[issue11906] test_argparse failure in interactive mode

Terry J. Reedy added the comment: http://docs.python.org/devguide/runtests.html "If you don’t have easy access to a command line, you can run the test suite from a Python or IDLE shell: >>> from test import autotest" However, argparse is the least of the test suite problems on Windows. (#9116,

[issue9116] test_capi.test_no_FatalError_infinite_loop crash on Windows

Terry J. Reedy added the comment: I just got 'fatal error' running python_d 3.3.a0 checked out yesterday. When I closed message, test continued (but really crashed later). -- nosy: +terry.reedy ___ Python tracker

[issue11990] redirected output - stdout writes newline as \n in windows

Peter Csapo added the comment: I have having the same issues as Jimbofbx. This seems to stem from changes due to issue 10841. All stdio is now opened in binary mode, in consideration that it is the TextIOWrapper's job to do endline translation. The problem here is that the newline mode = '\n'

[issue13741] *** glibc detected *** python: double free or corruption (!prev): 0x0000000001d53ad0 ***

New submission from Andrew : Hello, This program crashes after 12-24 hours of running. My OS is Ubuntu 11.10, I'm using Python 2.7.2, and gcc 4.6.1. Here's the error: *** glibc detected *** python: double free or corruption (!prev): 0x01d53ad0 *** === Backtrace: = /lib/x8

[issue13734] Add a generic directory walker method to avoid symlink attacks

Ross Lagerwall added the comment: > I'm currently leaning towards the simple 4-tuple approach I would also take that approach. It seems simplest to me. -- ___ Python tracker __

[issue13737] bugs.python.org/review's Django settings file DEBUG=True

Bithin A added the comment: The bugs.python.org/review is a running application and it is very bad to see debug error messages. -- ___ Python tracker ___ __

[issue13721] ssl.wrap_socket on a connected but failed connection succeeds and .peer_certificate gives AttributeError

Mads Kiilerich added the comment: I would find ValueError surprising here. socket.error or SSLError would be less surprising ... even though it technically isn't completely true. But isn't it more like a kind of RuntimeError to call getpeercert after the socket has been closed? -- _

[issue13734] Add a generic directory walker method to avoid symlink attacks

Nick Coghlan added the comment: Thanks for that Charles-François - do you mind if I adapt that for walkdir? The changes I would make are basically those that Antoine pointed out: - rather than replacing the dirpath entry, instead yield a 4-tuple that appends the dirfd attribute at the end* - i

[issue13740] winsound.SND_NOWAIT ignored on modern Windows platforms

New submission from Jelle Geerts : As the attached example Python script explains, the winsound.SND_NOWAIT flag doesn't do anything, at least not on modern Windows platforms. The updated MSDN documentation for PlaySound() states that SND_NOWAIT is not supported (it is ignored). Link: http://m

[issue13521] Make dict.setdefault() atomic

Filip Gruszczyński added the comment: Thanks, I'll try that. Like before I will probably find time next weekend. -- ___ Python tracker ___ __

[issue13521] Make dict.setdefault() atomic

Antoine Pitrou added the comment: > Do you have any idea, how I might this part? I mean how to check, that > this is called only once? Checking that __eq__ is called only once should be a good proxy. -- ___ Python tracker

[issue13642] urllib incorrectly quotes username and password in https basic auth

Michele Orrù added the comment: There's no need to port your patch over python3k, since urllib behaves differently with http passwords - as you can see in the doc http://docs.python.org/dev/py3k/library/urllib.request.html#examples I would be glad to finish your password on 2.7 as soon as po

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Terry J. Reedy added the comment: To make my idea work both now and after we switch docs.python.org, that url should not be used as either of the first two links. Rather there should be a docs.python.org/py2k (or whatever) that is the permanent 'latest Python 2 release' docs link. --

[issue13717] print fails on unicode '\udce5' surrogates not allowed

Atle Pedersen added the comment: Just wanted to say thanks for very fast response, and informative information. I respect your decision to close the bug as invalid. But my five cent is that it still feels like a bug, something that shouldn't happen. Especially since it's part of a very basic

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Georg Brandl added the comment: Sounds good. -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.py

[issue13521] Make dict.setdefault() atomic

Filip Gruszczyński added the comment: I understand you are talking about this call: mp->ma_lookup(mp, key, hash); I haven't noticed that earlier. I'll try to provide a better fix (for 2.7 first, after we agree it's good enough, I will provide one for 3.3). Do you have any idea, how I might t

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Terry J. Reedy added the comment: My specific suggestion is that the sidebar be time-independent (for at least a decade, until Python 4 ;-) and say Docs for other versions Current Python 3 Current Python 2 In development Everything else with stable links for each. -- ___

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Terry J. Reedy added the comment: Well, actually I do have an idea. When 3.3 comes out, I think docs.python.org should point to that, and another url should point to the most recent 2.7.x release. The url for all versions is already stable. Then the sidebar could have entries for most recent

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Antoine Pitrou added the comment: > Nobody said 3.2 was not stable... Well, the sidebar says it's in development while 2.7 is stable :) -- nosy: +pitrou ___ Python tracker ___

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Terry J. Reedy added the comment: http://docs.python.org/release/2.6.7/ has the same problem that the 3.x 'in development' version is called 3.2, while there is no link to current stable 3.2.2 except indirectly by going to 'old versions' (which perhaps should be 'other versions'). I do not kno

[issue13737] bugs.python.org/review's Django settings file DEBUG=True

Martin v. Löwis added the comment: I disagree that this is a security issue, or an issue at all. All source code of the site is in a public subversion repository, available for review to any attacker (as well as any security review) - and that is deliberately so because we fundamentally belie

[issue13026] Dis module - documentation of MAKE_FUNCTION

Terry J. Reedy added the comment: "For each keyword only default" is not in my text, but if you are referring to "keyword only name, default parameter object pairs ", yes. Here is a revised suggested wording: MAKE_FUNCTION(argc) Pushes a new function object on the stack. From bottom to top, t

[issue12760] Add create mode to open()

Charles-François Natali added the comment: > Just a small note: FileExistsError is raised, not exactly OSError, > when the file exists. I've updated the doc accordingly. -- stage: patch review -> commit review ___ Python tracker

[issue8184] multiprocessing.managers will not fail if listening ocket already in use

Charles-François Natali added the comment: Here's a patch addressing the multiple bind() problem on Windows. Note that this problem also affects other parts of the stdlib, which use SO_REUSEADDR when available. Also, there's an rather confusing comment in support.find_unused_port(): """

[issue13739] os.fdlistdir() is not idempotent

Charles-François Natali added the comment: > For some reason, the second changeset broke the OpenIndiana buildbots: > I have absolutely no idea of why this doesn't work. I suspect rewinddir() is a noop on OpenIndiana if readdir() hasn't been called. I'll revert this commit. > Also, wouldn't it

[issue13739] os.fdlistdir() is not idempotent

Antoine Pitrou added the comment: For some reason, the second changeset broke the OpenIndiana buildbots: http://www.python.org/dev/buildbot/all/builders/AMD64%20OpenIndiana%203.x/builds/2485 Also, wouldn't it be better to call rewinddir with the GIL released? (although I agree rewinddir shouldn

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Antoine Pitrou added the comment: Closing the issue as it's really a quirk rather than a bug. Thanks for reporting, though. -- resolution: -> wont fix stage: -> committed/rejected status: open -> closed ___ Python tracker

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Kay Hayen added the comment: I see, if it's refcount dependent, that explains why it changes from interpreter provided dictionary and self-created one. So, I take, I should always call "pickletools.optimize( cPickle.dumps( value ))" then. Thanks, Kay --

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Antoine Pitrou added the comment: > Regarding the "marshal", I presume, that somehow the dictionary when > created via "marshal" (or compile, if no ".pyc" is involved?) first > time is somehow less efficient to determine/stream that the one > "cPickle" created. I don't think so. marshal uses a

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Antoine Pitrou added the comment: I think the "premature" optimization comes down to that code: """ static int put(Picklerobject *self, PyObject *ob) { if (Py_REFCNT(ob) < 2 || self->fast) return 0; return put2(self, ob); } """ (put2() being the function which emits BINPUT) W

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Kay Hayen added the comment: Sending my attached file "stream.py" through "2to3.py" it shows that CPython 3.2 doesn't exihibit the issue for either protocol, which may be because it's now "unicode" key, but as it's the only value I tried, I can't tell. Hope this helps. Regarding the "marshal

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Antoine Pitrou added the comment: Ah, right. BINPUT is used to memoize objects in case a restructive structure happens. You can use pickletools.optimize() to remove the useless BINPUTs out of a pickle stream. Note that 3.x is more consistent and always emits the BINPUTs. It seems 2.x's cPick

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Kay Hayen added the comment: It seems that there is an extra "BINPUT 2", whatever it does. I am attaching a variant that does pickletools.dis on the 3 dumps. Protocol 2 : Dumping read const const stream '\x80\x02}q\x01U\x07modulesq\x02Ns.' 0: \x80 PROTO 2 2: }EMPTY_DICT 3:

[issue13721] ssl.wrap_socket on a connected but failed connection succeeds and .peer_certificate gives AttributeError

Antoine Pitrou added the comment: > A return value of None would still not indicate if we had a working > connection without certificate or a failed connection. That would be > annoying. Ah, right. Then raising e.g. a ValueError would be better. -- __

[issue13721] ssl.wrap_socket on a connected but failed connection succeeds and .peer_certificate gives AttributeError

Mads Kiilerich added the comment: > I'm a bit wary of API bloat here. Yes, but explicit is better than magic ... > Thanks. So fixing how getpeercert behaves and either raise a dedicated > error or return None would improve things here, right? Well ... that would at least make it theoretically

[issue10278] add time.wallclock() method

STINNER Victor added the comment: > on linux the underlying functionality is implemented in librt; the extension > doesn't check for this or links with -lrt. The changeset 35e4b7c4bafa changed configure.in to check clock_gettime(). It checks without and with librt: 7.7 +AC_CHECK_FUNCS(c

[issue13739] os.fdlistdir() is not idempotent

Roundup Robot added the comment: New changeset 36f2e236c601 by Charles-François Natali in branch 'default': Issue #13739: It's simpler and more direct to call rewinddir() at the http://hg.python.org/cpython/rev/36f2e236c601 -- ___ Python tracker

[issue13521] Make dict.setdefault() atomic

Antoine Pitrou added the comment: Looking at the patch again, I think this isn't enough. setdefault() will still call the lookup routine twice which, in the general case (i.e. lookdict() not lookdict_unicode()), can call arbitrary Python code through e.g. __eq__ methods. -- _

[issue13721] ssl.wrap_socket on a connected but failed connection succeeds and .peer_certificate gives AttributeError

Antoine Pitrou added the comment: > The only way to reliably implement the documented wrap_socket API > might thus be to maintain a flag in PySocketSockObject. Agreed. With the annoyance that the flag must be exposed to Python code, since ssl's wrap_socket is written in Python. It may be privat

[issue12760] Add create mode to open()

Antoine Pitrou added the comment: > Here's a new version of the patch that should address all the comments. Just a small note: FileExistsError is raised, not exactly OSError, when the file exists. -- ___ Python tracker

[issue13739] os.fdlistdir() is not idempotent

Roundup Robot added the comment: New changeset 7b2a178c028b by Charles-François Natali in branch 'default': Issue #13739: In os.listdir(), rewind the directory stream (so that listdir() http://hg.python.org/cpython/rev/7b2a178c028b -- nosy: +python-dev _

[issue13739] os.fdlistdir() is not idempotent

Ross Lagerwall added the comment: > I see two options: > 1. rewind the directory stream in fdlistdir() > 2. document this > > Here's a patch for option 1. Yeah, looks good. -- ___ Python tracker

[issue12760] Add create mode to open()

Charles-François Natali added the comment: Here's a new version of the patch that should address all the comments. -- Added file: http://bugs.python.org/file24179/open_create_x-3.patch ___ Python tracker _

[issue13721] ssl.wrap_socket on a connected but failed connection succeeds and .peer_certificate gives AttributeError

Mads Kiilerich added the comment: I won't claim to know more about socket error codes than what the Linux man pages says. According to them only send() can fail with ECONNRESET, even though POSIX Programmer's Manual man pages mentions many others. getpeername() is however not documented as be

[issue13739] os.fdlistdir() is not idempotent

Antoine Pitrou added the comment: > I see two options: > 1. rewind the directory stream in fdlistdir() > 2. document this > > Here's a patch for option 1. Agreed with that, and ok with the patch :) -- ___ Python tracker

[issue13521] Make dict.setdefault() atomic

Filip Gruszczyński added the comment: Bump! There was no activity here for two weeks. Is my patch for 2.7 ok or should I do something more about it? -- ___ Python tracker ___ _

[issue13739] os.fdlistdir() is not idempotent

New submission from Charles-François Natali : After a call to fdlistdir(), another call to fdlistdir() on the same file handle (but using a different FD, since the FD passed to fdlistdir() is closed) will return an empty list: """ $ cat ~/test_fdlistdir.py import os import sys fd = os.open(

[issue13026] Dis module - documentation of MAKE_FUNCTION

Sandro Tosi added the comment: Hi Arnaud, would you like to provide a patch to update the MAKE_FUNCTION opcode description? That would speed up a bit the fix-up. Also, I don't have that clear what "For each keyword only default, ..." mean: are those the default values for keyword arguments? s

[issue13673] PyTraceBack_Print() fails if signal received but PyErr_CheckSignals() not called

sbt added the comment: Trivial 3 lines patch. I guess there is still a race: if Ctrl-C is pressed after PyErr_CheckSignals() is called but before PyObject_Str() then the printing of any exception can still be suppressed. -- Added file: http://bugs.python.org/file24177/traceback_check

[issue13738] Optimize bytes.upper() and lower()

Antoine Pitrou added the comment: Thanks. -- resolution: -> fixed stage: patch review -> committed/rejected status: open -> closed ___ Python tracker ___ __

[issue13734] Add a generic directory walker method to avoid symlink attacks

Antoine Pitrou added the comment: > > Also be aware that symlinks mean sometimes you won't have a dirfd: if > > you have a symlink that points to another directory, you can't open that > > directory using openat from the symlink's directory. So if you follow > > symlinks (or have an option to do

[issue13738] Optimize bytes.upper() and lower()

Roundup Robot added the comment: New changeset 9683d59170ee by Antoine Pitrou in branch 'default': Issue #13738: Simplify implementation of bytes.lower() and bytes.upper(). http://hg.python.org/cpython/rev/9683d59170ee -- nosy: +python-dev ___ Python

[issue13735] The protocol > 0 of cPickle does not given stable dictionary values

Antoine Pitrou added the comment: - Can you use pickletools.dis to examine what differs in the pickles' bytecode? - Does it also apply to 3.3? > It appears at least some kind of efficiency might be missed out for > marshall as well. marshal doesn't use the pickle protocols, so it's unlikely to

[issue13734] Add a generic directory walker method to avoid symlink attacks

Charles-François Natali added the comment: Here's a possible walkfd() implementation. Example: """ $ cat /home/cf/testwalkfd.py import os import sys topfd = os.open(sys.argv[1], os.O_RDONLY) for rootfd, dirs, files in os.walkfd(topfd): print(rootfd, dirs, files) $ ./python ~/testwalkfd.py

[issue13738] Optimize bytes.upper() and lower()

Benjamin Peterson added the comment: LGTM. -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue13738] Optimize bytes.upper() and lower()

New submission from Antoine Pitrou : The current implementation has useless testing and copying. -- components: Interpreter Core files: bytesupper.patch keywords: patch messages: 150868 nosy: benjamin.peterson, flox, haypo, pitrou priority: low severity: normal stage: patch review status

[issue12760] Add create mode to open()

Antoine Pitrou added the comment: I don't think the "created()" method has to be exposed. People can inspect the "mode" attribute if they want to have that information. (besides, the semantics are misleading since a new file opened with "w" has also be created, but created() would return False

[issue13703] Hash collision security issue

Antoine Pitrou added the comment: Again, Roundup ate up some of the text: >PCbuild\amd64\python.exe -m timeit -s "import os" "os.urandom(16)" 100 loops, best of 3: 1.81 usec per loop (for the record, the Roundup issue is at http://psf.upfronthosting.co.za/roundup/meta/issue264 ) ---

[issue13703] Hash collision security issue

Antoine Pitrou added the comment: > Randomness is hard to do correctly > and is expensive. If we can avoid it, we should try very hard to do > so... os.urandom() is actually cheaper on Windows 7 here: 100 loops, best of 3: 1.78 usec per loop than on Linux: $ ./python -m timeit -s "import

[issue13734] Add a generic directory walker method to avoid symlink attacks

Antoine Pitrou added the comment: > Since walkdir is currently entirely based on returning filesystem > paths as strings (just like os.walk()) and hence shares the pervasive > symlink attack vulnerability, I'm particularly interested in the > question of whether or not the various *at APIs can b

[issue13733] Change required to sysconfig.py for Python 2.7.2 on OS/2

Antoine Pitrou added the comment: Well, the answer has probably been already given on python-dev: > I've been building Python 2.x for a while, and currently have binaries > of 2.6.5 available from http://os2ports.smedley.info > Unlike Andrew Mcintyre, I'm using libc for development > (http:/

[issue13703] Hash collision security issue

Changes by Terry J. Reedy : -- Removed message: http://bugs.python.org/msg150837 ___ Python tracker ___ ___ Python-bugs-list mailing l

[issue13703] Hash collision security issue

Changes by Terry J. Reedy : -- Removed message: http://bugs.python.org/msg150848 ___ Python tracker ___ ___ Python-bugs-list mailing l

[issue13122] Out of date links in the sidebar of the documentation index of versions 3.1 and 3.2

Sandro Tosi added the comment: Hi all, how can we fix it? or better, should we fix it? From a user POV, it is a weird to see 3.1 doc referring to 3.2 doc as "in development" then clicking on that link, being redirect to 3.3a0 doc and see there that 3.2 is stable (with no reference back to 3.1

[issue10278] add time.wallclock() method

Matthias Klose added the comment: on linux the underlying functionality is implemented in librt; the extension doesn't check for this or links with -lrt. -- nosy: +doko ___ Python tracker

[issue12760] Add create mode to open()

Charles-François Natali added the comment: I intend to commit this patch within a couple days (unless anyone objects of course). -- stage: -> patch review ___ Python tracker _

[issue13703] Hash collision security issue

Marc-Andre Lemburg added the comment: Christian Heimes wrote: > Marc-Andre: > Have you profiled your suggestion? I'm interested in the speed implications. > My gut feeling is that your idea could be slower, since you have added more > instructions to a tight loop, that is execute on every look

[issue13734] Add a generic directory walker method to avoid symlink attacks

Nick Coghlan added the comment: Another, possibly better, alternative would be to produce a tuple-subclass that adds a separate "dirfd" attribute to the (dirpath, subdirs, files) triple. I'll stop talking about the walkdir implications here. Instead, I've created a corresponding issue on walk

[issue13703] Hash collision security issue

Marc-Andre Lemburg added the comment: Tim Peters wrote: > > Tim Peters added the comment: > > [Marc-Andre] >> BTW: I wonder how long it's going to take before >> someone figures out that our merge sort based >> list.sort() is vulnerable as well... its worst- >> case performance is O(n log n),

[issue13703] Hash collision security issue

Paul McMillan added the comment: > Christian Heimes added the comment: > Ouch, the startup impact is large! Have we reached a point where "one size > fits all" doesn't work any longer? It's getting harder to have just one > executable for 500ms scripts and server processes that last for weeks.

[issue13737] bugs.python.org/review's Django settings file DEBUG=True

Changes by Georg Brandl : -- assignee: -> loewis nosy: +loewis priority: normal -> high title: bugs.python.org's Django settings file DEBUG=True -> bugs.python.org/review's Django settings file DEBUG=True ___ Python tracker

[issue13734] Add a generic directory walker method to avoid symlink attacks

Nick Coghlan added the comment: Since walkdir is currently entirely based on returning filesystem paths as strings (just like os.walk()) and hence shares the pervasive symlink attack vulnerability, I'm particularly interested in the question of whether or not the various *at APIs can be used

[issue13733] Change required to sysconfig.py for Python 2.7.2 on OS/2

Martin v. Löwis added the comment: Please consider this: the code that is there was specifically added by somebody porting Python to OS/2. So if it doesn't work for you, either OS/2 has changed (which I don't think is the case), or you are doing something wrong (such as using the wrong tool c

[issue12364] Deadlock in test_concurrent_futures

Ross Lagerwall added the comment: Thanks! -- assignee: -> rosslagerwall resolution: -> fixed stage: -> committed/rejected status: open -> closed type: -> behavior ___ Python tracker ___