Antoine Pitrou added the comment:
> Is there any reason not to add "-j0" for testall as well?
Have you looked at the patch? :)
> Are these really necessary in a push-race,
> post-local-merge, does Python crash-and-burn case?
Yes, they are.
If they are not significant, the
Antoine Pitrou added the comment:
> You've now merged any changes that have come in since you did your thorough
> tests, and you're trying to beat the other guy to the push. You want
> something that can run *fast* and just proves that the merge didn't hose
> Python
Changes by Antoine Pitrou :
--
nosy: +ezio.melotti, r.david.murray
___
Python tracker
<http://bugs.python.org/issue11653>
___
___
Python-bugs-list mailin
Changes by Antoine Pitrou :
--
nosy: +georg.brandl
___
Python tracker
<http://bugs.python.org/issue11653>
___
___
Python-bugs-list mailing list
Unsubscribe:
Antoine Pitrou added the comment:
I committed the "-j0" part of the patch in d8dd7ab6039d.
Brett made the point on #python-dev that a Makefile change doesn't help Windows
users. Instead, we may have a Python script somewhere that both "make test" and
"make qui
Antoine Pitrou added the comment:
> Sorry, I didn't read an error message very carefully. When I apply your patch
> I see:
>
> >>> from concurrent.futures import *
> >>> from time import *
> >>> t = ThreadPoolExecutor(5)
> >>> t
New submission from Antoine Pitrou :
Not sure if it's the desired behaviour, so I'm reporting it:
$ ./python -c "import atexit; atexit.register(lambda: 1/0)" && echo success
Error in atexit._run_exitfuncs:
ZeroDivisionError: division by zero
[36956 refs]
su
Antoine Pitrou added the comment:
Ok, here is a new patch with an additional test for the atexit hook. If you
don't object, I would like to start committing the test changes, and then the
code changes themselves.
--
stage: needs patch -> patch review
versions: +Python 3.2
Ad
Antoine Pitrou added the comment:
Oops, test didn't work under Windows. Here is a new patch.
--
Added file: http://bugs.python.org/file21361/cfpolling3.patch
___
Python tracker
<http://bugs.python.org/is
Changes by Antoine Pitrou :
Removed file: http://bugs.python.org/file21360/cfpolling3.patch
___
Python tracker
<http://bugs.python.org/issue11635>
___
___
Python-bug
Antoine Pitrou added the comment:
Brett, -W exists too and it seems it failed working here.
Skip, can you please try "make distclean" and then rebuild from scratch?
--
nosy: +pitrou
___
Python tracker
<http://bugs.python.o
Antoine Pitrou added the comment:
> Antoine> Skip, can you please try "make distclean" and then rebuild from
> Antoine> scratch?
>
> I did. The last output was after
Oops, sorry.
> I will try one more time later this evening with a capital "W"
Antoine Pitrou added the comment:
> We can either hack this to work by providing ContextDecorator with a
> way to get the underlying context manager to create a new copy of
> itself each time, or else revert to the 3.1 status quo and declare
> that context managers
Antoine Pitrou added the comment:
Tests now committed, here is a patch without them.
--
Added file: http://bugs.python.org/file21373/cfpolling4.patch
___
Python tracker
<http://bugs.python.org/issue11
Antoine Pitrou added the comment:
> Pretty vanilla. Install in my directory tree, get libraries from MacPorts:
>
> --prefix=/Users/skip/local --enable-shared LDFLAGS=-L/opt/local/lib
> CPPFLAGS=-I/opt/local/include
>
> I thought you had fixed the --enable-shared lin
Antoine Pitrou added the comment:
Which patch should be reviewed? They seem to be different. Senthil's patch
allows a redirect to ftp while Guido's doesn't.
Senthil's patch doesn't seem to fix urllib-inherited code, only urllib2- (see
FancyURLopener.redirect_in
New submission from Antoine Pitrou :
If you do:
./python -c "from concurrent.futures import *; from time import *; t =
ProcessPoolExecutor(1); t.submit(sleep, 60)"
and then kill the child process, the parent process doesn't notice and waits
endlessly for the child to return th
Antoine Pitrou added the comment:
Le jeudi 24 mars 2011 à 16:16 +, STINNER Victor a écrit :
> STINNER Victor added the comment:
>
> In the following example, if I kill the child process, the parent is
> immediatly done:
> ---
> from os import getpid
> from tim
Antoine Pitrou added the comment:
> Antoine Pitrou wrote:
> > It would be nice if it were enabled by default for fatal errors (and
> > asserts perhaps?).
>
> I feel like a broken record. This code hardcodes fd=2 as a write target on
> crash,
For fatal errors, you n
Antoine Pitrou added the comment:
> > Senthil's patch allows a redirect to ftp while Guido's doesn't.
>
> That is a good question. Should we? It doesn't look like ftp:
> participates in the vulnerability, but I'm not sure how useful it is
> either.
Antoine Pitrou added the comment:
After studying the multiprocessing code, it turns out that Queue.get() with a
timeout does its own rather high-frequency polling under Windows (see
Modules/_multiprocessing/pipe_connection.c). Therefore, here is an updated
patch which doesn't have a sec
Antoine Pitrou added the comment:
> The attached patch has both the code to make test skipping more
> obvious as well as eliminating the concept of expected skips.
I still don't like the idea that we have to hand-maintain lists of
"optional" or "required" platf
Antoine Pitrou added the comment:
> I can change it to 'required' and 'optional'.
>
> As for Antoine's comment, do you have another suggestion? I realize it
> isn't necessarily easier per se to manage these lists than the
> 'expected' li
Changes by Antoine Pitrou :
--
nosy: +ezio.melotti
___
Python tracker
<http://bugs.python.org/issue10966>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Antoine Pitrou :
This impacts higher-level APIs such as Queue.get() with a positive timeout.
I'm not sure whether it's easily possible to fix this. A Google search seems to
suggest that the pipe could be opened with "SYNCHRONIZE" access
Antoine Pitrou added the comment:
For the record, the polling code is in conn_poll() in
Modules/_multiprocessing/pipe_connection.c.
Windows named pipes seem to be created in pure Python in
Lib/multiprocessing/connection.py.
--
___
Python tracker
Changes by Antoine Pitrou :
--
nosy: +asksol, jnoller
versions: +Python 2.7, Python 3.1, Python 3.2, Python 3.3
___
Python tracker
<http://bugs.python.org/issue11
Antoine Pitrou added the comment:
> Agreed with nick's idea, the implicitly recreation of the context
> managers would confuse users.
Uh, why would it? That's exactly what I expect the decorator to do, and
I was astonished to discover
Antoine Pitrou added the comment:
> Because there is no *OBVIOUS* code or sign which can illustrate that
> context manager changes from a one-shot to a reusable.
I'm talking about the decorator, not the context manager. Surely there
is a way for the decorator to instantiate a
Antoine Pitrou added the comment:
Well, on my new setup (Windows 7 VM with a Python checkout located on an SMB
drive), the test sometimes passes and sometimes fails, regardless of the patch.
I suspect that maybe maildir requires atomicity guarantees that a network FS
won't pr
Antoine Pitrou added the comment:
Indeed, 61 seems to work. I don't understand the comment about one-second
granularity, shouldn't it be one-minute? (or why do you need 61?)
--
___
Python tracker
<http://bugs.python.
Antoine Pitrou added the comment:
> Hmm. You are right, I wasn't thinking clearly, and I copied that
> mtime setting call from another test. Now I have no idea why 61 would
> work, unless the clock between your virthost and your smb server is
> off by a minute?
No,
Antoine Pitrou added the comment:
before: os.path.getmtime('cur') = 1301075411.6942866
before: os.path.getmtime('new') = 1301075411.693287
after: os.path.getmtime('cur') = 1301075347.38
after: os.path.getmtime('new') = 1301075347.38
self._b
Antoine Pitrou added the comment:
> Hmm. 411-61=350. Three seconds difference looks a little odd. But
> doesn't explain 60 vs 61 making the difference in the test.
>
> Can you change it back to 60 (or even less) and see what the values
> look like when the test fails?
Changes by Antoine Pitrou :
--
nosy: +rhettinger
___
Python tracker
<http://bugs.python.org/issue11674>
___
___
Python-bugs-list mailing list
Unsubscribe:
Antoine Pitrou added the comment:
+1 for predictable behaviour and therefore zeroing of uninitialized arrays.
--
nosy: +pitrou
___
Python tracker
<http://bugs.python.org/issue11
Antoine Pitrou added the comment:
I've posted a review at http://bugs.python.org/review/11393/show.
--
___
Python tracker
<http://bugs.python.org/is
Changes by Antoine Pitrou :
--
nosy: +haypo
___
Python tracker
<http://bugs.python.org/issue11679>
___
___
Python-bugs-list mailing list
Unsubscribe:
Antoine Pitrou added the comment:
Are you sure about MP_EXCEPTION_HAS_BEEN_SET?
semaphore.c has a more sophisticated logic:
Py_BEGIN_ALLOW_THREADS
ResetEvent(sigint_event);
res = WaitForMultipleObjects(2, handles, FALSE, msecs);
Py_END_ALLOW_THREADS
Antoine Pitrou added the comment:
I've now pushed the patch. I hope this won't break anything, closing.
--
resolution: -> fixed
stage: patch review -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs
Changes by Antoine Pitrou :
--
nosy: -pitrou
___
Python tracker
<http://bugs.python.org/issue11236>
___
___
Python-bugs-list mailing list
Unsubscribe:
Antoine Pitrou added the comment:
> I updated the (Hg repo and the) patch to fix all Antoine's remarks.
Can you make the suggested changes to the tests? Thank you.
--
___
Python tracker
<http://bugs.python.org
Antoine Pitrou added the comment:
I have pushed a new branch "faulthandler-thread" in
http://hg.python.org/features/faulthandler/. It contains an implementation of
dump_tracebacks_later() using a watchdog thread, instead of alarm().
It has two advantages:
- it works under Windows
Antoine Pitrou added the comment:
One strong reason for having the test files in the build directory is ease of
cleanup, especially on the buildbots where crashes or hangs can lead to
progressive disk fillup (and some tests create very large files, e.g. 2GB).
See also 673a5afce4e0
Antoine Pitrou added the comment:
> As a more general policy question... where do we stand in regards to
> backwards compatibility of the AST? The ast module docs don't have any
> caveats to say that it may change between versions, but it obviously
> *can* change due to new la
Antoine Pitrou added the comment:
> I think these demo functions should be either be removed or
> incorporated into the docs.
+1 for either of that.
--
nosy: +pitrou
___
Python tracker
<http://bugs.python.org/i
Antoine Pitrou added the comment:
FWIW, Mercurial uses the following dance:
http://selenic.com/repo/hg/file/463aca32a937/mercurial/windows.py#l296
(Mercurial is under the GPL, so we can't copy that code verbatim; but it can
serve as an inspiration)
--
nosy: +p
Antoine Pitrou added the comment:
Thanks for the patch.
A couple of comments:
- this is a new feature, so can only go in in 3.x: no need to post a 2.7 patch
(unless this helps Gerhard for his standalone project)
- you need to document the new API in Doc/library/sqlite3.rst
About the patch
Antoine Pitrou added the comment:
> I did start to undertake a conversion of TESTFN to a named temporary,
> but it started to sprawl all over the place and came up against a
> number of corner cases (eg where tests deliberately wanted two
> filenames to be the same) so I gave up
Antoine Pitrou added the comment:
> Makes sense. So, what do you think about adding a --usetmp/-p flag to
> regrtest to honor mkdtemp's defaults even in a build dir? I'd add an
> atexit handler to clean it up but of course if it crashes and you've
> used the flag, y
Changes by Antoine Pitrou :
--
nosy: +loewis
___
Python tracker
<http://bugs.python.org/issue11696>
___
___
Python-bugs-list mailing list
Unsubscribe:
Antoine Pitrou added the comment:
Thanks for checking. Closing as fixed.
--
resolution: -> fixed
status: open -> closed
___
Python tracker
<http://bugs.python.org/
Antoine Pitrou added the comment:
I don't have a 2.5 checkout to test but the patch looks ok to me.
Under 2.7 I get a test failure, I suppose you'll have some merging work to do:
test test_urllib2 failed -- Traceback (most recent call last):
File "/home/antoine/cpyt
Antoine Pitrou added the comment:
Barry: does it allow to install Python into /usr/lib/whateverarch, or is it
just a partial fix for something slightly unrelated to this issue?
--
nosy: +pitrou
___
Python tracker
<http://bugs.python.
Antoine Pitrou added the comment:
It would be simpler to use "with" indeed.
Do you want to provide a patch?
--
nosy: +pitrou
versions: +Python 3.3 -Python 2.6, Python 3.1
___
Python tracker
<http://bugs.python.o
Antoine Pitrou added the comment:
> - speed up the test: because dump_backtraces_later() has now a
> subsecond resolution, we can use sleep of 50 ms instead of 1 sec
This is too short, there may be random failures on some slow buildbots.
IMO, 0.5s is the minimum you can use.
+
Antoine Pitrou added the comment:
> In short: If you have a socket with settimeout(1), then accept a
> connection on it, the new socket will have gettimeout()==None, but its
> state will still (internally) be non-blocking. The attached script
> demonstrates the issue.
This shou
Antoine Pitrou added the comment:
> The corresponding defect has a long (and bothersome) discussion. I am,
> however, surprised that this was not considered a "bug" and backported.
I think Martin's argument was that it could break compatibility.
> I can agree w
Antoine Pitrou added the comment:
> - speed up the test: because dump_backtraces_later() has now a
> subsecond resolution, we can use sleep of 50 ms instead of 1 sec
This is too short, there may be random failures on some slow buildbots.
IMO, 0.5s is the minimum you can use.
+
Antoine Pitrou added the comment:
> * Patch Py_DeleteFileW in posixmodule.c so that it renames before
> deleting: should solve the problem overall but obviously has a
> possible wider impact, in general and on performance in particular.
> This rename might be a simple renam
Antoine Pitrou added the comment:
> I also don't think there is a portable way to detect the NBIO attribute
> of a socket, so we still have a case of socket.gettimeout() not
> accurately reflecting the blocking state of the socket
Which case?
> I personally think that thi
Antoine Pitrou added the comment:
> socket.defaulttimeout(None)
> s = socket.socket()
> s.settimeout(0) #nonblocking
> s.bind()
> s2, a = s.accept()
> print s2.gettimeout() #prints ´none´, meaning blocking
> s2.receive(10) #raises EWOULDBLOCK error, since internally it i
Antoine Pitrou added the comment:
> Antoine, absolument. Please see attached file bug.py
Ah, thanks, I see. But this is not caused by this issue.
It seems to come from 12442ac3f7dd.
> Instead we should simply define it for python, and in accordance to
> established tradition, na
Antoine Pitrou added the comment:
What version of Python did you make your patch against? It failed applying
against the default branch.
You might want to make sure you are using an up-to-date source tree, see:
http://docs.python.org/devguide/setup.html#getting-the-source-code
Antoine Pitrou added the comment:
The patch looks good to me, thank you!
Gerhard, would you like to tackle this? Otherwise I'll commit in a couple of
days.
--
stage: patch review -> commit review
___
Python tracker
<http://bugs
Antoine Pitrou added the comment:
Nadeem,
> Can I ask that you not commit this
> patch until the CA has been submitted? I will need to clear it with my
> employer, and it might complicate things if the code in question has
> already been committed.
Apparently the PSF has r
Antoine Pitrou added the comment:
> OTOH, this is mostly a debugging aid and nothing that would be enabled
> in production. And premature optimization is the root of all evil...
Agreed.
--
___
Python tracker
<http://bugs.python.org/i
Antoine Pitrou added the comment:
I have now committed the semaphore implementation, so as to fix the issue.
Potential performance optimizations can still be discussed, of course (either
here or in a new issue, I'm not sure).
--
resolution: -> fixed
stage: -> committ
Antoine Pitrou added the comment:
There is a failure on FreeBSD:
==
FAIL: test_dump_tracebacks_later_repeat
(test.test_faulthandler.FaultHandlerTests
Antoine Pitrou added the comment:
Just a nit but:
"if timeout and 0 < timeout"
Could you use the usual coding style? ("if timeout and timeout > 0")
Putting constants on the left-side of comparisons is pointless in Python, and
even in C it's quite ugly a
Antoine Pitrou added the comment:
> "x86 Tiger custom" to learn more about test_threadsignals timeout
> (1 hour)
Please read the comments. There is no timeout anymore.
--
___
Python tracker
<http://bugs.py
Antoine Pitrou added the comment:
Agreed with Brian. There is generally no reliable way to terminate another
thread or process without cooperation from said thread or process. Especially
in the case of threads, terminating a thread while leaving the process alive
may leave some resources
Antoine Pitrou added the comment:
The problem with this approach is that it won't help concurrent.futures.
Detection of killed endpoints should ideally happen at a lower level, e.g. in
Process or Queue or Connection objects.
Speaking of which, I wonder why we have both multiprocessing
Antoine Pitrou added the comment:
Possible plan for POSIX, where a connection uses a pipe() or socketpair():
exploit the fact that an endpoint becomes ready for reading (indicating EOF)
when the other endpoint is closed:
>>> r, w = os.pipe()
>>> select.sele
Antoine Pitrou added the comment:
(certainly not easy, sorry)
--
___
Python tracker
<http://bugs.python.org/issue9205>
___
___
Python-bugs-list mailing list
Unsub
Antoine Pitrou added the comment:
> New changeset 8b1341d51fe6 by Victor Stinner in branch 'default':
> Issue #11393: Fix faulthandler_thread(): release cancel lock before join lock
> http://hg.python.org/cpython/rev/8b1341d51fe6
This is wrong, it should always be rele
Antoine Pitrou added the comment:
> I think it's best to remove all this inconsistency and fix it so that
> EPIPE is never generated and then backport it to 2.7, 3.1, 3.2.
>
> Attached is a patch which fixes it for poll, select, windows and adds
> two tests.
The pa
Antoine Pitrou added the comment:
> Here is an updated patch that adds read1() to BZ2File. This should fix things
> for issue10791 from the bz2 side. I also took the opportunity to clean up
> _read_block() to be more readable. As per Martin's suggestion on python-dev, I
>
New submission from Antoine Pitrou :
Here is a rewrite of multiprocessing.{PipeConnection,Connection} in pure
Python, for ease of maintenance and improvement.
--
components: Library (Lib)
files: mpconn.patch
keywords: patch
messages: 132816
nosy: asksol, brian.curtin, jnoller, pitrou
Changes by Antoine Pitrou :
--
dependencies: +Rewrite PipeConnection and Connection in pure Python
___
Python tracker
<http://bugs.python.org/issue9205>
___
___
Antoine Pitrou added the comment:
> The tests for readline() and readlines() expect a TypeError if size is None.
> Calling size.__index__() in this case raises an AttributeError instead.
> Should I
> change the tests to expect an AttributeError? Alternatively, something like
&g
Antoine Pitrou added the comment:
Thank you very much, Nadeem. The patch is now in.
--
resolution: -> fixed
stage: patch review -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs.python.o
Antoine Pitrou added the comment:
The patch here is totally out of date, following issue5863.
--
nosy: +nvawda
stage: patch review -> needs patch
versions: +Python 3.3 -Python 3.2
___
Python tracker
<http://bugs.python.org/iss
Antoine Pitrou added the comment:
The patch slows down Pipe() a bit, by the way. On my machine, each message sent
and received has an additional 10µs overhead. See attached benchmark script.
--
Added file: http://bugs.python.org/file21515/mpconn.patch
Antoine Pitrou added the comment:
The patch slows down Pipe() a bit, by the way. On my machine, each message sent
and received has an additional 10µs overhead. See attached benchmark script.
--
Added file: http://bugs.python.org/file21516/pipebench.py
Antoine Pitrou added the comment:
Woops, sorry for the duplicates...
--
___
Python tracker
<http://bugs.python.org/issue11743>
___
___
Python-bugs-list mailin
Changes by Antoine Pitrou :
Removed file: http://bugs.python.org/file21515/mpconn.patch
___
Python tracker
<http://bugs.python.org/issue11743>
___
___
Python-bugs-list m
Changes by Antoine Pitrou :
--
Removed message: http://bugs.python.org/msg132847
___
Python tracker
<http://bugs.python.org/issue11743>
___
___
Python-bugs-list m
New submission from Antoine Pitrou :
See
http://www.python.org/dev/buildbot/all/builders/x86%20XP-5%203.x/builds/2475/steps/test/logs/stdio
test test_ftplib failed -- Traceback (most recent call last):
File "D:\Buildslave\3.x.moore-windows\build\lib\test\test_ftplib.py", lin
New submission from Antoine Pitrou :
http://www.python.org/dev/buildbot/all/builders/x86%20XP-5%203.x/builds/2473/steps/test/logs/stdio
==
FAIL: testSmallReadNonBlocking
(test.test_socket.UnbufferedFileObjectClassTestCase
Antoine Pitrou added the comment:
10048 is errno.EADDRINUSE.
--
___
Python tracker
<http://bugs.python.org/issue11748>
___
___
Python-bugs-list mailing list
Unsub
Antoine Pitrou added the comment:
Committed, thank you.
--
resolution: -> fixed
stage: -> committed/rejected
status: open -> closed
versions: +Python 3.3
___
Python tracker
<http://bugs.python.or
Antoine Pitrou added the comment:
http://hg.python.org/cpython/rev/8a2d848244a2
--
___
Python tracker
<http://bugs.python.org/issue11748>
___
___
Python-bug
New submission from Antoine Pitrou :
subprocess and multiprocessing both have their own private modules for wrappers
of win32 functions: Modules/_multiprocessing/win32_functions.c and
PC/_subprocess.c.
It would be nice to group them in a common module (_win32?) that could be used
throughout
Antoine Pitrou added the comment:
Patch committed, thank you!
--
assignee: ghaering ->
resolution: -> fixed
stage: commit review -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs.python
New submission from Antoine Pitrou :
In
http://www.python.org/dev/buildbot/all/builders/AMD64%20OpenIndiana%203.x/builds/904/steps/test/logs/stdio
:
test test_subprocess failed -- Traceback (most recent call last):
File
"/export/home/buildbot/64bits/3.x.cea-indiana-amd64/build/Lib
Antoine Pitrou added the comment:
According to POSIX, EINVAL in select() means either "an invalid timeout
interval was specified" or "the nfds argument is less than 0 or greater than
FD_SETSIZE".
(http://www.opengroup.org/onlinepubs/007904875/functions/pselect.html)
Ad
Antoine Pitrou added the comment:
> Nadeem Vawda added the comment:
>
> > Is following change in GzipFile class enough:
> >
> > def read1(self, n):
> > return self.read(n)
> >
> > ? This satisfies TextIOWrapper to run readline correc
Antoine Pitrou added the comment:
Reopening, a failure appeared on some of the buildbots. I've made the failure
message a bit more explicit:
test_sqlite: testing with version '2.6.0', sqlite_version '3.6.12'
[...]
==
Antoine Pitrou added the comment:
Ok, it seems that expanding the value of bound parameters in the statement
passed to the trace callback is a recent SQLite feature:
http://www.sqlite.org/draft/releaselog/3_6_21.html
“The SQL output resulting from sqlite3_trace() is now modified to
2601 - 2700 of 16792 matches
Mail list logo
