69dce1c552. ran both master and 69dce1c552 using pyperformance with PGO: ➜ ~ python3.8 -m perf compare_to master.json 69dce1c552.json --table

[issue36478] backport of pickle fixes to Python 3.5.7 uses C99 for loops

Anthony Sottile added the comment: It's a rather oldish gcc in this case. (4.8.4 + whatever ubuntu patches). Here's a full (successful) build log (including the combinations of flags) after patching: https://launchpadlibrarian.net/416985438/buildlog_ubuntu-trusty-amd64.python3.5_3.5.7-1

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > If your patch uses __length_hint__, it is bug. I’m not sure I understand this comment, PEP424 says “This is useful for presizing containers when building from an iterable.“ This patch uses __length_hint__ to presize the list container for a l

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > If your patch uses __length_hint__, it is bug. iterator will return 2**1000 for __length_hint__, but produce no item on iteration. It raises an OverflowError because of the goto https://github.com/python/cpython/pull/12718/files#d

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > In such case, current behavior works. And your patch will raise > OverflowError. Try [x for x in range(2**1000)] in a REPL. It doesn’t raise anything, it tries to create a list that will eventually exceed PY_SIZE_MAX, but it only crashes once it r

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > This patch makes it slow for small iterators That is a one-off cost for the __length_hint__ of the range object specifically. Objects with a known length (lists, sets, tuples) would not have that overhead. I can run a more useful set of benchmarks agai

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > This might cause a MemoryError when the __length_hint__ of the source returns > a too large value, even when the actual size of the comprehension is smaller, > e.g.: The current implementation of list comprehensions raise neither a memoryerror or

[issue36551] Optimize list comprehensions with preallocate size and protect against overflow

anthony shaw added the comment: > How about converting `[x for x in it]` to `[*it]` in AST? I should have been more explicit, this patch improves the performance of all list comprehensions that don’t have an if clause. Not just [x for x in y] but: d = {} # some sort of dictionary [f

[issue35488] pathlib Path.match does not behave as described

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

<    1   2   3   4   5   6   7   8   >