[issue10618] regression in subprocess.call() command quoting

2010-12-14 Thread R. David Murray
R. David Murray rdmur...@bitdance.com added the comment: Tim: we just do our best to guess, and try to err on the conservative side. But things will always happen. Given benjamin's reply I'm closing the issue. Mercurial would have to add conditional code now anyway no matter what we do.

[issue10618] regression in subprocess.call() command quoting

2010-12-13 Thread Mads Kiilerich
Changes by Mads Kiilerich m...@kiilerich.com: -- nosy: +kiilerix ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue10618 ___ ___ Python-bugs-list

[issue10618] regression in subprocess.call() command quoting

2010-12-04 Thread Tim Golden
Tim Golden m...@timgolden.me.uk added the comment: I'm not quite sure how anyone's supposed to determine which bugs are likely to have been worked around and which haven't :) I'm also unsure why a clear bugfix shouldn't make it into a minor version release. Surely this isn't the only one to do

[issue10618] regression in subprocess.call() command quoting

2010-12-04 Thread Éric Araujo
Changes by Éric Araujo mer...@netwok.org: -- nosy: +eric.araujo ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue10618 ___ ___ Python-bugs-list

[issue10618] regression in subprocess.call() command quoting

2010-12-03 Thread Dirkjan Ochtman
New submission from Dirkjan Ochtman dirk...@ochtman.nl: http://mercurial.selenic.com/bts/issue2534 is a regression in 2.7.1 relative to 2.7, around the way commands are called (it seems to concern subprocess.call()). The error seems to be raised from mercurial's util.system() call, which uses

[issue10618] regression in subprocess.call() command quoting

2010-12-03 Thread Georg Brandl
Georg Brandl ge...@python.org added the comment: The culprit seems to be r83957, which was done to fix #2304. Assigning to Tim, who committed that revision. -- assignee: - tim.golden nosy: +georg.brandl, tim.golden ___ Python tracker

[issue10618] regression in subprocess.call() command quoting

2010-12-03 Thread Georg Brandl
Georg Brandl ge...@python.org added the comment: In util.system(), Mercurial adds its own pair of quotes: if os.name == 'nt': cmd = '%s' % cmd That will result in one level of quoting too much. Now it seems unfortunate that this change was done in a minor version. It is definitely

[issue10618] regression in subprocess.call() command quoting

2010-12-03 Thread Dirkjan Ochtman
Dirkjan Ochtman dirk...@ochtman.nl added the comment: Ask, Jesse, sorry, I got confused about multiprocessing and subprocess. Taking you out of the nosy now. -- nosy: -asksol, jnoller ___ Python tracker rep...@bugs.python.org

[issue10618] regression in subprocess.call() command quoting

2010-12-03 Thread Benjamin Peterson
Benjamin Peterson benja...@python.org added the comment: 2010/12/3 Georg Brandl rep...@bugs.python.org: Georg Brandl ge...@python.org added the comment: In util.system(), Mercurial adds its own pair of quotes:    if os.name == 'nt':        cmd = '%s' % cmd That will result in one level