New submission from Tim Koopman:
When using rmtree with a symlink to a directory as path, it will first
follow the symlink, (try to) remove all the contents of the source
directory and then raise the exception OSError: [Errno 20] Not a
directory.
Expected behaviour: The function should only
Changes by Tim Koopman:
--
title: shutils.rmtree fails on symlink, after deleting contents -
shutil.rmtree fails on symlink, after deleting contents
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1669
__
New submission from Ronald Oussoren:
The scriptlet below hangs (as expected) but is also not interuptable by
Ctrl+C, at least on Linux and Mac OS X:
from Queue import Queue, Empty, Full
q = Queue()
q.get(True)
This is due to Threading.Condition.wait not being interuptable.
--
Thomas Herve added the comment:
Thanks a lot!
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1549
__
___
Python-bugs-list mailing list
Unsubscribe:
Facundo Batista added the comment:
This is not a bug, just a comment of a code you did.
If you think that there's actually a problem in HTMLParser.py, tell us
what you do, what you get, and what you expect to get.
When submitting changes, it's very useful to send the diff only, as in a
patch.
Changes by Christian Heimes:
Removed file: http://bugs.python.org/file8999/trunk_select_epoll3.patch
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1657
__
___
Python-bugs-list
Changes by Christian Heimes:
Removed file: http://bugs.python.org/file9009/trunk_select_epoll_kqueue.patch
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1657
__
___
Changes by Christian Heimes:
Added file: http://bugs.python.org/file9010/trunk_select_epoll_kqueue.patch
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1657
__
___
Ismail Donmez added the comment:
Hi Martin,
Actually the only problem is how can I get wctype functionality with
8-bit strings, any example is appreciated.
This bug itself is invalid because --with-wctype-functions is
deprecated. But as I said I just hope removing that doesn't regress
Turkish
Christian Heimes added the comment:
The GHOP task was a success. Georg will merge the new docs soonish.
--
resolution: - fixed
status: open - closed
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1570
__
Christian Heimes added the comment:
Georg, do we need more docs?
--
assignee: tiran - georg.brandl
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1497
__
___
Christian Heimes added the comment:
The PCbuild9 directory works for 2.6 and 3.0.
--
resolution: - out of date
status: open - closed
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1465
__
Christian Heimes added the comment:
I've removed the disabled test from py3k. There is no chance it is going
to work on all systems.
--
status: open - closed
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1377
__
New submission from Quentin Gallet-Gilles:
The title says it all.
I also updated the country codes and other TLDs as things have evolved a
bit since 2002.
--
components: Demos and Tools
files: world_tool.patch
messages: 58874
nosy: quentin.gallet-gilles
severity: normal
status: open
Changes by Georg Brandl:
--
status: pending - closed
versions: +3rd party -Python 2.5
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1665
__
___
Python-bugs-list
New submission from Joseph Armbruster:
Subprocess contains a minor temporary file getter that failed in one of
my terminals due to the method used to obtain a temporary directory
location.
Patch attached.
--
components: Tests
files: subprocess.patch
messages: 58875
nosy:
New submission from Joseph Armbruster:
test_pep277 failed on one of my windows boxes. It looks as if an import
unicode is missing.
See attached patch.
--
components: Tests
files: testpep277.patch
messages: 58876
nosy: JosephArmbruster
severity: normal
status: open
title: test_pep277
Joseph Armbruster added the comment:
had to rush this one as i'm at work right now and did not verify it.
disregard that patch!
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1673
__
Christian Heimes added the comment:
Fixed in r59580
Please test it, I don't have access to a Windows box right now.
--
nosy: +tiran
resolution: - fixed
status: open - pending
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1673
Joseph Armbruster added the comment:
I used str( instead of unicode(
python -E -tt ../lib/test/regrtest.py test_pep277
test_pep277
1 test OK.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1673
__
Christian Heimes added the comment:
Thx :)
--
keywords: +patch, py3k
nosy: +tiran
priority: - normal
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1671
__
___
Changes by Joseph Armbruster:
--
title: test_pep277 missing unicode import - test_pep277 fails
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1673
__
___
Guido van Rossum added the comment:
Tickle the interest of one of the many folks with commit privileges.
Since 2.6 isn't going to be released for months there's no great hurry,
but waiting until the last minute would be a mistake.
__
Tracker [EMAIL PROTECTED]
Guido van Rossum added the comment:
If there aren't too many differences between the 2.6 and 3.0 version of
decimal.py and your patch, do 2.6 first, then the next time we merge
stuff into 3.0 from the trunk it'll be forward-ported automatically.
Though you'd have to start by backporting
Ismail Donmez added the comment:
Funnily,
print .encode(iso-8859-9).decode(iso-8859-9).upper()
works, but
print .encode(iso-8859-9).upper().decode(iso-8859-9)
not.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1609
Guido van Rossum added the comment:
Funnily,
print .encode(iso-8859-9).decode(iso-8859-9).upper()
works, but
print .encode(iso-8859-9).upper().decode(iso-8859-9)
not.
You'll have to debug this yourself.
__
Tracker [EMAIL PROTECTED]
Guido van Rossum added the comment:
Maybe the original author (Barry Warsaw) can review this?
--
assignee: - barry
nosy: +barry, gvanrossum
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1671
__
Ismail Donmez added the comment:
I guess so, I will no longer spam this bug. Thanks for the suggestions.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1609
__
___
Guido van Rossum added the comment:
Two easy ways to get the functionality using 8-bit strings, assuming
you've already set your locale properly:
(1) If your data is already an 8-bit string (i.e. isinstance(data,
str)), simply use data.upper() or data.lower()
(2) If your data is Unicode (i.e.
Joseph Armbruster added the comment:
and.. after performing an svn update :-)
rt test_pep277
Deleting .pyc/.pyo files ...
36 .pyc deleted, 0 .pyo deleted
python -E -tt ../lib/test/regrtest.py test_pep277
test_pep277
1 test OK.
About to run again without deleting .pyc/.pyo first:
Press any key
Guido van Rossum added the comment:
This can't be fixed directly -- the pthreads mutex is not an
interruptable system call.
However there's a simple (though expensive) work-around: use a very long
timeout. The timeout version of waiting for a lock is a busy-wait with a
short sleep (much less
Quentin Gallet-Gilles added the comment:
You're welcome :-)
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1671
__
___
Python-bugs-list mailing list
Unsubscribe:
Joseph Armbruster added the comment:
Here's some other quick tests for posterity:
TEST 1 [from a downloaded msi from a while ago]
Python 2.5.1 (r251:54863, Apr 18 2007, 08:51:08) [MSC v.1310 32 bit
(Intel)] on win32
Type help, copyright, credits or license for more information.
import os
p =
Jeffrey Yasskin added the comment:
Right. Will do.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1623
__
___
Python-bugs-list mailing list
Unsubscribe:
Barry A. Warsaw added the comment:
The change looks good to me, thanks! I made a minor change to your
patch (i.e. .keys() not needed in sorted()), did some whitespace
normalization, and fixed up the --dump output. Otherwise, the patch is
quentin's and I've committed it to the tree.
r59581
New submission from Eric Moyer:
I installed the python-2.5.1.msi stable package from the python.org
website using a super-user account on my system. I installed for all
users and used the default location and chose to install everything,
not omitting any component.
After installation, I moved
Raghuram Devarakonda added the comment:
Please try to include stack trace in bug reports. I reproduced the error
on my Linux (SuSE 10).
marvin:tmp$ ls -l dirlink testdir
lrwxrwxrwx 1 raghu users7 2007-12-20 10:10 dirlink - testdir/
testdir:
total 0
-rw-r--r-- 1 raghu users 0 2007-12-20
Quentin Gallet-Gilles added the comment:
That was quick, thanks!
I'll make sure to remember that sorted() and keys() aren't needed together.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1671
__
Guido van Rossum added the comment:
I think you misunderstand raw_input(). It just returns
sys.stdin.readline().rstrip(\n). I don't think it's worth fixing.
The 'q' processing works fine as long as you didn't type \r first. :-)
--
nosy: +gvanrossum
resolution: - wont fix
status:
Joseph Armbruster added the comment:
Here are two examples of why I created this issue:
example 1: that is the output I received when pressing Return 5 times
using license()
python
Python 3.0a2 (py3k:59579M, Dec 20 2007, 08:46:46) [MSC v.1500 32 bit
(Intel)] on win32
Type help, copyright,
Eric Moyer added the comment:
I tried adding my normal account to the administrators group and that
did not fix the problem.
And my system:
I am running fully patched Windows XP professional SP 2 on a Presario
V2000 (AMD Turion 64 ML-37 with 480 MB ram (the other 32 MB RAM is used
by the video
Guido van Rossum added the comment:
I believe its original intent was to only ignore variables starting with
PYTHON, but I'm not 100% sure. Ignoring PATH seems counterintuitive.
I'm not sure about THREADDEBUG -- I'm thinking that variable should be
renamed PYTHONTHREADDEBUG?
BTW, doesn't this
Guido van Rossum added the comment:
I agree with Tesiph, more useful behavior would be to raise an error
immediately because the argument is not a directory. If you wanted to
remove the think linked to, you could use rmtree(foo/.,
ignore_errors=True).
--
nosy: +gvanrossum
Raghuram Devarakonda added the comment:
I am ok with disallowing symlinks in rmtree() because if it were to be
allowed, the semantics are not clear. In addition, neither 'rmdir' nor
'rm -rf' delete the target directory.
The attached patch would raise error if symbolic link is passed to
rmtree.
Eric Moyer added the comment:
I tried rolling back the computer and installing it on my normal account
with added super-user privileges and though the install succeeded, I
still couldn't run it. However, the normal super-user could run it just
fine.
Looks like this is some weird
Guido van Rossum added the comment:
Fixed in 2.6.
Committed revision 59582.
3.0 will follow at the next merge.
--
keywords: +patch
nosy: +gvanrossum
resolution: - accepted
status: open - closed
versions: +Python 2.6 -Python 3.0
__
Tracker [EMAIL
Guido van Rossum added the comment:
Very odd. Can you experiment with input() and sys.stdin.readline() to
see what exactly gets returned? It seems that somehow the CRLF - LF
translation isn't working, and that should be fixed somewhere else,
not in license().
I imagine you could just do this:
Guido van Rossum added the comment:
Thanks for the patch. I think it should raise IOError, not ValueError,
and it should use the onerror() handling used for all other errors.
Also, can you include an update for the docs in the Doc tree?
__
Tracker [EMAIL
Joseph Armbruster added the comment:
Here's the results, i'll look into it briefly.
import sys
x = input()
Hi:
len(x)
3
x = sys.stdin.readline()
Hi:
len(x)
4
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1667
__
Guido van Rossum added the comment:
len(x)
Please print repr(x). That should be safe even if it contains control
characters.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1667
__
Joseph Armbruster added the comment:
import sys
x = input()
Hi:
repr(x)
'Hi:'
x = sys.stdin.readline()
Hi:
repr(x)
'Hi:\\n'
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1667
__
Guido van Rossum added the comment:
import sys
x = input()
Hi:
repr(x)
'Hi:'
x = sys.stdin.readline()
Hi:
repr(x)
'Hi:\\n'
Hm, that works as expected (except I meant Hi: to be the prompt, but
that's okay :-). Why doesn't it inside license()? Can you put a print
repr(key) in the
Joseph Armbruster added the comment:
Ok, just did as so:
sys.stdout.write(prompt)
sys.stdout.flush()
key = sys.stdin.readline()
print(repr(key))
print(len(key))
if key not in ('', 'q'):
Results:
All Python releases are Open Source (see http://www.opensource.org for
Hit Return for more, or q
Thomas Herve added the comment:
Some remarks:
* the name of the function used for PyArg_ParseTupleAndKeywords in
register, modify, unregister is set to control instead of the good name.
* there is a leak in pyepoll_new if the parsing of arguments fails.
* the indentation is sometimes tabs,
thekorn added the comment:
Thanks a lot, this works.
Markus
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1665
__
___
Python-bugs-list mailing list
Unsubscribe:
Guido van Rossum added the comment:
Ah d'oh!
The 3.0 code was wrong. I've fixed it now in SVN by going back to input()
Committed revision 59583.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1667
__
Joseph Armbruster added the comment:
Looks good :-)
Python 3.0a2 (py3k:59579M, Dec 20 2007, 08:46:46) [MSC v.1500 32 bit
(Intel)] on win32
Type help, copyright, credits or license for more information.
license()
A. HISTORY OF THE SOFTWARE
==
Python was created in the
Raghuram Devarakonda added the comment:
Index: Lib/shutil.py
===
--- Lib/shutil.py (revision 59581)
+++ Lib/shutil.py (working copy)
@@ -156,6 +156,16 @@
elif onerror is None:
def onerror(*args):
New submission from Isaac Morland:
There appears to be a race condition in os.makedirs. Suppose two
processes simultaneously try to create two different directories with a
common non-existent ancestor. For example process 1 tries to create
a/b and process 2 tries to create a/c. They both
Raghuram Devarakonda added the comment:
I don't think os.makedirs() can do anything here. It should be caller's
responsibility to check for this kind of issues.
--
nosy: +draghuram
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1675
Christian Heimes added the comment:
Some remarks:
* the name of the function used for PyArg_ParseTupleAndKeywords in
register, modify, unregister is set to control instead of the good name.
Fixed
* there is a leak in pyepoll_new if the parsing of arguments fails.
Fixed
* the
Isaac Morland added the comment:
The only thing I found in the bug database concerning os.makedirs was
Issue 766910 (http://bugs.python.org/issue766910). I realized
os.makedirs had a race condition because in my application I want to
create directories but it's perfectly fine if they already
Isaac Morland added the comment:
I would suggest that the need to create directories that may already
exist (really ensure existence of directories) is not exclusive to
trace.py. I am suggesting this be added as an option to os.mkdir and
os.makedirs. See Issue 1675.
--
nosy: +ijmorlan
Guido van Rossum added the comment:
I think we can fix this as follows: whenever it calls os.mkdir() and an
error is returned, check if that is EISDIR or EEXISTS, and if so, check
that indeed it now exists as a directory, and then ignore the error.
Moreover, I'd like to do this for the ultimate
Changes by Isaac Morland:
Added file: http://bugs.python.org/file9016/makedirs.py
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1675
__
___
Python-bugs-list mailing list
Guido van Rossum added the comment:
Can you rephrase this as svn diff output?
Also, mkdir() is a confusing name for the helper -- I'd call it
forgiving_mkdir() or something like that.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1675
Isaac Morland added the comment:
Yes, I'm really combining two things here - the race condition, which I
argue is a (minor) bug, and a feature request to be able to ensure
exists a directory.
I have not produced a proper Python patch before and I have other things
to do so this will take longer
Martin v. Löwis added the comment:
print .encode(iso-8859-9).upper().decode(iso-8859-9)
does not
Please get your types right. is a byte string (in Python 2.x).
encode: unicode - string
decode: string - unicode
That you still can apply .encode to the byte string is a bug/pit fall in
Ismail Donmez added the comment:
Tried like ,
unicode(iii).encode(iso-8859-9).upper()
doesn't work, I'll ask on python users list. Thanks.
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1609
__
Martin v. Löwis added the comment:
As of r59585, _PyLong_FitsInLong() is no longer.
--
resolution: - fixed
status: open - closed
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1666
__
New submission from Kevin Walzer:
I've just updated to Tk 8.5 on OS X (Leopard, 10.5.1) and have rebuilt
Python to link to the new version of Tk. I'm seeing tons of weird error
messages in my logs when I run IDLE:
12/20/07 8:18:46 PM [0x0-0xa50a5].org.python.IDLE[1300] Break on
New submission from Isaul Vargas:
When running Python 2.5.1 stable in Windows, you can press Ctrl-C as
many times as you want and it will always output Keyboard Interrupt in
the interpreter.
Python 3.0a+ will quit if you press ctrl-c too many times. The last
release of 3.0a2 can handle many
Changes by Martin v. Löwis:
--
status: pending - closed
__
Tracker [EMAIL PROTECTED]
http://bugs.python.org/issue1673
__
___
Python-bugs-list mailing list
Unsubscribe:
Martin v. Löwis added the comment:
I would expect that this is a Tk bug, primarily, not a Python bug. If
somebody could reproduce it with Tk alone, it would be good to report it
at http://sourceforge.net/tracker/?group_id=12997
--
nosy: +loewis
__
74 matches
Mail list logo
