Add "daemon" argument to threading.Thread constructor

[tebeka]

BPO	6064
Nosy	@rhettinger, @tebeka, @pitrou
Files	threading.diff: Patch threading.diff: Diff againt r73465 threading.py threading3k.diff: Diff 6064-multiprocessing-daemon-kwarg.patch: Patch for multiprocessing threading3k.diff: patch for threading multiprocessing3k.diff: patch for multiprocessing

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2011-02-25.22:08:28.985>
created_at = <Date 2009-05-19.19:14:00.577>
labels = ['easy', 'type-feature', 'library']
title = 'Add "daemon" argument to threading.Thread constructor'
updated_at = <Date 2011-02-25.22:51:15.316>
user = 'https://github.com/tebeka'

bugs.python.org fields:

activity = <Date 2011-02-25.22:51:15.316>
actor = 'tebeka'
assignee = 'none'
closed = True
closed_date = <Date 2011-02-25.22:08:28.985>
closer = 'pitrou'
components = ['Library (Lib)']
creation = <Date 2009-05-19.19:14:00.577>
creator = 'tebeka'
dependencies = []
files = ['14029', '14312', '14313', '14314', '15376', '18152', '18153']
hgrepos = []
issue_num = 6064
keywords = ['patch', 'easy']
message_count = 21.0
messages = ['88088', '88094', '88142', '88169', '89465', '89467', '89470', '89471', '89472', '89473', '89474', '89475', '89481', '89836', '95585', '111343', '111344', '111346', '111347', '129443', '129451']
nosy_count = 9.0
nosy_names = ['rhettinger', 'tebeka', 'pitrou', 'jnoller', 'ptn', 'conf', 'asksol', 'bbrazil', 'steven']
pr_nums = []
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue6064'
versions = ['Python 3.3']

[tebeka]

It would be nice if threading.Thread constructor will have a "daemon"
argument as well.

This way we'll be able to write
Thread(target=some_function, daemon=1).start()

Instead of currently writing
t = Thread(target=some_function)
t.daemon = True
t.start()

[rhettinger]

+1

[tebeka]

Attaching a patch against trunk (at revision 72805).

[tebeka]

I think that
self.__daemonic = daemon or self._set_daemon()
is wrong, need to thing about it.

[ptn]

+1, but I can't apply the patch cleanly:

$ patch -p0 < threading.diff
patching file Doc/library/threading.rst
Hunk #1 succeeded at 198 (offset -17 lines).
Hunk #2 succeeded at 211 (offset -17 lines).
Hunk #3 succeeded at 230 (offset -17 lines).
patching file Lib/threading.py
Hunk #1 succeeded at 407 with fuzz 1 (offset -16 lines).
Hunk #2 FAILED at 416.
1 out of 2 hunks FAILED -- saving rejects to file Lib/threading.py.rej
patching file Lib/test/test_threading.py
Hunk #1 succeeded at 350 with fuzz 2 (offset 14 lines).

[tebeka]

I'm attaching a new diff (svn diff > threading.diff), hope this one's OK.

[ptn]

Nope. Our working copies seem to be different. I'm uploading mine,
which I just update to revision 73468. Please diff yours against that
and against HEAD too, just in case.

Hunk 2 on threading.py fails because your attributes use two leading
underscores and mine use only one. I have no idea why hunk4 fails.

[tebeka]

I'm diffing against the 2.7 branch, I guess your comes from the 3.2?

Will checkout 3.2 and do it there as well.

[tebeka]

Attaching a diff against the py3k branch on revision 73468

[ptn]

I'm diffing against the 2.7 branch, I guess your comes from the 3.2?
*forehead meets desktop*

Patch applies cleanly and all tests pass. I used it for a while and
everything seemed OK. After someone else tests, I say we go for it :)

[conf]

+1 to the py3k diff. :)
Hey, I think this daemon property should be set as a keyword argument of
the Thread constructor.

[ptn]

+1 on making it a keyword-only argument.

[pitrou]

+1 from me as well.

[jnoller]

+1 to this, and I'd need to make the same patch/change to
multiprocessing.Process as well

[asksol]

Patch for multiprocessing, adding the daemon kwarg attached.

[steven]

Hello, I've taken the liberty of updating the two patch files here for python3.2 at revision 83065. Apologies if I'm treading on anyones toes here. I'm taking part in a europython sprint and this is the first time I've tried this.

I've modified the multiprocessing patch. The documentation didn't agree with the behaviour of the code, so I've changed the Doc file. I've also added an extra test for the default behaviour.

I've reviewed the code for PEP-8 (as an exercise for me) and run the tests and they passed ok.

Steven

[steven]

Here's the patch for multiprocessing based on python 3 revision 83065, all credit to the original author, all problems probably mine!
Steven

[bbrazil]

I've updated the patch for 3.2 and improved the documentation formatting.

[bbrazil]

Looks like it lost my patch, but Steven's does the same.

One tweak: in threading.rst put all of the class signature on one line or the html formatting will be a bit off.

[pitrou]

I corrected the patches (they were breaking the property of inheriting daemon by default) and committed them in r88618. Thank you!

[tebeka]

Thank you!

On Fri, Feb 25, 2011 at 2:08 PM, Antoine Pitrou <report@bugs.python.org> wrote:

Antoine Pitrou <pitrou@free.fr> added the comment:

I corrected the patches (they were breaking the property of inheriting daemon by default) and committed them in r88618. Thank you!

----------
resolution:  -> fixed
stage: patch review -> committed/rejected
status: open -> closed

---

Python tracker <report@bugs.python.org>
<http://bugs.python.org/issue6064\>

---