Change by Thomas Wouters :
--
nosy: +ned.deily
priority: normal -> release blocker
___
Python tracker
<https://bugs.python.org/issue33497>
___
___
Python-
Thomas Wouters added the comment:
We do use Setup/Setup.local at Google (and have for many years now), and I find
it very useful. (FWIW, the 'we would use' comment in msg294174 was about the
new '*disabled*' feature, not about Setup files in general.) Avoiding local
chang
Thomas Wouters added the comment:
New changeset 500a419a7a68c32650717b213f0f5ab0461bb16b by T. Wouters in branch
'3.6':
[3.6] bpo-32591: fix abort in _PyErr_WarnUnawaitedCoroutine during shutdown
(GH-5337) (#6536)
https://github.com/python/cpyt
New submission from Thomas Wouters :
The fix for issue #31752 (changeset 5ef883b096895a84123760859f0f34ad37bf2277
for 2.7, as I ran into this while upgrading to 2.7.15) includes assertions that
are easily triggered from user code:
>>> import datetime, numpy
>>> datetime
Thomas Wouters added the comment:
This patch includes assertions that are easily triggered from user code:
https://bugs.python.org/issue35021
--
nosy: +twouters
___
Python tracker
<https://bugs.python.org/issue31
Thomas Wouters added the comment:
For the record, Raymond, I think you're wrong about this. Itertools isn't
always a solution to every problem, and it makes for a very awkward way around
a silly limitation in min() and max(). Their API is already awkward -- because
they already take
New submission from Thomas Wouters:
unittest.installHandler incorrectly assumes signal.SIGINT will always be set to
a callable object, rather than signal.SIG_DFL or signal.SIG_IGN. This breaks if
the test is being executed in an environment that e.g. ignores signal.SIGINT,
like a bash script
Thomas Wouters added the comment:
Patch attached.
--
keywords: +patch
stage: -> patch review
Added file: http://bugs.python.org/file26604/inthandler.patch
___
Python tracker
<http://bugs.python.org/issu
New submission from Thomas Wouters:
test_subprocess's test_send_signal assumes SIGINT is not being ignored, by
spawning a new process and expecting it to have the default SIGINT handler .
SIGINT can be set to SIG_IGN if the process running the test inherited it from
the parent pr
Thomas Wouters added the comment:
Patch attached to (briefly) set signal.SIGINT to the default in the test.
It may make sense to add a feature to subprocess.Popen to ignore/unignore
signals (it should only need to support SIG_DFL and SIG_IGN, not other signal
handlers, considering there'
Thomas Wouters added the comment:
Updated patch.
--
Added file: http://bugs.python.org/file26610/inthandler.patch
___
Python tracker
<http://bugs.python.org/issue15
Changes by Thomas Wouters :
Removed file: http://bugs.python.org/file26604/inthandler.patch
___
Python tracker
<http://bugs.python.org/issue15505>
___
___
Python-bug
Thomas Wouters added the comment:
No, using preexec_fn leaves all kinds of races when mixing threads and
processes (which is a bad idea, but people still sometimes do.) And no,
restore_signals doesn't fix this; restore_signals only resets the signals the
*Python interpreter itself* ig
New submission from Thomas Wouters:
In a submodule of a package, replacing the parent package in sys.modules during
import of the package (meaning the only reference to it is in sys.modules) and
then importing itself (or something from itself) crashes the interpreter:
centurion:~/python
Changes by Thomas Wouters :
--
nosy: +twouters
___
Python tracker
<http://bugs.python.org/issue17621>
___
___
Python-bugs-list mailing list
Unsubscribe:
Thomas Wouters added the comment:
The implementation in the patch preserves the exception context. It's probably
the thing that took the most code, and it's why there's two new opcodes in the
patch :) It's covered in the tests, too.
--
__
Changes by Thomas Wouters :
--
stage: needs patch -> patch review
___
Python tracker
<http://bugs.python.org/issue15002>
___
___
Python-bugs-list mai
Thomas Wouters added the comment:
Matthias, I think this is already fixed for Python 3.3 and later (at least.)
There may still be problems in 2.7, but I'm not sure if it's worth fixing them
there. Can you see if you still have problems, and if so, show us how to
reproduce them? (
Thomas Wouters added the comment:
FYI, this broke building in a separate object directory (again!) for multiple
reasons: it's running the script without specifying $(srcdir), and it's writing
to $(srcdir)/Include/opcode.h (where $(srcdir) may be unwritable), and it's
picki
Thomas Wouters added the comment:
Here's a minimal patch to at least make the current mechanism work when using a
separate build directory.
I still don't like the fact that this is importing opcode.py in a different
Python than the target Python. Nor do I like that the script
Thomas Wouters added the comment:
You should put the definition closer to the Windows one (right after or before
it) rather than further down the file. Other than that, looks good.
--
___
Python tracker
<http://bugs.python.org/issue21
Thomas Wouters added the comment:
Looks good.
--
___
Python tracker
<http://bugs.python.org/issue21275>
___
___
Python-bugs-list mailing list
Unsubscribe:
Thomas Wouters added the comment:
Looks good.
--
___
Python tracker
<http://bugs.python.org/issue21276>
___
___
Python-bugs-list mailing list
Unsubscribe:
Thomas Wouters added the comment:
Good fix. Do remove the 'first curses header check' comment you add, and don't
forget to regenerate configure (and maybe pyconfig.h.in? I don't know if
that'll change.)
--
Thomas Wouters added the comment:
I don't understand the story with ffi_convenience here. Perhaps someone else on
python-dev remembers what it was for and whether we need it for any platforms,
still?
--
___
Python tracker
<http://bugs.py
New submission from Thomas Wouters:
The new xattr support in shutil causes shutil.copytree and shutil.copy2 to fail
inelegantly on (source) filesystems that do not support xattrs (like NFS):
# /home/twouters does not support xattrs
>>> os.listxattr("/home/twouters/foo")
Tra
Thomas Wouters added the comment:
ping (you know why :)
--
___
Python tracker
<http://bugs.python.org/issue15505>
___
___
Python-bugs-list mailing list
Unsub
Thomas Wouters added the comment:
Updated patch with test.
--
Added file: http://bugs.python.org/file28926/shutil.patch
___
Python tracker
<http://bugs.python.org/issue17
Changes by Thomas Wouters :
Removed file: http://bugs.python.org/file28898/shutil.patch
___
Python tracker
<http://bugs.python.org/issue17076>
___
___
Python-bugs-list m
Changes by Thomas Wouters :
Removed file: http://bugs.python.org/file28926/shutil.patch
___
Python tracker
<http://bugs.python.org/issue17076>
___
___
Python-bugs-list m
Thomas Wouters added the comment:
Now really updated the patch.
--
Added file: http://bugs.python.org/file28927/shutil.patch
___
Python tracker
<http://bugs.python.org/issue17
New submission from Thomas Wouters:
At some point (probably in 3.2) the support for shared modules built using
Modules/Setup was broken, for two reasons:
- Python no longer considers 'foomodule.so' when looking for a module called
'foo', but Modules/makesetup still appen
Thomas Wouters added the comment:
Additionally, the deprecation notice for 'U' in the table of modules should
mention that the 'U' mode, not universal newlines, is deprecated because it is
unnecessary. Or 'controlled by the newline argument' perhap
New submission from Thomas Wouters:
Here is a preliminary implementation of PEP 463, minus mandatory parentheses
and with the most straightforward precedence rule: equal to if-expr -- which
means this:
A if C else B except E: D
is parsed as
A if C else (B except E: D)
(because of
Thomas Wouters added the comment:
http://hg.python.org/features/pep-463-except-expr now contains a version of the
patch that makes the parentheses mandatory, similar to generator expressions.
(Leaving the old patch for reference.)
--
hgrepos: +218
Changes by Thomas Wouters :
Added file: http://bugs.python.org/file34205/5c078eb8da39.diff
___
Python tracker
<http://bugs.python.org/issue20739>
___
___
Python-bug
Thomas Wouters added the comment:
Here's a preliminary fix (also see
http://codereview.appspot.com/96125/show )
--
assignee: -> twouters
keywords: +patch
nosy: +twouters
Added file: http://bugs.python.org/file14553/import_lock_fork_deadl
New submission from Thomas Wouters :
In order to properly handle multiple threads and fork()-calls from C code
(rather than os.fork()), Python should provide pthread_atfork()-like
functionality: a function to call before a fork, to acquire any locks
that need to be acquired, and a pair of
Thomas Wouters added the comment:
Checked in the patch to fix the forks-through-os.fork() cases, which
should be most of them. Forks from other C code will need some more work,
created http://bugs.python.org/issue6923 to track that.
--
resolution: -> fixed
status: open ->
New submission from Thomas Wouters:
Similar to http://bugs.python.org/issue14509, Python 3.3 conflates Py_DEBUG and
non-NDEBUG builds, creating build failures when building with 'CFLAGS=-UNDEBUG
./configure --without-pydebug'. (assert statements are only compiled out when
NDEBUG i
Thomas Wouters added the comment:
Yes, I already had the same kinds of failures fixed for 3.2, before; this only
affects 3.3 and later.
--
___
Python tracker
<http://bugs.python.org/issue17
Thomas Wouters added the comment:
Unfortunately there is no "release mode". There's Py_DEBUG mode, and the
absence of Py_DEBUG. And there's NDEBUG, and the absence of NDEBUG, which
controls the assert macro. Py_DEBUG unsets NDEBUG, but *not* setting Py_DEBUG
doesn't
New submission from Thomas Wouters:
In response to a question asked at Brett Cannon's Python 3.3 talk at PyCon, it
occurs to me the iterators in the itertools module should participate in
generator sending, so that you can do this:
def report_first_ten(g):
s = itertools.islice(
Thomas Wouters added the comment:
After writing a simplistic implementation of this for itertools.islice, I'm not
sure if this is actually useful. While it's nice for symmetry, I have a hard
time imagining how to use it -- so I can't write tests for the new feature, and
it may
New submission from Thomas Wouters:
The documentation of '%r' in
http://docs.python.org/2/library/stdtypes.html#string-formatting-operations
links to the wrong repr, the module
(http://docs.python.org/2/library/repr.html#module-repr) instead of the builtin
function (http://docs.py
Thomas Wouters added the comment:
I left some style comments in the rietveld review. However, a bigger questions
is whether we should change list at all. I think we should; the change is
fairly straightforward, there is some value in it and it's a good idea to keep
bytearray and other mu
Thomas Wouters added the comment:
Yes. The test doesn't run with more than the requested "units", but if you
run it without -M it does run with a very small number, in order to make
sure the tests don't bitrot.
The mechanism could do with an overhaul though; I wrote it in 2
Thomas Wouters added the comment:
Rewriting the tests shouldn't block this specific issue, no. Also, don't use
multiprocessing for it. I would just use subprocess to start a separate process
(which might after all be OOM-killed,) check the exitcode, and record its
stderr in case
Thomas Wouters added the comment:
The latest patch looks good to me. No need to do the additional AST refactoring
if it's going to make PEP 492's implementor's life harder (but I do read
Guido's comment as a reason to check this in sooner rather than later :>) So,
unl
Thomas Wouters added the comment:
FYI, I meant last Monday, but I forgot it was May 4th (Dutch Memorial day) and
May 5th (Dutch Liberation day), so that got in the way :P
Should we keep this bug open for docs changes, or is there a separate issue for
that
New submission from Thomas Wouters:
The attached patch adds a --testonly mode to the 'perf.py' benchmark runner,
which runs the selected benchmarks in the fastest possible way with just one
interpreter (without filtering out slow benchmarks, or benchmarks that can't
run in
201 - 251 of 251 matches
Mail list logo
