[issue36760] subprocess.run fails with capture_output=True and stderr=STDOUT
New submission from Joe Borg : Reading from https://docs.python.org/3/library/subprocess.html#subprocess.CompletedProcess """ If you ran the process with stderr=subprocess.STDOUT, stdout and stderr will be combined in this attribute, and stderr will be None. """ But, if you run `run()` with `capture_output=True`, you get the following exception: """ ValueError: stdout and stderr arguments may not be used with capture_output. """ So, it seems impossible to get the combined outputs of stdout and stderr with `run()`. -- components: Library (Lib) messages: 341158 nosy: Joe.Borg priority: normal severity: normal status: open title: subprocess.run fails with capture_output=True and stderr=STDOUT type: behavior versions: Python 3.7 ___ Python tracker <https://bugs.python.org/issue36760> ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue4705] python3.0 -u: unbuffered stdout
Joe Borg added the comment: Can I confirm this is still in the trunk? I have 3.3.2 and am suffering from the fact that `-u` isn't setting stdin to unbuffered. I'm have to run a flush every command, which is awful. -- nosy: +Joe.Borg, georg.brandl versions: +Python 3.2, Python 3.3 ___ Python tracker <http://bugs.python.org/issue4705> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue18868] Python3 unbuffered stdin
New submission from Joe Borg:
I'm in need of an unbuffered stdin for Python3. Using the '-u' flag worked
fine in Python2. But, it seems, Python3's stdin is always buffered; as seen in
http://bugs.python.org/issue4705.
This is not always desirable. For example:
#!/bin/python3
import os, subprocess, time
with open("%s/unbuffered_test.log" % (os.getenv("HOME")), "w") as f:
with subprocess.Popen(["%s/unbuffered_test.sh" % (os.getenv("HOME"))],
stdin=subprocess.PIPE, stdout=f, stderr=f) as p:
p.stdin.write(bytes("test\n", encoding="utf-8"))
time.sleep(10)
Where unbuffered_test.sh is:
#!/bin/sh
read INPUT
echo $INPUT
exit 0
Running with -u in Python2 sees the log file populated before the 10 seconds
are up. This isn't the case in Python3. This making controlling applications,
using subprocess, basically impossible; without putting p.stdin.flush() after
each command (which does work in the example above).
I ran this example in Python3.3.2.
--
messages: 196359
nosy: Joe.Borg
priority: normal
severity: normal
status: open
title: Python3 unbuffered stdin
versions: Python 3.1, Python 3.2, Python 3.3, Python 3.4, Python 3.5
___
Python tracker
<http://bugs.python.org/issue18868>
___
___
Python-bugs-list mailing list
Unsubscribe:
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue18868] Python3 unbuffered stdin
Changes by Joe Borg : -- type: -> behavior ___ Python tracker <http://bugs.python.org/issue18868> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue21481] Argpase Namespace object methods __eq__ and __ne__ raise TypeError when comparing to None
New submission from Joe Borg:
See example:
>>> import argparse
>>> a = argparse.ArgumentParser()
>>> b = a.parse_args([])
>>> if b != None:
... print "hey"
File "", line 2
print "hey"
^
SyntaxError: invalid syntax
>>>
>>> if b != None:
... print("hey")
...
Traceback (most recent call last):
File "", line 1, in
File "/cfd/software/Python/340/lib/python3.4/argparse.py", line 1202, in
__ne__
return not (self == other)
File "/cfd/software/Python/340/lib/python3.4/argparse.py", line 1199, in
__eq__
return vars(self) == vars(other)
TypeError: vars() argument must have __dict__ attribute
--
components: Library (Lib)
messages: 218326
nosy: Joe.Borg
priority: normal
severity: normal
status: open
title: Argpase Namespace object methods __eq__ and __ne__ raise TypeError when
comparing to None
versions: Python 2.7, Python 3.1, Python 3.2, Python 3.3, Python 3.4, Python 3.5
___
Python tracker
<http://bugs.python.org/issue21481>
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue21481] Argpase Namespace object methods __eq__ and __ne__ raise TypeError when comparing to None
Joe Borg added the comment: I believe this comes from doing vars(None). But why would this be happening if Namespace is empty. -- ___ Python tracker <http://bugs.python.org/issue21481> ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue16767] Cannot install Python 2.7 in Wine 1.4.1
New submission from Joe Borg: Trying to install Python 2.7 (in particular) under Wine 1.4.1, but keep getting the "Installer has ended prematurely" error. I've also tried with 2.6 and 3.2 and different versions of wine; same error. Also tried with winetricks; again, same error. I've used the msiexec command to log the output of the installer and will attach. Thanks. -- components: Windows files: install.log messages: 178061 nosy: Joe.Borg priority: normal severity: normal status: open title: Cannot install Python 2.7 in Wine 1.4.1 versions: Python 2.7 Added file: http://bugs.python.org/file28421/install.log ___ Python tracker <http://bugs.python.org/issue16767> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue21887] Python3 can't detect Tcl
New submission from Joe Borg: Trying to configure 3.4.1 on Cent OS 6.4. I have built Tcl and Tk, using the prefix /scratch/root. I can confirm the builds with: $ find /scratch/root/ -name "tcl.h" /scratch/root/include/tcl.h $ find /scratch/root/ -name "tk.h" /scratch/root/include/tk.h But, when configuring Python, they aren't picked up: $ ./configure --prefix=/scratch/root --with-tcltk-includes=/scratch/root/include --with-tcltk-libs=/scratch/root/lib | grep tcl checking for --with-tcltk-includes... /scratch/root/include checking for --with-tcltk-libs... /scratch/root/lib checking for UCS-4 tcl... no I've tried to make install with this, but then get the usual exception from _tkinter. -- components: Build messages: 221964 nosy: Joe.Borg priority: normal severity: normal status: open title: Python3 can't detect Tcl versions: Python 3.4 ___ Python tracker <http://bugs.python.org/issue21887> ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue21887] Python3 can't detect Tcl/Tk 8.6.1
Changes by Joe Borg : -- title: Python3 can't detect Tcl -> Python3 can't detect Tcl/Tk 8.6.1 ___ Python tracker <http://bugs.python.org/issue21887> ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue17321] Better way to pass objects between imp.find_module() and imp.load_module()
New submission from Joe Borg: If I want to use imp to find some load modules, I have to do it in quite an "unpythonic" way: >>> name = "os" >>> file, pathname, description = imp.find_module(name) >>> imp.load_module(name, file, pathname, description) Can there not be a better way to pass the tuple (or other object) between find_module() and load_module()? I'm happy to patch this myself and submit, if it's not already been thought about. -- messages: 183220 nosy: Joe.Borg priority: normal severity: normal status: open title: Better way to pass objects between imp.find_module() and imp.load_module() versions: Python 2.7 ___ Python tracker <http://bugs.python.org/issue17321> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue17321] Better way to pass objects between imp.find_module() and imp.load_module()
Joe Borg added the comment: Thanks for the swift feedback guys, if this is deprecated (i.e. not being carried to Python 3) then close the bug; I'll look at `importlib` and see if it carries the same problem. -- type: behavior -> ___ Python tracker <http://bugs.python.org/issue17321> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue17321] Better way to pass objects between imp.find_module() and imp.load_module()
Joe Borg added the comment: Hi Brett, I missed the fact that it's deprecated as it's not mentioned in the 2.* docs http://docs.python.org/2/library/imp.html. I can see it is in the 3.*. Thanks for the feedback. -- ___ Python tracker <http://bugs.python.org/issue17321> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
