[issue29537] Alternative fix for BUILD_MAP_UNPACK_WITH_CALL opcode in 3.5

Nick Coghlan added the comment: For easier cross-referencing, note that http://bugs.python.org/issue29514 is the issue proposing a test case that ensures future maintainers are aware of the practical problems created by bumping the bytecode magic number in a maintenance release. --

[issue29537] Alternative fix for BUILD_MAP_UNPACK_WITH_CALL opcode in 3.5

Changes by Nick Coghlan : -- type: -> behavior ___ Python tracker ___ ___

[issue29537] Alternative fix for BUILD_MAP_UNPACK_WITH_CALL opcode in 3.5

Changes by Nick Coghlan : -- assignee: -> ncoghlan ___ Python tracker ___ ___

[issue26184] Add versionchanged note for error when create_module() is not defined by loaders

Nick Coghlan added the comment: That should actually be a versionchanged note now - converting this to an easy documentation issue to cover that change. -- assignee: -> docs@python components: +Documentation -Interpreter Core keywords: +easy nosy: +docs@python stage: test needed ->

[issue14243] tempfile.NamedTemporaryFile not particularly useful on Windows

Nick Coghlan added the comment: John, your problem sounds different - if you're opening the files in binary mode, then you'll be getting a default buffer that's probably 4k or 8k in size, so if you're writing less content than that, the subprocess won't see anything until you explicitly

[issue27788] platform module's version number doesn't match its docstring

Berker Peksag added the comment: Perhaps it's now time to drop that __version__ attribute in 3.7? We've removed the __version__ attribute from the email module in issue 22508. -- nosy: +berker.peksag stage: -> patch review ___ Python tracker

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Changes by Raymond Hettinger : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___

[issue26184] raise an error when create_module() is not defined by exec_module() is for loaders

Matthias Bussonnier added the comment: should be closed by #28026, it now raises an ImportError. I think this can be closed. Unless one want to change the text in Starting in Python 3.6 it will be an error > Starting in Python 3.6 it will be an error To use the past tense. -- nosy:

[issue28556] typing.py upgrades

Changes by Mariatta Wijaya : -- pull_requests: +242 ___ Python tracker ___ ___

[issue28556] typing.py upgrades

Changes by Mariatta Wijaya : -- pull_requests: +241 ___ Python tracker ___ ___

[issue14243] tempfile.NamedTemporaryFile not particularly useful on Windows

Changes by Mark Lawrence : -- nosy: -BreamoreBoy ___ Python tracker ___ ___

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Louie Lu added the comment: Raymond: comment has changed, pushed on to GitHub. -- ___ Python tracker ___ ___

[issue29632] argparse values for action in add_argument() should be flags instead of (or in addition to) strings

paul j3 added the comment: Those strings are defined in a 'parser.registry' dictionary, with expressions like self.register('action', 'store', _StoreAction) (Users can also register their own custom classes. The registry can also be used for the 'type' attribute.) So you can already do:

[issue29639] test suite intentionally avoids referring to localhost, destroying abstraction away from IPv6 vs IPv4

New submission from Gregory P. Smith: I am working on fixing our test suite to run on IPv6 only hosts (which are becoming a reality). Many failures today occur because of hard coded 127.0.0.1 values. This is wrong. We should refer to "localhost" The "solution" to

[issue18792] test_ftplib timeouts

Gregory P. Smith added the comment: https://bugs.python.org/issue29639 opened to track undoing this when appropriate. -- ___ Python tracker ___

[issue18792] test_ftplib timeouts

Gregory P. Smith added the comment: FYI - hardcoding these addresses is now causing me problems as I try to get our test suite passing on IPv6 only hosts. "localhost" is correct. IMNSHO if for some reason a system cannot resolve "localhost" into a correct address, it should be banned from

[issue29632] argparse values for action in add_argument() should be flags instead of (or in addition to) strings

Changes by paul j3 : -- nosy: +paul.j3 ___ Python tracker ___ ___ Python-bugs-list

[issue29622] ast module doesn't support optional fields of Parser/Python.asdl

Changes by Matthias Bussonnier : -- pull_requests: +240 ___ Python tracker ___

[issue29463] Add `docstring` field to AST nodes

Changes by Matthias Bussonnier : -- pull_requests: +238 ___ Python tracker ___

[issue29637] ast.get_docstring(): AttributeError: 'NoneType' object has no attribute 'expandtabs'

Changes by Matthias Bussonnier : -- pull_requests: +239 ___ Python tracker ___

[issue29637] ast.get_docstring(): AttributeError: 'NoneType' object has no attribute 'expandtabs'

Changes by Matthias Bussonnier : -- nosy: +mbussonn ___ Python tracker ___ ___

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Raymond Hettinger added the comment: The code looks fine. Please change the comment to something like: /* Reduce the number of repetitions when maxlen would be exceeded */ -- ___ Python tracker

[issue29638] Spurious failures in test_collections in releak hunting mode after typing is imported

New submission from Ivan Levkivskyi: This command: ./python -c 'import runpy, typing; runpy.run_module("test")' -R 5:5 test_collections Sometimes gives spurious failures like this: test_collections leaked [0, 0, 3, -3, 3] memory blocks, sum=3 I think this is because ABC caches of

[issue28556] typing.py upgrades

Changes by Ivan Levkivskyi : -- pull_requests: +237 ___ Python tracker ___ ___

[issue29549] Improve docstring for str.index

Serhiy Storchaka added the comment: Write docstrings in following style: "index($self, sub, start=0, end=sys.maxsize, /)\n" "--\n" "\n" "\n" "\n" "" Extract the summary line. Don't use the metavariable "S" since it no longer defined in the signature. --

[issue22273] abort when passing certain structs by value using ctypes

Vinay Sajip added the comment: > Perhaps it should instead use two specific flags, TYPEFLAG_HASBITFIELD and > TYPEFLAG_HASUNION This seems better at first sight. It's not making any suitability decisions (apart from doing the unrolling), and the meaning of these flags will be less volatile

[issue29191] liblzma is missing from pcbuild.sln

Changes by Steve Dower : -- versions: +Python 3.6 ___ Python tracker ___ ___

[issue29624] Python 3.5.3 x86 web installer cannot install launcher

Changes by Steve Dower : -- pull_requests: +235 ___ Python tracker ___ ___

[issue29624] Python 3.5.3 x86 web installer cannot install launcher

Changes by Steve Dower : -- pull_requests: +236 ___ Python tracker ___ ___

[issue28909] Adding LTTng-UST tracing support

Peter McCormick added the comment: Hi Łukasz, thank you for the feedback! > "PyTrace" is already a name in use for a different purpose. I understand the > itch to make the name more "right" but I am in general not a fan of renaming > "PyDTrace" to anything else now. It was a placeholder from

[issue22273] abort when passing certain structs by value using ctypes

Eryk Sun added the comment: Perhaps it should instead use two specific flags, TYPEFLAG_HASBITFIELD and TYPEFLAG_HASUNION, which are propagated unconditionally from the base class and fields. As a base case, a union itself is flagged TYPEFLAG_HASUNION. Arrays are unrolled on X86_64 only if

[issue29514] Add a test case that prevents magic number changes in minor releases

Changes by Carson Lam : -- nosy: +Carson Lam ___ Python tracker ___ ___ Python-bugs-list

[issue27593] Deprecate sys._mercurial and create sys._git

Steve Dower added the comment: It looks to me like we want: branch=`git name-rev --name-only HEAD` revision=`git rev-parse HEAD` tag=`git name-rev --tags --name-only HEAD` Unless we're planning on leaving out the tag? My PR 262 makes the Windows build changes in master, but doesn't change

[issue27593] Deprecate sys._mercurial and create sys._git

Changes by Steve Dower : -- pull_requests: +234 ___ Python tracker ___ ___

[issue14243] tempfile.NamedTemporaryFile not particularly useful on Windows

Changes by Jakub Wilk : -- nosy: +jwilk ___ Python tracker ___ ___ Python-bugs-list mailing

[issue29637] ast.get_docstring(): AttributeError: 'NoneType' object has no attribute 'expandtabs'

New submission from Jakub Wilk: With git master (4c78c527d215c37472145152cb0e95f196cdddc9) I get this: >>> import ast >>> ast.get_docstring(ast.parse('')) Traceback (most recent call last): File "", line 1, in File "/home/jwilk/opt/lib/python3.7/ast.py", line 203, in get_docstring text

[issue28909] Adding LTTng-UST tracing support

Francis Deslauriers added the comment: I am finally having the time to work in this. > A nit: the name LTTng-UST is rather unfriendly, especially when used without > the dash and in all lowercase characters. Given that we're using "dtrace" and > "systemtap", it would be simpler to just use

[issue16355] inspect.getcomments() does not work in the interactive shell

Marco Buttu added the comment: Hello Vajrasky, the doc patch LGTM. Looking at the David's comment in Rietveld, it seems that he does not want the test patch to be applyed. Can you please make a pull request? Thank you very much -- ___ Python

[issue29414] Change 'the for statement is such an iterator' in Tutorial

Marco Buttu added the comment: Hello Jim, do you have the time to make a pull request? Let me know, otherwise I will do it -- ___ Python tracker ___

[issue29191] liblzma is missing from pcbuild.sln

Mo Jia added the comment: I think let user add the liblzma project by hand is not good enough. While the build.bat don't have this problem. So they should work similar. -- nosy: +Mo.Jia versions: -Python 3.5, Python 3.6 ___ Python tracker

[issue29455] Mention coverage.py in trace module documentation

Changes by Marco Buttu : -- pull_requests: +233 ___ Python tracker ___ ___

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Changes by Serhiy Storchaka : -- status: open -> closed ___ Python tracker ___

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Fabien Dubosson added the comment: > I'm going to close this issue. I was doing to do so, here was my message: > See shells sources. According to the manpage `pwd -L` just uses PWD from > environment. I looked directly at `pwd` sources, and

[issue22273] abort when passing certain structs by value using ctypes

Vinay Sajip added the comment: > It occurs to me that in the 1st pass, it also needs to propagate the > non-argument flag from any field that has it set. So does that mean disallowing a structure which contains a union? What about if the final structure is large enough to require passing in

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

R. David Murray added the comment: Ah, that should all be implementable from your python shell, then, no need for support in the os module. For reference, when I said 'cd man page', this is what I was referring to: http://www.unix.com/man-page/posix/1posix/cd/ I'm going to close this issue.

[issue27413] Add an option to json.tool to bypass non-ASCII characters.

Changes by Daniel Himmelstein : -- pull_requests: +232 ___ Python tracker ___

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Fabien Dubosson added the comment: > Most os functions are thin wrappers around system calls. There is no system > call that works as 'cd' without '-P'. I would like to believe in this, but then if `cd` is some bash internal, how does `/usr/bin/pwd -L` find it back? > Since yours is the

[issue29615] SimpleXMLRPCDispatcher._dispatch mangles tracebacks when invoking RPC calls through _dispatch

Changes by Petr MOTEJLEK : -- pull_requests: +231 ___ Python tracker ___ ___

[issue29636] Specifying indent in the json.tool command

New submission from Daniel Himmelstein: The utility of `python -m json.tool` would increase if users could specify the indent level. Example use case: newlines in a JSON document are important for readability and the ability to open in a text editor. However, if the file is large, you can

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Serhiy Storchaka added the comment: > Just wondering, do you know what is the function called by `cd`/`cd -L` then? > It doesn't seems to be a bash internal tweak because `/usr/bin/pwd -L` is > able to get the symlink path, and this binary is not part of bash. See shells sources. According to

[issue29624] Python 3.5.3 x86 web installer cannot install launcher

Steve Dower added the comment: I have built an updated web installer which should allow us to replace just the one file on the server to fix the issue. The only downside is that it is not compatible with existing installs of 3.5.3 (that is, you can't run the executable to Modify/Repair/Remove

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

R. David Murray added the comment: Since yours is the itch, I'm afraid you are going to have to be the one to figure out how this could be implemented :) -- ___ Python tracker

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Fabien Dubosson added the comment: Thanks for the additional information! > The 'cd' man page says that 'cd -P' should "perform actions equivalent to the > chdir() function". Just wondering, do you know what is the function called by `cd`/`cd -L` then? It doesn't seems to be a bash internal

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Serhiy Storchaka added the comment: 'cd' is not an external command and has no manpage. It is a shell builtin command. Most os functions are thin wrappers around system calls. There is no system call that works as 'cd' without '-P'. If you implement a shell in Python, you perhaps need an

[issue14243] tempfile.NamedTemporaryFile not particularly useful on Windows

John Florian added the comment: I just stumbled onto this though I'm not writing for Windows. Instead, I'm on Fedora 25 with Python 3.5.2 and I went nearly crazy tracing down what seemed to be inconsistent behavior. My use case has Python using NamedTemporaryFile(delete=True) in a CM to

[issue29625] colorsys.rgb_to_hsv always returns saturation as 0 (python2.7 only)

Mark Dickinson added the comment: No problem. I did spend some time wondering whether `rgb_to_hsv` still gives the correct results if any of the inputs happens to be an integer (though still in the range [0, 1]). It looks as though it does, though. --

[issue10379] locale.format() input regression

Changes by Garvit Khatri : -- pull_requests: +229 ___ Python tracker ___ ___

[issue29624] Python 3.5.3 x86 web installer cannot install launcher

Changes by Steve Dower : -- pull_requests: +228 ___ Python tracker ___ ___

[issue22273] abort when passing certain structs by value using ctypes

Eryk Sun added the comment: > I'm not sure using this flag impacts on consistency with CFFI I meant consistency with respect to supported argument types. If someone contributed a workaround to CFFI, then I would rather port it, but it looks like they're also waiting for this to be addressed

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Changes by R. David Murray : -- type: behavior -> enhancement ___ Python tracker ___

[issue29549] Improve docstring for str.index

Mariatta Wijaya added the comment: Thanks for the PR, Lisa. Serhiy, Raymond, could you both review this? Thanks. I can do the merge and backport once it gets both of your approval :) Also,should this be applied to 3.5 too? -- ___ Python tracker

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

R. David Murray added the comment: The function is chdir, not cd. The 'cd' man page says that 'cd -P' should "perform actions equivalent to the chdir() function". So, you are asking for a different function, which should *not* be named 'os.chdir'. You'll have to make a case for it being

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Fabien Dubosson added the comment: The previous example (sorry, first time using this platform, I'm a little bit messy here), shows that `os.chdir` resolves symlinks by default, as opposed to what `cd` does in bash. This means it is not possible to change the directory to a symlink folder. A

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

New submission from Fabien Dubosson: When using bash, the `cd` function does not follow symlinks by default, but `cd -P` does. The `os.chdir` function behaves like `cd -P` preventing to be able to change directory to a symlink folder. Initial setup (make a `/tmp/to/dst` symlink pointing to

[issue29635] os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks

Changes by Fabien Dubosson : -- components: Interpreter Core nosy: StreakyCobra priority: normal severity: normal status: open title: os.chdir() acts like `cd -P` by default, it should offer an option to not follow symlinks type: behavior versions: Python

[issue24024] str.__doc__ needs an update

Changes by Garvit : -- pull_requests: +227 ___ Python tracker ___ ___ Python-bugs-list

[issue29622] ast module doesn't support optional fields of Parser/Python.asdl

Changes by INADA Naoki : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___

[issue28814] Deprecation notice on inspect.getargvalues() is incorrect

Matthias Bussonnier added the comment: Thanks Nick for finding this ! -- ___ Python tracker ___ ___

[issue29549] Improve docstring for str.index

Lisa Roach added the comment: I'll just go ahead and make my PR, let me know what else needs to be done. Serhiy, if you could point me in the direction of how to write the docstring so that it is in the Argument Clinic style I would be happy to take a look. --

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Louie Lu added the comment: Serhiy: yes, your advice is better than checking inside the loop. I have updated this to the commit, thanks! -- ___ Python tracker

[issue29549] Improve docstring for str.index

Changes by Lisa Roach : -- pull_requests: +225 ___ Python tracker ___ ___

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Serhiy Storchaka added the comment: Wouldn't be better to update the number of repeats before the loop rather than checking at every iteration? if (deque->maxlen >= 0 && n * size > deque->maxlen) n = (deque->maxlen + size - 1) / size; -- assignee: -> rhettinger nosy: +rhettinger,

[issue22273] abort when passing certain structs by value using ctypes

Vinay Sajip added the comment: > We need more feedback on this suggested flag, especially to stay consistent > with CFFI if possible. Undoubtedly, more feedback would be very helpful. I'm not sure using this flag impacts on consistency with CFFI particularly, since it's an internal

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

Changes by Louie Lu : -- pull_requests: +223 ___ Python tracker ___ ___ Python-bugs-list

[issue29634] Reduce deque repeat execution when maxlen exist and size is not 1

New submission from Louie Lu: There is a XXX in v3.5.0 shows that need to dealing with deque maxlen setting case in deque_repeat. Although there have common case for deque size 1 with maxlen, other size of deque with maxlen still using for-loop to extend the deque, without any detection.

[issue29598] Write unit tests for pdb module

Xavier de Gaye added the comment: > Note that there is a bug in find_function() (it opens a file with default > encoding). Please open an issue. -- ___ Python tracker

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

STINNER Victor added the comment: I don't completely reject the issue. I may come back later if I find a way to make it even more efficient. But I prefer to close the issue as REJECTED to make it clear that right now with the current implementation and benchmark results, it's not worth it.

[issue29465] Modify _PyObject_FastCall() to reduce stack consumption

Changes by STINNER Victor : -- title: Add _PyObject_FastCall() to reduce stack consumption -> Modify _PyObject_FastCall() to reduce stack consumption ___ Python tracker

[issue29632] argparse values for action in add_argument() should be flags instead of (or in addition to) strings

R. David Murray added the comment: The obvious thing to do would be to make the Action subclasses have public names. I personally would continue to use the strings, though, as they are easier to type. (The same would be true if an enum were introduced (that I'd continue to use the stings

[issue29633] MSI installer does not pass values as SecureProperty from UI

New submission from aldehoff: This is a revival of issue 1298962 (http://bugs.python.org/issue1298962) from 2009. I can reproduce the problem with an MSI created by cx_Freeze and bdist_msi on a Windows 7 64bit system in a company domain. Python is at version 3.6.0, cx_Freeze is at 5.0.1. I am

[issue28911] Clarify the behaviour of assert_called_once_with

Changes by Arne de Laat : -- pull_requests: +222 ___ Python tracker ___ ___

[issue28814] Deprecation notice on inspect.getargvalues() is incorrect

Nick Coghlan added the comment: Thanks mbusson, berker.peksag! -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue27840] functools.partial: don't copy keywoard arguments in partial_call()?

STINNER Victor added the comment: I reopen the issue to propose to add a comment explaining why the dictionary must always be copied: https://github.com/python/cpython/pull/253 -- pull_requests: +221 resolution: rejected -> status: closed -> open

[issue29581] __init_subclass__ causes TypeError when used with standard library metaclasses (such as ABCMeta)

Nick Coghlan added the comment: Note that the publisher side workaround for this is relatively straightforward: __init_subclass__ implementations that want to be compatible with arbitrary metaclasses will need to take any additional parameters as class attributes (which they may delete),

[issue22273] abort when passing certain structs by value using ctypes

Eryk Sun added the comment: I had suggested inheriting the TYPEFLAG_NONARGTYPE flag in StructUnionType_new. It requires a minor change to get basedict unconditionally, and then assign if (basedict) dict->flags |= basedict->flags & TYPEFLAG_NONARGTYPE; We need more feedback on

[issue29581] __init_subclass__ causes TypeError when used with standard library metaclasses (such as ABCMeta)

Changes by Nick Coghlan : -- nosy: +Martin.Teichmann ___ Python tracker ___ ___

[issue29581] __init_subclass__ causes TypeError when used with standard library metaclasses (such as ABCMeta)

Nick Coghlan added the comment: Oops, and now I see the typo in the example code, it seems you're right. I was misremembering an earlier more decorator-like variant of the design where we didn't rely on passing the __init_subclass__ arguments through the metaclass constructor. --

[issue29581] __init_subclass__ causes TypeError when used with standard library metaclasses (such as ABCMeta)

Nick Coghlan added the comment: No, the filtering is only applied to the __new__ and __init__ calls on the metaclass, not to the __init_subclass__ call. Showing the last part of an interactive session where I ran the above commands: === >>> class

[issue23890] assertRaises increases reference counter

Nick Coghlan added the comment: Victor's PR takes a more pragmatic approach to address this problem: rather than adding the ability to clear the frames for an arbitrary exception tree, it just uses a try/finally in a couple of key unittest function definitions to clear the offending circular

[issue23890] assertRaises increases reference counter

Changes by Nick Coghlan : -- stage: needs patch -> commit review ___ Python tracker ___

[issue28911] Clarify the behaviour of assert_called_once_with

Changes by Arne de Laat : -- pull_requests: +220 ___ Python tracker ___ ___

[issue22273] abort when passing certain structs by value using ctypes

Vinay Sajip added the comment: Just a thought - the TYPEFLAG_NONARGTYPE needs to be copied from the base class if set there, right? -- ___ Python tracker

[issue29606] urllib FTP protocol stream injection

Changes by STINNER Victor : -- nosy: +christian.heimes, haypo ___ Python tracker ___

[issue18423] Document limitations on -m

Nick Coghlan added the comment: The limitation noted in #18422 was specific to 3.3, as the missing API was added to NamespaceLoader in 3.4. So the current documentation is accurate for recent versions of Python. -- nosy: +ncoghlan resolution: -> out of date stage: needs patch ->

[issue29624] Python 3.5.3 x86 web installer cannot install launcher

Steve Dower added the comment: Creating a repaired installer is possible, though risky (it's easy when I still have the original build handy, but at this stage a rebuild will produce different hashes and make things worse). But I intend to look into that as soon as I can. If the launcher is

[issue28911] Clarify the behaviour of assert_called_once_with

Changes by Arne de Laat : -- pull_requests: +219 ___ Python tracker ___ ___

[issue28598] RHS not consulted in `str % subclass_of_str` case.

Changes by Serhiy Storchaka : -- pull_requests: +218 ___ Python tracker ___ ___

[issue29632] argparse values for action in add_argument() should be flags instead of (or in addition to) strings

New submission from Pedro: The possible values for action are currently: 'store' 'store_true' 'store_false' 'store_const' 'append' 'append_const' 'count' 'help' 'version' a subclass of argparse.Action The string values are evidently for backward compatibility with optparse. However, adding the

[issue28598] RHS not consulted in `str % subclass_of_str` case.

Changes by Serhiy Storchaka : -- pull_requests: -57 ___ Python tracker ___ ___

[issue28810] Document bytecode changes in 3.6

Changes by Ivan Levkivskyi : -- pull_requests: +217 ___ Python tracker ___ ___

  1   2   >