[issue6721] Locks in python standard library should be sanitized on fork

Richard Oudkerk added the comment: > conn = MySQLConn() > start_thread1(conn) > start_thread2(conn): > while True: >if os.fork() == 0: # child > raise Exception('doom') # triggers destructor There is no guarantee here that the lock will be held at the time of the fork. So even

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Added file: http://bugs.python.org/file25784/3665.patch ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Added file: http://bugs.python.org/file25783/re_unicode_escapes.diff ___ Python tracker ___ ___ Python-bugs-list

[issue6721] Locks in python standard library should be sanitized on fork

Gregory P. Smith added the comment: We could make any later attempt to acquire or release a lock that was reinitialized while it was held raise an exception. Such exception raising behavior should be conditional at lock construction time; some code (such as logging) never wants to deal with o

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Removed file: http://bugs.python.org/file25782/3665.patch ___ Python tracker ___ ___ Python-bugs-list mailing lis

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Removed file: http://bugs.python.org/file25781/re_unicode_escapes.diff ___ Python tracker ___ ___ Python-bugs-lis

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Added file: http://bugs.python.org/file25782/3665.patch ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue3665] Support \u and \U escapes in regexes

Changes by Serhiy Storchaka : Added file: http://bugs.python.org/file25781/re_unicode_escapes.diff ___ Python tracker ___ ___ Python-bugs-list

[issue3665] Support \u and \U escapes in regexes

Serhiy Storchaka added the comment: I don't think it is worth to target it for 2.7 and 3.2 (it's new feature, not bugfix), but for 3.3 it will be very useful. Since PEP 393 conversion to the surrogate pairs is no longer relevant. -- components: +Regular Expressions, Unicode nosy: +sto

[issue14973] restore python2 unicode literals in "ur" strings

Serhiy Storchaka added the comment: See issue3665. -- nosy: +storchaka title: restore python2 unicode literals in "ru" strings -> restore python2 unicode literals in "ur" strings ___ Python tracker __

[issue14965] super() and property inheritance behavior

josmiley added the comment: >>> class DerivedProp(BaseProp): ... @property ... def p(self): ... return super(DerivedProp, self).p * 2 ... @p.setter ... def p(self, value): ... BaseProp.p.__set__(self,value / 2) -- nosy: +josmiley

[issue14673] add sys.implementation

Changes by Eric Snow : Removed file: http://bugs.python.org/file25779/issue14673_full.diff ___ Python tracker ___ ___ Python-bugs-list mailing

[issue14673] add sys.implementation

Eric Snow added the comment: Guess I should have tested the docs before posting the patch. ;) -- Added file: http://bugs.python.org/file25780/issue14673_full.diff ___ Python tracker ___

[issue14007] xml.etree.ElementTree - XMLParser and TreeBuilder's doctype() method missing

Roundup Robot added the comment: New changeset a29ae1c2b8b2 by Eli Bendersky in branch 'default': Issue #14007: make XMLParser a real subclassable type exported from _elementtree. +cleanups http://hg.python.org/cpython/rev/a29ae1c2b8b2 -- ___ Python

[issue14673] add sys.implementation

Eric Snow added the comment: This patch incorporates types.SimpleNamespace and makes the repr show all attributes. Tests and docs for types.SimpleNamespace are also included. The patch skips 3 tests (which I added) that are failing. One has a segfault due to a recursive repr (where I had a

[issue14958] IDLE 3 and PEP414 - highlighting unicode literals

Roger Serwy added the comment: Attached is a backport to 2.7. -- keywords: +patch versions: +Python 2.7 Added file: http://bugs.python.org/file25778/issue14958_27.patch ___ Python tracker _

[issue14974] rename packaging.pypi to packaging.index

Éric Araujo added the comment: We could use “catalog” (see how the interest group for PyPI is named catalog-sig), but it sounds a bit abstract to me, and at the same time fails to convey that this module deals with finding things (in an index/catalog) and also download them (from PyPI or thir

[issue14968] Section "Inplace Operators" of :mod:`operator` should be a subsection

Éric Araujo added the comment: Good catch, +1 for the patch. -- nosy: +eric.araujo, sandro.tosi stage: -> commit review versions: +Python 2.7, Python 3.2 ___ Python tracker ___

[issue14965] super() and property inheritance behavior

Changes by Éric Araujo : -- nosy: +eric.araujo stage: -> needs patch versions: +Python 3.3 ___ Python tracker ___ ___ Python-bugs-lis

[issue14950] Make packaging.install API less confusing and more extensible

Éric Araujo added the comment: BTW I suggest you work off a d2 checkout; I have not ported all changes from Montreal sprints to cpython/p7g at the moment. -- title: packaging.install does not have a clear API. -> Make packaging.install API less confusing and more extensible _

[issue14950] packaging.install does not have a clear API.

Éric Araujo added the comment: To summarize the high-level API needed: install from an index (argument is a version requirement as string), install archive file, install from directory (containing setup.cfg or setup.py). When #8668 is implemented, an “editable” option will be nice too. (Need

[issue14957] Improve docs for str.splitlines

Changes by Éric Araujo : -- keywords: +easy nosy: +eric.araujo stage: -> needs patch ___ Python tracker ___ ___ Python-bugs-list mail

[issue14499] Extension module builds fail with Xcode 4.3 on OS X 10.7 due to SDK move

Changes by Éric Araujo : -- nosy: +eric.araujo ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue6721] Locks in python standard library should be sanitized on fork

lesha added the comment: Re "threading locks cannot be used to protect things outside of a single process": The Python standard library already violates this, in that the "logging" module uses such a lock to protect the file/socket/whatever, to which it is writing. If I had a magic wand that

[issue6721] Locks in python standard library should be sanitized on fork

Gregory P. Smith added the comment: threading locks cannot be used to protect things outside of a single process. Any code using them to do that is broken. In your examples you are suggesting a class that wants to do one or more mysql actions within a destructor and worried that the __del__

[issue6721] Locks in python standard library should be sanitized on fork

lesha added the comment: Deadlocks are dandy, but corruption is cruel. -- ___ Python tracker ___ ___ Python-bugs-list mailing list Uns

[issue6721] Locks in python standard library should be sanitized on fork

Gregory P. Smith added the comment: Anyone using a preexec function in subprocess has already declared that they like deadlocks so that isn't an issue. :) -- ___ Python tracker

[issue14969] Use __suppress_context__ in contextlib.ExitStack.__exit__

Nick Coghlan added the comment: Electronic contributor forms are actually on the PSF wishlist. We'll get there some day (maybe after the website update). Getting back to the problem at hand, I think you might be on to something with the idea of exploiting PEP 409 to handle this. Specifically,

[issue6721] Locks in python standard library should be sanitized on fork

lesha added the comment: > I think forkall() on Solaris acts like that, but the normal fork() > function does not. Only the thread which performs fork() will survive > in the child process. Sorry, brain fail. A slightly more contrived failure case is this: subprocess.Popen( ..., preexe

[issue14975] import unicodedata, DLL load failed on Python 2.7.3

Martin v. Löwis added the comment: I do not believe that this is a problem in Python. Instead, it appears that your Python installation got corrupted somehow. I recommend to run a virus scanner. -- nosy: +loewis ___ Python tracker

[issue6721] Locks in python standard library should be sanitized on fork

Richard Oudkerk added the comment: > a) fork() is called with the DB lock held by thread1. > b) Some time passes before the child gets to exec(). > c) In that time, the child's thread2 gets to doWork(). > d) Simultaneously, the parent's doWork is still running and holding a lock. > e) Thanks to

[issue14959] ttk.Scrollbar in Notebook widget freezes

Ned Deily added the comment: Another data point: I just tried the same test using a Python 3.2.3 linked with an X11 Tk 8.5 (MacPorts) rather than the ActiveState Cocoa Tk 8.5.11 (python.org Python). I was not able to get the test to fail with X11 Tk so that tends to support the idea that the

[issue14971] (unittest) loadTestsFromName does not work on method with a decorator

Michael Foord added the comment: Whilst functools.wraps would fix the problem it still sounds like a bug (or at the very least a reasonable feature request). -- ___ Python tracker

[issue14972] listcomp with nested classes

Changes by Hynek Schlawack : -- nosy: +hynek ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python

[issue6721] Locks in python standard library should be sanitized on fork

lesha added the comment: I am really alarmed by the reinit_locks patches. I scanned the comment history, and looked at the patch. I may have missed something, but it looks to me like the basic behavior is this: "After fork(), all locks are replaced by brand-new lock objects that are NOT lock

[issue14975] import unicodedata, DLL load failed on Python 2.7.3

New submission from YF : Before I use the Python 2.7.2 on Windows XP, today I tried to upgrade to the Python 2.7.3, but encountered a problem. When running any .py file encountered "import unicodedata" or directly run the command always returns: Traceback (most recent call last): .. imp

[issue14974] rename packaging.pypi to packaging.index

Hynek Schlawack added the comment: While I can see your point, I think that "index" is way too generic. I also think that the pypi term is overloaded with both meanings. -- nosy: +hynek ___ Python tracker ___

[issue14814] Implement PEP 3144 (the ipaddress module)

Roundup Robot added the comment: New changeset bd2c2def77a7 by Hynek Schlawack in branch 'default': #14814: Remove stale __hex__ method from ipaddress http://hg.python.org/cpython/rev/bd2c2def77a7 -- ___ Python tracker

[issue14974] rename packaging.pypi to packaging.index

New submission from Alexis Metaireau : PyPI is the name of a particular index, whereas "index" is a generic term. So ISTM that it would be better to use the latter, semantically-wise. -- assignee: alexis components: Distutils2 messages: 162027 nosy: alexis, tarek priority: normal severi

[issue14972] listcomp with nested classes

R. David Murray added the comment: This is doubtless a result of the way the class namespace scope is handled, coupled with the fact that in Python3 list comprehensions have a local scope. The class scope is a somewhat unique beast. I agree that this is unfortunate, but I have a feeling tha

[issue14973] restore python2 unicode literals in "ru" strings

New submission from rurpy the second : PEP 414 proposes restoring the "u" string prefix (semantically as a "noop") to make porting from Python2 easier. I would like to propose that "ru"-strings also interpret embedded "\u" unicode literals in the python2 fashion (as a single unicode chara

[issue14972] listcomp with nested classes

Westley Martínez added the comment: $ python Python 3.2.3 (default, Apr 23 2012, 23:35:30) [GCC 4.7.0 20120414 (prerelease)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> class A: ... x = 42 ... y = [x for _ in '1'] ... Traceback (most recent call

[issue14972] listcomp with nested classes

Changes by Alex Gaynor : -- nosy: +alex ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/

[issue14972] listcomp with nested classes

Florent Xicluna added the comment: Simpler test case: class A: x = 42 y = [x for _ in '1'] The semantics of list comprehension changed with Python 3. However, I do not see this specific behavior documented somewhere. http://docs.python.org/dev/whatsnew/3.0.html#changed-syntax

[issue14972] listcomp with nested classes

New submission from josmiley : # this runs with python2.7, not with python3.2 class Foo(object): class Bar(object): pass Attr = [Bar()for n in range(10)] # solved in this way ... class Foo(object): class Bar(object): pass Attr = [] for n in range(10): Attr

[issue14969] Restore sys.exc_clear()?

alon horev added the comment: Another possible solution is to explicitly set an exception's __supress_context__ attribute to False (right now impossible because it's the default value). If a user can 'turn on' the flag when attaching a different exception (raise X from Y), why not allow 'tur

[issue14635] telnetlib uses select instead of poll - limited to FD_SETSIZE fds

Akintayo Holder added the comment: Hi, telnet.read_until() and telnet.expect() will use select.poll() instead of select.select() on systems where poll() is available. The patch also includes updates to test_telnetlib, the read_until() tests were changed to test the case where poll() is unava

[issue6721] Locks in python standard library should be sanitized on fork

Richard Oudkerk added the comment: Attached is an updated version of Charles-François's reinit_locks.diff. Changes: * Handles RLock by assuming that if self->count != 0 when we acquire the lock, then the lock must have been reinitialized by PyThread_ReInitLocks(). * Applies existing fork t

[issue14814] Implement PEP 3144 (the ipaddress module)

Hynek Schlawack added the comment: It's funny how raising the test coverage _always_ uncovers lurking bugs in obscure branches. :) Patch attached, let me know if I got it wrong – would commit otherwise. A few quick side-note: does the __version__ variable make sense in stdlib? -- Add

[issue14971] (unittest) loadTestsFromName does not work on method with a decorator

R. David Murray added the comment: I don't think this is documented anywhere (and should be). I believe what you need to do is use functools.wraps on your wrapper function. -- assignee: -> docs@python components: +Documentation -None nosy: +docs@python, michael.foord, r.david.murray

[issue14938] 'import my_pkg.__init__' creates duplicate modules

Ronan Lamy added the comment: That would force the Loaders to know how to convert a module name into a file path, which isn't the case now since FileLoader.get_filename() is just a shim that returns self.path. So I'd rather add an optional argument to FileLoader. Actually, I feel that the cle

[issue14971] (unittest) loadTestsFromName does not work on method with a decorator

New submission from Alessandro Piccione : Steps to reproduce the problem. 1. Create a module (ex. name it "test") 2. Create a file in it (ex. mainTest.py) with a class (ex. MainTest) derived from TestCase. 3. Create a test method in that class, (ex. test_base), prefix it with "test". 4. Create

[issue14962] When changing IDLE configuration all text in shell window loses highlighting

Ned Deily added the comment: LGTM, thanks. Applied for release in 2.7.4, 3.2.4, and 3.3.0. -- nosy: +ned.deily resolution: -> fixed stage: -> committed/rejected status: open -> closed versions: +Python 2.7 ___ Python tracker

[issue14962] When changing IDLE configuration all text in shell window loses highlighting

Roundup Robot added the comment: New changeset 1a4e99460438 by Ned Deily in branch '2.7': Issue #14962: Update text coloring in IDLE shell window after changing http://hg.python.org/cpython/rev/1a4e99460438 New changeset 9d0c3a835bfe by Ned Deily in branch '3.2': Issue #14962: Update text color

[issue14970] -v command line option is broken

Brett Cannon added the comment: Not a problem. =) If you want to know where an import originated from, you can probably do something as simple as overload builtins.__import__ with a version that does a quick stack look to see where the previous call is coming from. -- __

[issue14962] When changing IDLE configuration all text in shell window loses highlighting

Roger Serwy added the comment: Attached is a patch to fix this issue. The ModifiedColorDelegator already marks everything before "iomark" as SYNC'ed. This is good, as the ColorDelegator should not be trying to colorize STDOUT text as Python code. Resetting the ColorDelegator in _rmcolorize

[issue14970] -v command line option is broken

Kristján Valur Jónsson added the comment: Oh dear, silly me. I misunderstood the point of -v _and_ misunderstood the output of 2.7 I was quoting. I was looking for a way to find out where the import originated from, not which file would be imported as a result. Import errors, along with pickl

[issue14970] -v command line option is broken

Brett Cannon added the comment: I am setting this as pending since I consider the total output acceptable, but if Kristján has specific issues he wants to bring up or change he still can. -- status: open -> pending ___ Python tracker

[issue14970] -v command line option is broken

Brett Cannon added the comment: The information is still there, just in a different output line (i.e. http://hg.python.org/cpython/file/c7b16e2be71a/Lib/importlib/_bootstrap.py#l735 outputs the same info, just on its own line). I couldn't keep the old format as the code has been shifted aroun

[issue14938] 'import my_pkg.__init__' creates duplicate modules

Brett Cannon added the comment: I see what you mean about the discrepancy, but you don't need to complicate the constructor to get the desired result. If you have is_package() check if the module name ends in __init__ to skip the package check and just say False (e.g. only if the path ends in

[issue14970] -v command line option is broken

Kristján Valur Jónsson added the comment: Yes, I found that. The line in question is, however, this: _verbose_message('import {!r} # {!r}', name, loader) (_bootstrap.py:1254). Unfortunately, I see no way to get at the line from which the import occurred here. The loader itself is not us

[issue14952] Cannot run regrtest with amd64/debug on windows

Changes by Kristján Valur Jónsson : -- resolution: -> fixed status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailin

[issue14952] Cannot run regrtest with amd64/debug on windows

Roundup Robot added the comment: New changeset c7b16e2be71a by Kristjan Valur Jonsson in branch 'default': Issue #14952: Fix incorrect output dll names for win64/debug builds, causing http://hg.python.org/cpython/rev/c7b16e2be71a -- nosy: +python-dev ___

[issue14952] Cannot run regrtest with amd64/debug on windows

Kristján Valur Jónsson added the comment: I've found what is wrong. There is a problem with _multiprocessing.pyd on 64 bit debug builds. Why this manifests itself as it does, I don´t know. Someone must be silencinng the proper import error. this is most likely a build config error, that I wil

[issue14970] -v command line option is broken

Martin v. Löwis added the comment: See http://hg.python.org/cpython/file/default/Lib/importlib/_bootstrap.py for the source of SourceFileLoader. -- nosy: +loewis ___ Python tracker

[issue14970] -v command line option is broken

New submission from Kristján Valur Jónsson : Using the -v command line option to diagnose import problem is no longer useful. In stead of lines like this in version 2.7: import UserDict # from D:\p4\games\branches\development\MAIN\eve\dust\tool\bin/../../../../carbon/src/stackless/Lib/UserDic

[issue14963] Use an iterative implementation for contextlib.ExitStack.__exit__

Nick Coghlan added the comment: Interesting - it turns out we can't fully reproduce the behaviour of nested with statements in ExitStack (see the new reference test I checked in, as well as #14969) I added one technically redundant variable to the implementation to make it more obviously cor

[issue14963] Use an iterative implementation for contextlib.ExitStack.__exit__

Roundup Robot added the comment: New changeset c0c7618762e5 by Nick Coghlan in branch 'default': Close #14963: Use an iterative algorithm in contextlib.ExitStack.__exit__ (Patch by Alon Horev) http://hg.python.org/cpython/rev/c0c7618762e5 -- resolution: -> fixed stage: test needed ->

[issue14963] Use an iterative implementation for contextlib.ExitStack.__exit__

Roundup Robot added the comment: New changeset fc73e6ea9e73 by Nick Coghlan in branch 'default': Issue #14963: Added test cases for contextlib.ExitStack exception handling behaviour (Initial patch by Alon Horev) http://hg.python.org/cpython/rev/fc73e6ea9e73 -- nosy: +python-dev __

[issue14969] Restore sys.exc_clear()?

New submission from Nick Coghlan : When adding the test case for #14963, I discovered that contextlib.ExitStack can't *quite* reproduce the exception handling of nested with statements. The problem arises when the original exception gets suppressed by one of the handlers, but an outer handler

[issue14964] distutils2.utils.resolve_name cleanup

Ronny Pfannschmidt added the comment: i missused hg export, here is a corrected patch -- Added file: http://bugs.python.org/file25773/resolve_name.patch ___ Python tracker ___ _

[issue14968] Section "Inplace Operators" of :mod:`operator` should be a subsection

New submission from Lars Buitinck : The section "Inplace Operators" of the module docs for operator now show up in TOC at http://docs.python.org/dev/library/. I don't think that's intended as it does not describe a separate module. -- assignee: docs@python components: Documentation fil

[issue14964] distutils2.utils.resolve_name cleanup

Ronny Pfannschmidt added the comment: updated the patch with more detailed errors -- Added file: http://bugs.python.org/file25771/resolve_name.patch ___ Python tracker ___ _

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

Ronny Pfannschmidt added the comment: an example of creating a wrong error would be something like the following: there is a package foo.bar, which does a wrong import in __init__.py we want to resolve the name foo.bar.something we'd get the error that foo has no attribute bar -- ___

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

Ronny Pfannschmidt added the comment: to correctly implement it we need the : separator back with the separator the import specification is no longer ambigious, and we can use one exact import, and an error will always be an error -- ___ Python trac

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

Changes by Tarek Ziadé : -- resolution: -> duplicate superseder: -> Improve error reporting for packaging.util.resolve_name ___ Python tracker ___ _

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

Tarek Ziadé added the comment: can you give an example of a bad error ? -- ___ Python tracker ___ ___ Python-bugs-list mailing list U

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

Tarek Ziadé added the comment: the current code works as expected. Why not leaving it like this since your change seem to be comsetics only ? -- ___ Python tracker ___

[issue14967] distutils2.utils.resolve_name cannot be implemented to give correct errors in all situations

New submission from Ronny Pfannschmidt : due to the lack of a marker that denotes where the module ends and the attribute starts, unrelated import errors can break the try&error chain at unexpected places and the code can pass on to the recursive getattr chain, giving a completely different er

[issue14964] distutils2.utils.resolve_name cleanup

Ronny Pfannschmidt added the comment: my change is a unrelated cleanup but the ideas in the patch look good the attribute error addition seems relevant i'll adapt my patch to raise the original import error for toplevel import failure and accumuplate the attrbute name so it can raise the corr

[issue14909] Fix incorrect use of *Realloc() and *Resize()

Changes by Kristján Valur Jónsson : -- resolution: -> fixed status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailin

[issue14909] Fix incorrect use of *Realloc() and *Resize()

Roundup Robot added the comment: New changeset 588ea940e5e3 by Kristjan Valur Jonsson in branch 'default': Issue #14909: A number of places were using PyMem_Realloc() apis and http://hg.python.org/cpython/rev/588ea940e5e3 -- nosy: +python-dev ___ Pyt

[issue14007] xml.etree.ElementTree - XMLParser and TreeBuilder's doctype() method missing

Eli Bendersky added the comment: * Another problem: the C implementation of XMLParser does not take 3 positional args, but only 2. Although the 'html' arg is documented as "unsupported", it should still be taken and silently ignored, similarly to the Python version -- ___

[issue14899] Naming conventions and guidelines for packages and namespace packages

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue14963] Use an iterative implementation for contextlib.ExitStack.__exit__

alon horev added the comment: that was indeed trickier, but overriding the __context__ attribute did the trick. -- Added file: http://bugs.python.org/file25770/14963.2.patch ___ Python tracker ___

[issue14945] Setup & Usage documentation for selected stdlib modules

Changes by A.M. Kuchling : -- nosy: +akuchling ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue14944] Setup & Usage documentation for pydoc, idle & 2to3

Changes by A.M. Kuchling : -- nosy: +akuchling ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue10376] ZipFile unzip is unbuffered

Serhiy Storchaka added the comment: The patch updated to reflect Martin's stylistic comments. Sorry for the delay, Martin. I have not received an email with your review from 2012-05-13, and only today accidentally discovered your comments in Rietveld. It seems to have been some bug in Rietvel