[issue25509] PyImport_ImportModule inaccurately described

New submission from Memeplex: The documentation (for 3.5) states that "[PyImport_ImportModule] is a simplified interface to PyImport_ImportModuleEx()" but the current implementation calls PyImport_Import instead, which is a higher level interface that takes into account import hooks. Older

[issue25510] fileinput.FileInput.readline() always returns str object at the end even if in 'rb' mode

New submission from Ryosuke Ito: In Python3, fileinput.FileInput.readline() always returns str object at the end, even if in 'rb' mode. Here's a test code. import fileinput fi = fileinput.input('test_fileinput.py', mode='rb') while True: line = fi.readline() assert isinstance(line,

[issue25510] fileinput.FileInput.readline() always returns str object at the end even if in 'rb' mode

Changes by Ryosuke Ito : -- type: -> behavior ___ Python tracker ___ ___

[issue25510] fileinput.FileInput.readline() always returns str object at the end even if in 'rb' mode

Changes by Ryosuke Ito : -- components: +Library (Lib) versions: +Python 3.5 ___ Python tracker ___

[issue25498] Python 3.4.3 core dump with simple sample code

Stefan Krah added the comment: We should ultimately do something like I suggested in msg235256. Everything else is a hack (N.B.: the issues that 1da9630e9b7f tried to fix were also hacks, so it isn't really the fault of that commit). -- nosy: +skrah

[issue25507] IDLE: user code 'import tkinter; tkinter.font' should fail

Mark Roseman added the comment: This (restructuring/refactoring to minimize the subprocess imports) does definitely sound like the right approach. There will be other benefits to breaking up PyShell a bit too.. -- ___ Python tracker

[issue25508] LogRecord attributes are not tuple, when logging only dict

R. David Murray added the comment: I believe this is due to the way % formatting works, and so is a doc bug. Vinay will know for sure, of course :) -- nosy: +r.david.murray, vinay.sajip ___ Python tracker

[issue25510] fileinput.FileInput.readline() always returns str object at the end even if in 'rb' mode

Serhiy Storchaka added the comment: Would be nice to add a test in Lib/test/test_fileinput.py. -- nosy: +serhiy.storchaka stage: -> test needed ___ Python tracker

[issue25510] fileinput.FileInput.readline() always returns str object at the end even if in 'rb' mode

R. David Murray added the comment: We need to think carefully about backward compatibility here. -- nosy: +r.david.murray ___ Python tracker ___

[issue25482] signal.set_wakeup_fd() doesn't work if the signal don't have handler

Atsuo Ishimoto added the comment: > Benjamin Peterson added the comment: > > The OP probably wants something like the Linux-specific signalfd() syscall. Yes. Long explanation: I have an console text editor written in Python/curses(http://kaaedit.github.io/). This editor didn't work with

[issue25511] multiprocessing pool blocks SIGTERM from being handled

New submission from David Jones: This is probably related to #21913, but more specifically concerns the documentation. I have a sub process of a larger program that handles a SIGTERM sent by the main process for a clean shutdown. However, if I launch a parallel task in the sub process, via

[issue8231] Unable to run IDLE without write-access to home directory

Mark Roseman added the comment: Better, but alas still not quite. On further investigation, the issue is that a new instance of idleConf is instantiated in the subprocess, which then calls mkdtemp() returning a different name. You can see this by doing 'restart shell' and noting that it will

[issue25509] PyImport_ImportModule inaccurately described

Brett Cannon added the comment: Semantic change was probably because of the lack of import hook support which is required since Python 3.3. So the docs just need to be fixed rather than the semantics. -- ___ Python tracker

[issue25489] sys.exit() caught in exception handler

Brian Sutherland added the comment: On Wed, Oct 28, 2015 at 02:49:55PM +, R. David Murray wrote: > > R. David Murray added the comment: > > Using sys.exit means you are depending on garbage collection to clean > up all of your program's resources. In the general case this is a bad > idea.

[issue25501] Use async/await through asyncio docs

Brett Cannon added the comment: Once 3.4.4 launches the need to keep the docs synced with a version that doesn't support async/await goes away. And worrying about 3.3 isn't necessary since asyncio was added in 3.4. So once 3.4.4 is released and we close the 3.4 branch to bugfixes can we

[issue17006] Add advice on best practices for hashing secrets

Changes by Joshua Bronson : -- nosy: +jab ___ Python tracker ___ ___ Python-bugs-list

[issue25469] multiprocessing .Condition.notify(_all) function has O(N) time complexity where N is the number of wait() calls with a timeout since the last notify(_all) call

Vilnis Termanis added the comment: I realise that having type bug type of "performance" means it's less likely to be looked at than if I had marked is as "behaviour" or "resource usage" (which would not be completely unfair, I think). Also, the patch should be applicable to 2.7 and 3.2

[issue25509] PyImport_ImportModule inaccurately described

Changes by Brett Cannon : -- nosy: +brett.cannon ___ Python tracker ___ ___

[issue25513] collections.abc.Iterable don't implement __bool__

New submission from yoch: collections.abc.Iterable don't implement the __bool__ method. This may seriously degrade performance in case __len__ is not efficient. I suggest to implement it as : class Iterable: ... def __bool__(self): try: next(iter(self))

[issue25513] collections.abc.Iterable don't implement __bool__

R. David Murray added the comment: This is as designed. The only method needed for something to qualify as an Iterable is __iter__, thus this is the only method defined on the ABC. If you want your Iterable to be usable in a boolean test, add the __bool__ method to your concrete class. In

[issue25512] apparent memory leak using ctypes

New submission from Martin Smith: Attached file demonstrates a memory leak arising in ctypes (I think). Leak occurs in both 2.7 and 3.5, though at different rates. Humble apologies if I've screwed up somewhere. Very nervous about submitting this. -- components: ctypes files:

[issue24778] mailcap.findmatch: document shell command Injection danger in filename parameter

Bernd Dietzel added the comment: My patch for mailcap.py. Please check and apply my patch please. 1) I have removed the os.system() calls for security reasons. 2) New "findmtach_list()" function witch returns the commandline as a [list] witch can be passed to subprocess instead of passing it

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

Laura Creighton added the comment: Note that the full path name of the file that is doing the shadow is of important interest to teachers. We get machines will all sorts of garbage written on them -- the amount of hell caused by a misnamed turtle.py is hard to measure -- and anything as helps

[issue25501] Use async/await through asyncio docs

Changes by Brett Cannon : -- resolution: -> wont fix status: open -> closed ___ Python tracker ___

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

New submission from Mark Roseman: When we create e.g. string.py that shadows a stdlib module needed by IDLE, it would be nice if a better error message could be shown, pointing to that cause. Original message: lac at smartwheels:~/junk$ echo "print ('hello there')" >string.py lac at

[issue25501] Use async/await through asyncio docs

Guido van Rossum added the comment: Honestly I think it's better if most people keep using coroutine/yield-from instead of async/await for a few more releases; their code will be more portable, since it takes forever to update old datacenters. We put in async/await with an eye towards the

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

Laura Creighton added the comment: s/machines will/machines with/ (I was tired.) -- ___ Python tracker ___

[issue25041] document AF_PACKET socket address format

Camilla Montonen added the comment: Provided patch provides documentation for the AF_PACKET address_family. -- keywords: +patch nosy: +Winterflower Added file: http://bugs.python.org/file40898/af_packet_doc.patch ___ Python tracker

[issue25515] Always use os.urandom for generating uuid4s

Alex Gaynor added the comment: (Note that the speed difference would be even bigger on a recent python, 2.7.3 was before the file descriptor was cached for os.urandom) -- ___ Python tracker

[issue25515] Always use os.urandom for generating uuid4s

New submission from Alex Gaynor: Right now uuid4 can be implemented one of 3 ways: - If there's a libuuid (and it's not OS X's) it uses that. - Fallback to os.urandom - If that raises an exception, fall back to the random module I propose to simplify this to _just_ use os.urandom always.

[issue25515] Always use os.urandom for generating uuid4s

Barry A. Warsaw added the comment: On Oct 29, 2015, at 10:30 PM, Alex Gaynor wrote: >Right now uuid4 can be implemented one of 3 ways: If you're hacking on the uuid module, keep in mind issue22807 -- nosy: +barry ___ Python tracker

[issue25515] Always use os.urandom for generating uuid4s

Changes by John Mark Vandenberg : -- nosy: +John.Mark.Vandenberg ___ Python tracker ___ ___

[issue25516] threading.Condition._is_owned() is wrong when using threading.Lock

New submission from Nir Soffer: When using threading.Lock, threading.Condition._is_owned is assuming that the calling thread is owning the condition lock if the lock cannot be acquired. This check is completely wrong if another thread owns the lock. >>> cond =

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

Terry J. Reedy added the comment: Laura, I did not understand from your python-list post that you were complaining about shadowing disabling IDLE itself, as opposed to user code. That is partly because you followed up on Peter Otten complaining because IDLE tries to run user code the same

[issue25498] Python 3.4.3 core dump with simple sample code

Martin Panter added the comment: Before the Issue 22896 changes, PyObject_AsWriteBuffer() was used, which requests a buffer with the PyBUF_WRITABLE flag. Currently we use “w*” argument parsing format, which also uses PyBUF_WRITABLE. (Incidentally, I suspect the C-contiguity check is redundant

[issue11796] Comprehensions in a class definition mostly cannot access class variable

Anne Rosa Wangnick added the comment: With https://bugs.python.org/issue13557 closed, this one should be reopened to amend 6.2.8 Generator expressions: "[However, the leftmost for clause is immediately evaluated] (in the local context)[, so that an error produced by it can be seen ...]"

[issue25156] shutil.copyfile should internally use os.sendfile when possible

desbma added the comment: I played a bit with Unix domain sockets, and it appears you can not open them like a file with open(). So they do no work with the current implementation of shutil.copyfile anyway. -- ___ Python tracker

[issue25482] signal.set_wakeup_fd() doesn't work if the signal don't have handler

Benjamin Peterson added the comment: The OP probably wants something like the Linux-specific signalfd() syscall. -- ___ Python tracker ___

[issue25506] test_pprint reuses test_user_dict

New submission from John Mark Vandenberg: test_pprint defined test_user_dict twice, once for UserDict, and the second for UserList. -- components: Library (Lib), Tests files: test_pprint-fix.diff keywords: patch messages: 253663 nosy: John.Mark.Vandenberg, serhiy.storchaka priority:

[issue25505] undefined name 'window' in Tools/scripts/fixdiv.py

Changes by John Mark Vandenberg : -- nosy: +serhiy.storchaka ___ Python tracker ___ ___

[issue25498] Python 3.4.3 core dump with simple sample code

Changes by eryksun : Added file: http://bugs.python.org/file40890/ctypes_from_buffer_2.patch ___ Python tracker ___

[issue25498] Python 3.4.3 core dump with simple sample code

Changes by Serhiy Storchaka : -- assignee: -> serhiy.storchaka priority: normal -> high stage: needs patch -> patch review ___ Python tracker

[issue25506] test_pprint reuses test_user_dict

Roundup Robot added the comment: New changeset 0e0f3f5195f8 by Serhiy Storchaka in branch '3.5': Issue25506: Fixed a copy-pasting error in test_pprint. https://hg.python.org/cpython/rev/0e0f3f5195f8 New changeset 30b6c3bbefc7 by Serhiy Storchaka in branch 'default': Issue25506: Fixed a

[issue25506] test_pprint reuses test_user_dict

Serhiy Storchaka added the comment: Thank you John. -- assignee: -> serhiy.storchaka resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker

[issue24765] Move .idlerc to %APPDATA%\IDLE on Windows

Terry J. Reedy added the comment: I do not consider the presence of .idlec in $HOME on Windows to be a problem in itself. There are numerous applications, unix-derived and otherwise, that do the same with config files. Very few, if any, have the issue of sharing between multiple

[issue25504] undefined name 'modules' in Tools/freeze/checkextensions_win32.py

New submission from John Mark Vandenberg: All versions of Tools/freeze/checkextensions_win32.py have had an error due to accessing 'modules.sourceFiles' instead of 'module.sourceFiles'. https://hg.python.org/cpython/diff/8e9d5e5103f5/Tools/freeze/checkextensions_win32.py#l1.96 Presumably

[issue25502] unnecessary re-imports

Roundup Robot added the comment: New changeset dec0a26d0b4d by Benjamin Peterson in branch '3.5': remove duplicated imports (closes #25502) https://hg.python.org/cpython/rev/dec0a26d0b4d New changeset 695c50c8cc92 by Benjamin Peterson in branch 'default': merge 3.5 (#25502)

[issue25503] inspect.getdoc does find inherited property __doc__

Serhiy Storchaka added the comment: Thank you John for your report and your patch. Committed tests are slightly changed because "contradiction" was originally purposed to test properties. -- assignee: -> serhiy.storchaka resolution: -> fixed stage: -> resolved status: open ->

[issue25505] undefined name 'window' in Tools/scripts/fixdiv.py

Changes by John Mark Vandenberg : Added file: http://bugs.python.org/file40889/fixdiv-remove-truncate.diff ___ Python tracker ___

[issue25505] undefined name 'window' in Tools/scripts/fixdiv.py

New submission from John Mark Vandenberg: Tools/scripts/fixdiv.py 's `FileContext.truncate` has used `window` since it was created, when it is supposed to use self.window. https://hg.python.org/cpython/annotate/60f290a7eae8/Tools/scripts/fixdiv.py#l195 `truncate` is unused, so an alternative

[issue13828] Further improve casefold documentation

Mark Summerfield added the comment: I think the str.casefold() docs are fine as far as they go, rightly covering what it _does_ rather than _how_, yet providing a reference for the details. But what they lack is more complete information. For example I discovered this: >>> x = "files and

[issue25503] inspect.getdoc does find inherited property __doc__

Roundup Robot added the comment: New changeset bbf00faf25ff by Serhiy Storchaka in branch '3.5': Issue #25503: Fixed inspect.getdoc() for inherited docstrings of properties. https://hg.python.org/cpython/rev/bbf00faf25ff New changeset e80d1e9737d4 by Serhiy Storchaka in branch 'default': Issue

[issue25506] test_pprint reuses test_user_dict

John Mark Vandenberg added the comment: Sorry; the first patch includes unrelated fixes. -- Added file: http://bugs.python.org/file40892/test_pprint-fix.diff ___ Python tracker

[issue8231] Unable to run IDLE without write-access to home directory

Terry J. Reedy added the comment: OK: call GetUserCfgDir once on creation of idleConf instance and set .userdir attribute. Replace repeated calls in PyShell and Editor with attributes accesses. I tested that, with patch, IDLE starts and rewrites both breakpoints.lst and recent-files.lst as

[issue25156] shutil.copyfile should internally use os.sendfile when possible

desbma added the comment: Thanks for the comment. > Also, the os.sendfile() doc suggests that some platforms only support writing > to sockets, so I definitely think a backup plan is needed. You are right, the man page clearly says: > Applications may wish to fall back to read(2)/write(2) in

[issue25507] IDLE: user code 'import tkinter; tkinter.font' should fail

New submission from Terry J. Reedy: In python, 'import tkinter; tkinter.font' fails with AttributeError. In IDLE, it does not, which is a bug, This lead to a Stackoverflow question 'Why does my code run in IDLE but not with Python'? The issue is that importing modules in a package has the

[issue25508] LogRecord attributes are not tuple, when logging only dict

New submission from Ondrej Sejvl: Hi, in doc https://docs.python.org/3.4/library/logging.html#logrecord-attributes there is argsYou shouldn’t need to format this yourself. The tuple of arguments merged into msg to produce message. But when I log message with 1 arg - a dict - in

[issue24765] Move .idlerc to %APPDATA%\IDLE on Windows

eryksun added the comment: On Windows, how about creating a junction (_winapi.CreateJunction in 3.5, or the shell's mklink /j) from ~\.idlerc to %APPDATA%\idle? If ~\.idlerc already exists, it could be moved to %APPDATA%\idle before creating the junction. New code would directly use

[issue25444] Py Launch Icon

Nils Lindemann added the comment: i see now what you mean. I have this icon too (py 3.5.0 idle icon.png). Something is misconfigured with the icons in the python 3.5.0 windows installer, also python files in explorer have no icons. Thanks for your hint with the python ideas list, i will post

[issue8231] Unable to run IDLE without write-access to home directory

Terry J. Reedy added the comment: I cannot currently think of any reason why the subprocess *needs* to access idleConf, so I want to put this on hold while investigating whether the accesses can be eliminated as part of #25507. -- dependencies: +IDLE: user code 'import tkinter;

[issue25507] IDLE: user code 'import tkinter; tkinter.font' should fail

Terry J. Reedy added the comment: #8231 will benefit if indirect imports of configHandler and idleConf into the user process can be eliminated. I do not currently believe it is needed. Not importing PyShell will go part way. AutoComplete imports idleConf to get the popup delay. run imports

[issue8231] Unable to run IDLE without write-access to home directory

Terry J. Reedy added the comment: Without being able to make my homedir read-only, I don't, of course, see the messages. Since run does not use idleConf (I am now sure), you should have been able to proceed after clicking away the admittedly obnoxious repeat messages. The old extraneous

[issue25482] signal.set_wakeup_fd() doesn't work if the signal don't have handler

Atsuo Ishimoto added the comment: Thank you for good sample. This should be written in PEP 475. But installing new signal handler can cause different behavior. As in my previous example, if other 3rd party libraries(ncurses in my case) need to handle the signal, signal is consumed by Python

[issue25482] signal.set_wakeup_fd() doesn't work if the signal don't have handler

STINNER Victor added the comment: It's very easy to get the same behaviour on Python 3.4 (without PEP 475) and Python 3.5 (with PEP 475). Just add the following code at the top of your example: import signal def sighandler(signum, frame): raise InterruptedError

[issue25507] IDLE: user code 'import tkinter; tkinter.font' should fail

Terry J. Reedy added the comment: The subprocess command is built in PyShell.ModifiedInterpreter method build_subprocess_arglist (l.404). It imports run and calls run.main(args). The test for this issue would be to import run and check that sys.modules does not contain configHandler and that

[issue25495] binascii documentation incorrect

Martin Panter added the comment: Thanks for bringing this up, it has often bugged me. My understanding, based on the original wording and places where this is used, is that the data is often pre-processed into 57-byte chunks, rather than post-processing it. Maybe it is worthwhile restoring

[issue25515] Always use os.urandom for generating uuid4s

Roundup Robot added the comment: New changeset 24bdc4940e81 by Benjamin Peterson in branch '2.7': always use os.urandom for the uuid4 algorithm (closes #25515) https://hg.python.org/cpython/rev/24bdc4940e81 New changeset 70be1f9c9255 by Benjamin Peterson in branch '3.5': always use os.urandom

[issue25504] undefined name 'modules' in Tools/freeze/checkextensions_win32.py

Roundup Robot added the comment: New changeset 31fd613a40aa by Benjamin Peterson in branch '2.7': fix usage of undefined name (#25504) https://hg.python.org/cpython/rev/31fd613a40aa New changeset 114fb81a08e3 by Benjamin Peterson in branch '3.4': fix usage of undefined name (#25504)

[issue25504] undefined name 'modules' in Tools/freeze/checkextensions_win32.py

Changes by Benjamin Peterson : -- resolution: -> fixed status: open -> closed ___ Python tracker ___

[issue25507] IDLE: user code 'import tkinter; tkinter.font' should fail

Terry J. Reedy added the comment: To continue: CallTips.Calltips has a similar here-or-there logic that decides where to execute the actual get-calltip-logic. In this case, get_entity and get_argspec are already functions rather than methods (I changed get_entity from method to function in

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

Terry J. Reedy added the comment: "I think that removing '' from sys.path, so that IDLE can run, is better than a nicer warning that it cannot run. This, of course, requires that sys not be shadowed, so that sys.path can be accessed." So after importing sys, we should check sys.__file__, or

[issue25514] better startup error messages in IDLE when stdlib modules shadowed

Terry J. Reedy added the comment: Either tkinter and tkinter.messagebox or subprocess must be imported to display a message when IDLE is started from an icon, and there is no console. With subprocess, '''python -i -c "print('Cannot start IDLE (or whatever)')"''' should work. If nothing

[issue25515] Always use os.urandom for generating uuid4s

Donald Stufft added the comment: This looks like a good idea to me, faster and more secure seems like a total win. -- ___ Python tracker ___

[issue25516] threading.Condition._is_owned() is wrong when using threading.Lock

Nir Soffer added the comment: The issue was introduced in this commit: commit 8cb1ccbb8b9ed01c26d2c5be7cc86682e525dce7 Author: Guido van Rossum Date: Thu Apr 9 22:01:42 1998 + New Java-style threading module. The doc strings are in a separate module. --

[issue25512] apparent memory leak using ctypes

Martin Panter added the comment: I am running your script with the Python 3.5.0 from Arch Linux on an x86-64 computer. The reported numbers slowly increase and then plateau at 221 million: $ python ctype_stressor.py 1 0.000e+00 3.8010880e+07 10 2.8538793e+00 1.4648934e+08

[issue25516] threading.Condition._is_owned() is wrong when using threading.Lock

Changes by Nir Soffer : -- keywords: +patch Added file: http://bugs.python.org/file40900/0001-Issue-25516-threading.Condition._is_owned-fails-when.patch ___ Python tracker