[issue42339] official embedded Python fails to import certain modules

Change by Christian Heimes : -- assignee: -> steve.dower components: +Windows nosy: +paul.moore, steve.dower, tim.golden, zach.ware ___ Python tracker ___

[issue42341] xml.dom.minidom parsing omits Text nodes in top level

New submission from Mike Frysinger : $ python3 Python 3.8.5 (default, Aug 2 2020, 15:09:07) [GCC 9.3.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>> from xml.dom import minidom # Lets parse a simple XML file with comment & text nodes in the top level.

[issue23436] xml.dom.minidom.Element.ownerDocument is hidden

Change by Mike Frysinger : -- title: xml.dom.minidom.Element.ownerDocument is hiden -> xml.dom.minidom.Element.ownerDocument is hidden ___ Python tracker ___

[issue42253] xml.dom.minidom.rst missing standalone documentation

Change by Mike Frysinger : -- assignee: -> docs@python components: +Documentation nosy: +docs@python title: xml.dom.minidom.rst missed informations -> xml.dom.minidom.rst missing standalone documentation ___ Python tracker

[issue42334] Subclassing int and complex with keyword arguments weird

Dennis Sweeney added the comment: Here's an example: class A(complex): def __init__(self, *args, test): self.test = test def __new__(cls, *args, test): return super().__new__(cls, *args) >>> a = A(1, test="TEST") >>> a (1+0j) >>> a.test 'TEST' >>> b = A(1, 1,

[issue42334] Subclassing int and complex with keyword arguments weird

Dennis Sweeney added the comment: This is because int, str, and complex are immutable. If I have class MyInt(int): def __init__(self, stuff): pass then when I call MyInt("19"), the string "19" is passed to the constructor int.__new__ before the overridden

[issue42141] Speedup various dict inits

Change by Inada Naoki : -- resolution: -> rejected ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue42296] Infinite loop uninterruptable on Windows in 3.10

Eryk Sun added the comment: Note that this issue only applies to a process that has a single Python thread. With multiple Python threads, the signal handler gets called when the main thread acquires the GIL. As far as Windows console-driven SIGINT and SIGBREAK are concerned, C

[issue42340] KeyboardInterrupt should come with a warning

Change by Benjamin Fogle : -- keywords: +patch pull_requests: +22151 stage: -> patch review pull_request: https://github.com/python/cpython/pull/23255 ___ Python tracker ___

[issue42340] KeyboardInterrupt should come with a warning

Change by Benjamin Fogle : Added file: https://bugs.python.org/file49595/sigint_condition_1.py ___ Python tracker ___ ___ Python-bugs-list

[issue42340] KeyboardInterrupt should come with a warning

Change by Benjamin Fogle : Added file: https://bugs.python.org/file49597/sigint_tempfile.py ___ Python tracker ___ ___ Python-bugs-list

[issue42340] KeyboardInterrupt should come with a warning

Change by Benjamin Fogle : Added file: https://bugs.python.org/file49598/sigint_zipfile.py ___ Python tracker ___ ___ Python-bugs-list

[issue42340] KeyboardInterrupt should come with a warning

New submission from Benjamin Fogle : This is related to bpo-29988, and I'm happy to move this to there. I made this a separate issue because this is a workaround, not a fix as was being discussed there. Also unlike bpo-29988, this is not restricted to context managers or finally blocks.

[issue42340] KeyboardInterrupt should come with a warning

Change by Benjamin Fogle : Added file: https://bugs.python.org/file49596/sigint_condition_2.py ___ Python tracker ___ ___ Python-bugs-list

[issue42339] official embedded Python fails to import certain modules

New submission from Jelle Geerts : This problem happened with 'python-3.8.6-embed-amd64.zip' when trying to import certain modules. Note that this problem does NOT happen with Python from 'python-3.7.9-embed-amd64.zip' (its output is also attached below). It happened with 'import socket',

[issue42319] The `functools.singledispatchmethod` example in the document cannot run

hongweipeng added the comment: Thanks,so close this one due to duplicate issue. -- stage: patch review -> resolved status: open -> closed ___ Python tracker ___

[issue42246] Implement PEP 626

STINNER Victor added the comment: > New changeset 877df851c3ecdb55306840e247596e7b7805a60a by Mark Shannon in > branch 'master': > bpo-42246: Partial implementation of PEP 626. (GH-23113) This change introduced reference leaks: https://buildbot.python.org/all/#builders/384/builds/100 5

[issue42296] Infinite loop uninterruptable on Windows in 3.10

Steve Dower added the comment: This looks like the smallest change I can make to fix it: diff --git a/Include/internal/pycore_pystate.h b/Include/internal/pycore_pystate.h index 0cd5550cfd..9ff740b87b 100644 --- a/Include/internal/pycore_pystate.h +++ b/Include/internal/pycore_pystate.h @@

[issue17852] Built-in module _io can lose data from buffered files in reference cycles

Change by Jakub Stasiak : -- nosy: +jstasiak ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue40939] Remove the old parser

Pablo Galindo Salgado added the comment: > By the way, is there a replacement for this function? The unbound project > uses it to display a SyntaxError when PyRun_SimpleFile() fails. There is no replacement for the function because that function returned CST nodes and those not exist

[issue42319] The `functools.singledispatchmethod` example in the document cannot run

Jakub Stasiak added the comment: There's an earlier issue created that relates to this (https://bugs.python.org/issue39679) and there are some possible solutions to get singledispatchmethod and classmethod work there. -- nosy: +jstasiak ___

[issue42338] Enable Debug Build For Python Native Modules in Windows, with Visual Studio Toolchain

New submission from Jeff Moguillansky : Hi, We developed a Python module that interfaces with native code via Cython. We currently build on Windows with Visual Studio Toolchain. We encounter the following issues when trying to build a debug version: 1) 3rd party modules installed via PIP are

[issue42336] Make PCbuild/build.bat build x64 by default

Tim Peters added the comment: +1. If you're feeling more ambitious, it would also be good to change build.bat and rt.bat to use the same "which platform?" spellings and with the same defaults. -- nosy: +tim.peters ___ Python tracker

[issue42337] Skip building web installers on Windows

New submission from Steve Dower : In future releases, we won't be advertising the web-based installers anymore. They never got much use, and the shorter list of downloads will be more beneficial to users than saving 10-20MB worth of bandwidth. The web site change has already been made, and

[issue42336] Make PCbuild/build.bat build x64 by default

New submission from Steve Dower : Now that the 64-bit download is the default on python.org, we should make the build command default to 64-bit as well. This should only affect contributors, so should be non-controversial. But we should add a "-x86" option to the batch file as well.

[issue1635741] Py_Finalize() doesn't clear all Python objects at exit

Christian Heimes added the comment: See bpo-42333 for _ssl module. The PR introduces module state and ports the _ssl extension to multiphase initialization. -- nosy: +christian.heimes ___ Python tracker

[issue42333] Port ssl module to heap types and module state (PEP 573)

Christian Heimes added the comment: I underestimated the effort but it's done: +567 −446 1,013 lines changed I even got rid of PyState_FindModule(). -- ___ Python tracker

[issue40939] Remove the old parser

Guido van Rossum added the comment: Honestly that code seems poorly thought out. If running it returns -1, an exception was presumably reported, but not necessarily SyntaxError -- so parsing it may not produce an error at all. The functionality needed is in PyRun_InteractiveOneObjectEx(),

[issue1635741] Py_Finalize() doesn't clear all Python objects at exit

STINNER Victor added the comment: See bpo-27400 for the _datetime module: strptime_module variable is not cleared by Py_Finalize(). -- ___ Python tracker ___

[issue42296] Infinite loop uninterruptable on Windows in 3.10

Guido van Rossum added the comment: Can you think of a fix? (Presumably restore some code that was deleted from 3.9?) -- ___ Python tracker ___

[issue41987] singledispatchmethod raises an error when relying on a forward declaration

Guido van Rossum added the comment: > Does anyone know why the treatment of unresolved references was changed in > 3.9? Probably to prepare for 3.10, where `from _future__ import annotations` is the default. > Also, I'm a bit puzzled about something from the previously mentioned Integer >

[issue42317] Docs of `typing.get_args`: Mention that due to caching of typing generics the order of arguments for Unions can be different from the one of the returned tuple

Guido van Rossum added the comment: Exactly! -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue42317] Docs of `typing.get_args`: Mention that due to caching of typing generics the order of arguments for Unions can be different from the one of the returned tuple

Dominik V. added the comment: Thinking more about it, I came to realize that it's not the Union that sits at the root of this behavior, but rather the caching performed by generic types in general. So if we consider ``` L1 = List[Union[int, str]] L2 = List[Union[str, int]] ``` then

[issue32426] Tkinter: reference document of possible cursor names

Terry J. Reedy added the comment: A link here is not a patch. The current tkinter doc for the cursor option in https://docs.python.org/3/library/tkinter.html#tk-option-data-types badly needs updating. """ The standard X cursor names from cursorfont.h can be used, without the XC_ prefix.

[issue42273] Using LazyLoader leads to AttributeError

Brett Cannon added the comment: You can ignore the half sentence. I was contemplating closing this issue when I decided to leave it open in case someone wanted to propose something and another core dev wanted to take it on. But everything is working as I expect it to and you may want to do

[issue42273] Using LazyLoader leads to AttributeError

Change by Brett Cannon : -- nosy: -brett.cannon ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue42317] Docs of `typing.get_args`: Mention that due to caching of typing generics the order of arguments for Unions can be different from the one of the returned tuple

Change by Dominik V. : -- keywords: +patch pull_requests: +22150 stage: needs patch -> patch review pull_request: https://github.com/python/cpython/pull/23254 ___ Python tracker

[issue42335] Python Crashes Exception code 0xc0000374 ntdll.dll

New submission from Charles Staton : Hello, I am experiencing crashes of Python 3.8.5 (32 bit) on multiple identical Windows 10 x64 (Enterprise 2016 LTSB) machines. The Crashes are of the sort with the Windows popup "Python has stopped working" and there is no traceback. The crashes occur

[issue42246] Implement PEP 626

Mark Shannon added the comment: New changeset cc75ab791dd5ae2cb9f6e0c3c5f734a6ae1eb2a9 by Mark Shannon in branch 'master': bpo-42246: Eliminate jumps to exit blocks by copying those blocks. (#23251) https://github.com/python/cpython/commit/cc75ab791dd5ae2cb9f6e0c3c5f734a6ae1eb2a9

[issue41987] singledispatchmethod raises an error when relying on a forward declaration

Ryan Sobol added the comment: Also, I'm a bit puzzled about something from the previously mentioned Integer class and its use of __future__.annotations. Why is it possible to declare an Integer return type for the add() method, but only possible to declare an "Integer" forward reference for

[issue42273] Using LazyLoader leads to AttributeError

E. Paine added the comment: Sorry Brett to readd you to the nosy for this, but we only got half a sentence in msg380718 (which is surely not what you intended?). While I agree with you that this is not a bug, I do feel at least a note in the docs would be helpful to explain the implications

[issue41987] singledispatchmethod raises an error when relying on a forward declaration

Ryan Sobol added the comment: Does anyone know why the treatment of unresolved references was changed in 3.9? -- ___ Python tracker ___

[issue42334] Subclassing int and complex with keyword arguments weird

autospamfighter added the comment: I tried some more classes and str is weird, but dict and set work fine. very weird -- ___ Python tracker ___

[issue42329] typing classes do not have __name__ attributes in 3.7+

Gregory P. Smith added the comment: working as intended, they aren't classes. -- resolution: -> rejected stage: needs patch -> resolved status: open -> closed ___ Python tracker

[issue42334] Subclassing int and complex with keyword arguments weird

New submission from autospamfighter : I was trying to make a custom complex class that looked something like this and it failed. I replaced complex with int and it also failed, but I tried float and it worked. ---fail class A(complex): def __init__(self, test): super().__init__()

[issue42333] Port ssl module to heap types and module state (PEP 573)

Change by hai shi : -- nosy: +shihai1991 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue42308] Add threading.__excepthook__ similar to sys.__excepthook__

STINNER Victor added the comment: PR merged, thanks Mario. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker ___

[issue42308] Add threading.__excepthook__ similar to sys.__excepthook__

STINNER Victor added the comment: New changeset 750c5abf43b7b1627ab59ead237bef4c2314d29e by Mario Corchero in branch 'master': bpo-42308: Add threading.__excepthook__ (GH-23218) https://github.com/python/cpython/commit/750c5abf43b7b1627ab59ead237bef4c2314d29e --

[issue42333] Port ssl module to heap types and module state (PEP 573)

Change by Christian Heimes : -- keywords: +patch pull_requests: +22149 pull_request: https://github.com/python/cpython/pull/23253 ___ Python tracker ___

[issue42333] Port ssl module to heap types and module state (PEP 573)

New submission from Christian Heimes : Move all objects to module state. Convert all types and extensions to heap types. -- messages: 380837 nosy: christian.heimes priority: normal severity: normal stage: patch review status: open title: Port ssl module to heap types and module state

[issue42332] add __weakref__ to types.GenericAlias

Guido van Rossum added the comment: I think it's reasonable to consider this a bug to be fixed. -- ___ Python tracker ___ ___

[issue42321] Limitations of building a static python library on Windows (MSVC)

Steve Dower added the comment: > Does a statically built Windows python.exe support C extension modules at all? I've given this some thought in the past, and I suspect the answer is "no" unless you were to also statically link the extension modules into the same executable (which will

[issue42332] add __weakref__ to types.GenericAlias

Tin Tvrtković added the comment: It would be great if we could get this into 3.9. -- ___ Python tracker ___ ___ Python-bugs-list

[issue42327] Add PyModule_Add()

STINNER Victor added the comment: > PyModule_AddObjectRef() is just Py_XINCREF() followed by PyModule_Add(). But > since most values added to the module are new references, Py_XINCREF() is > usually not needed. There is no general rule. I saw two main cases. (A) Create an object only to

[issue41987] singledispatchmethod raises an error when relying on a forward declaration

Guido van Rossum added the comment: So the biggest difference I see is that ForwardRef._evaluate() has grown a recursive_guard argument in 3.9. This makes me think that in 3.8, only one level of evaluation was happening, and in 3.8, we keep evaluating until we don't see a string or

[issue42246] Implement PEP 626

Change by Mark Shannon : -- pull_requests: +22148 pull_request: https://github.com/python/cpython/pull/23251 ___ Python tracker ___

[issue42317] Docs of `typing.get_args`: Mention that due to caching of typing generics the order of arguments for Unions can be different from the one of the returned tuple

Ken Jin added the comment: Dominik, would you like to submit a PR for this :) ? -- ___ Python tracker ___ ___ Python-bugs-list

[issue42329] typing classes do not have __name__ attributes in 3.7+

Guido van Rossum added the comment: So shall we just close this? -- ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue42327] Add PyModule_Add()

Serhiy Storchaka added the comment: PyModule_Add() allows to make such macro much simpler: #define MOD_ADD(name, expr) \ do { \ if (PyModule_Add(mod, name, expr) < 0) { \ return -1; \ } \ } while (0) PyModule_AddObjectRef() is just Py_XINCREF() followed by

[issue42332] add __weakref__ to types.GenericAlias

Change by Ken Jin : -- keywords: +patch nosy: +kj nosy_count: 2.0 -> 3.0 pull_requests: +22147 stage: -> patch review pull_request: https://github.com/python/cpython/pull/23250 ___ Python tracker

[issue40939] Remove the old parser

STINNER Victor added the comment: > There is no mention of PyParser_SimpleParseFile() removal in What's New in > Python 3.10. By the way, is there a replacement for this function? The unbound project uses it to display a SyntaxError when PyRun_SimpleFile() fails. Petr Menšík asked: "Could

[issue42312] sys.prefix is set incorrectly on Mac OS X

Ronald Oussoren added the comment: > In any case, do you think that there is a way to get the behavior I am > looking for (basically, invoke a Python interpreter that has sys.prefix set > to the venv but without a symbolic link)? Not that I know. Using the python binary in the virtual

[issue40939] Remove the old parser

STINNER Victor added the comment: I reopen the issue. Would you mind to explicitly list function removed from the C API in What's New in Python 3.10? https://docs.python.org/dev/whatsnew/3.10.html#id4 I'm talking about the commit 1ed83adb0e95305af858bd41af531e487f54fee7. For example, the

[issue42312] sys.prefix is set incorrectly on Mac OS X

Michael Ferguson added the comment: > The way sys.prefix is calculated on macOS ensures that the correct sys.prefix > is calculated even if you copy the binary to a different location. That's > functionality I don't want to drop. I agree with you that it's important for the Python

[issue38823] Improve stdlib module initialization error handling.

STINNER Victor added the comment: New changeset b5cc05bbe681dbe06d5ec6d34318815d1c1ad6c5 by Victor Stinner in branch 'master': bpo-38823: Always build _ctypes with wchar_t (GH-23248) https://github.com/python/cpython/commit/b5cc05bbe681dbe06d5ec6d34318815d1c1ad6c5 --

[issue42260] [C API] Add PyInterpreterState_SetConfig(): reconfigure an interpreter

STINNER Victor added the comment: New changeset ef75a625cdf8377d687a04948b4db9bc1917bf19 by Victor Stinner in branch 'master': bpo-42260: Initialize time and warnings earlier at startup (GH-23249) https://github.com/python/cpython/commit/ef75a625cdf8377d687a04948b4db9bc1917bf19 --

[issue42260] [C API] Add PyInterpreterState_SetConfig(): reconfigure an interpreter

Change by STINNER Victor : -- pull_requests: +22146 pull_request: https://github.com/python/cpython/pull/23249 ___ Python tracker ___

[issue38823] Improve stdlib module initialization error handling.

Change by STINNER Victor : -- pull_requests: +22145 pull_request: https://github.com/python/cpython/pull/23248 ___ Python tracker ___

[issue38823] Improve stdlib module initialization error handling.

STINNER Victor added the comment: New changeset d19fa7a337d829e3dab3e9f919f5dcf09cf6f6ba by Victor Stinner in branch 'master': bpo-38823: Fix refleaks in _ctypes extension init (GH-23247) https://github.com/python/cpython/commit/d19fa7a337d829e3dab3e9f919f5dcf09cf6f6ba --

[issue42327] Add PyModule_Add()

STINNER Victor added the comment: In practice, PyModule_AddRef(mod, obj) behaves as PyModule_Add(mod, Py_NewRef(obj)) if I understood correctly. -- ___ Python tracker ___

[issue42327] Add PyModule_Add()

STINNER Victor added the comment: If PyModule_Add() is added, I would suggest to rename PyModule_AddObjectRef() to PyModule_AddRef() for consistency. IMHO PyModule_AddObjectRef() remains useful even if PyModule_Add() is added. -- ___ Python

[issue42327] Add PyModule_Add()

STINNER Victor added the comment: I'm using more and more often such macro: #define MOD_ADD(name, expr) \ do { \ PyObject *obj = (expr); \ if (obj == NULL) { \ return -1; \ } \ if (PyModule_AddObjectRef(mod, name, obj) < 0) { \

[issue42327] Add PyModule_Add()

STINNER Victor added the comment: Oh, I just rejected PR 17298. Copy of my message: --- I added PyModule_AddObjectRef() which uses strong references, rather than only stealing a reference on success. I also enhanced the documentation to show concrete examples:

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

Change by Andrew Svetlov : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed versions: +Python 3.8 ___ Python tracker ___

[issue38823] Improve stdlib module initialization error handling.

Change by STINNER Victor : -- pull_requests: +22144 pull_request: https://github.com/python/cpython/pull/23247 ___ Python tracker ___

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

Andrew Svetlov added the comment: New changeset f37628eb7117f222de24a6931aa7639e668cb7b0 by Jakub Stasiak in branch '3.8': [3.8] bpo-42237: Fix os.sendfile() on illumos (GH-23154). (GH-23246) https://github.com/python/cpython/commit/f37628eb7117f222de24a6931aa7639e668cb7b0 --

[issue42332] add __weakref__ to types.GenericAlias

New submission from Tin Tvrtković : For example, dict[int, int] cannot be used with singledispatch because types.GenericAlias doesn't support weak references. I'm an author of a third party library (https://github.com/Tinche/cattrs) for which this functionality would be useful. Here's a

[issue32426] Tkinter.ttk Widget does not define wich option exists to set the cursor

Alessandro Piccione added the comment: As suggested the documentation for "cursor" is here: https://www.tcl.tk/man/tcl/TkCmd/cursors.htm Close [SOLVED] -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker

[issue24831] Load average in test suite too high

STINNER Victor added the comment: No activity for 5 years, so far this issue is more an open discussion than a list of concrete changes, so I close the issue as out of date. If some tests are too CPU or I/O intensive, you can skip them using the -x option. Use --ignorefile to only skip

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

Change by Jakub Stasiak : -- pull_requests: +22143 pull_request: https://github.com/python/cpython/pull/23246 ___ Python tracker ___

[issue32426] Tkinter.ttk Widget does not define wich option exists to set the cursor

Pat Thoyts added the comment: The Tk documentation for the acceptable cursor names is the cursors manual page. https://www.tcl.tk/man/tcl/TkCmd/cursors.htm Tk does not provide a way to get all these names in script. This should probably be closed. -- nosy: +patthoyts

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

miss-islington added the comment: New changeset 7ae19ef5cf5d4f464588133bd48f81d80ea54aee by Miss Islington (bot) in branch '3.9': bpo-42237: Fix os.sendfile() on illumos (GH-23154) https://github.com/python/cpython/commit/7ae19ef5cf5d4f464588133bd48f81d80ea54aee --

[issue42246] Implement PEP 626

Change by Mark Shannon : -- pull_requests: +22142 stage: -> patch review pull_request: https://github.com/python/cpython/pull/23245 ___ Python tracker ___

[issue41116] build on macOS 11 (beta) does not find system-supplied third-party libraries

Sijmen J. Mulder added the comment: With './configure; make' of master on macOS 11 on Apple Silicon I get the library detection failure. The quick and easy fix was to amend inc_dirs and lib_dirs in setup.py: if MACOS: +sysroot = macosx_sdk_root() +

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

Change by miss-islington : -- nosy: +miss-islington nosy_count: 4.0 -> 5.0 pull_requests: +22141 pull_request: https://github.com/python/cpython/pull/23244 ___ Python tracker

[issue42237] test_socket.SendfileUsingSendfileTest fails on illumos

Andrew Svetlov added the comment: New changeset fd4ed57674c675e05bd5d577dd5047a333c76c78 by Jakub Stasiak in branch 'master': bpo-42237: Fix os.sendfile() on illumos (GH-23154) https://github.com/python/cpython/commit/fd4ed57674c675e05bd5d577dd5047a333c76c78 -- nosy: +asvetlov

[issue42246] Implement PEP 626

Mark Shannon added the comment: New changeset 877df851c3ecdb55306840e247596e7b7805a60a by Mark Shannon in branch 'master': bpo-42246: Partial implementation of PEP 626. (GH-23113) https://github.com/python/cpython/commit/877df851c3ecdb55306840e247596e7b7805a60a --

[issue16879] distutils.command.config uses fragile constant temporary file name

Amir Mohamadi added the comment: Can I open a PR for it? -- nosy: +Amir ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue42329] typing classes do not have __name__ attributes in 3.7+

Gregory P. Smith added the comment: Not a big deal if we don't, I just found it odd so I figured I'd pose the question. That it's been in three releases and only just now come up is pretty telling that isn't critical. The code in question was trying to identify public functions in a module by

[issue42331] 'base64' has no attribute 'decodestring'

Karthikeyan Singaravelan added the comment: The tracker is for issues related to CPython. This seems to be an issue with gsutil on Python 3.9 compatibility. This seems to have been fixed as below : https://github.com/GoogleCloudPlatform/gsutil/issues/1118 -- nosy: +xtreak