Hi All,
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
How do I do that?
cheers,
Chris
On 06/01/2010 11:19, Chris Withers wrote:
Hi All,
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
How do I do that?
Bug triaging is one of Python's "big needs" and anything you do to help on this score would be much appreciated. Particularly reviewing new and outstanding issues.
I assumed there would be RSS feeds for bug tracker activity but can't easily find these on the tracker. There is a bot that posts activity to #python-dev, so there must be some way of getting this information.
All the best,
Michael
cheers,
Chris
Michael Foord wrote:
I assumed there would be RSS feeds for bug tracker activity but can't easily find these on the tracker. There is a bot that posts activity to #python-dev, so there must be some way of getting this information.
Yeah, email-out is what I'm really after... I have it for my own Roundup instance so it can't be that hard to do ;-)
Roché, you guys host the bug tracker, right? Is there email-out set up for it?
Chris
On Wed, 2010-01-06 at 11:30 +0000, Chris Withers wrote:
Michael Foord wrote:
I assumed there would be RSS feeds for bug tracker activity but can't easily find these on the tracker. There is a bot that posts activity to #python-dev, so there must be some way of getting this information.
Yeah, email-out is what I'm really after... I have it for my own Roundup instance so it can't be that hard to do ;-)
Roché, you guys host the bug tracker, right? Is there email-out set up for it?
We do, but we don't administer it. There are a few administrators taking care of it and you should be able to reach them by logging your request here: http://psf.upfronthosting.co.za/roundup/meta/ or post it to the Infrastructure mailing list: infrastructure@python.org
Michael Foord wrote:
I assumed there would be RSS feeds for bug tracker activity but can't easily find these on the tracker. There is a bot that posts activity to #python-dev, so there must be some way of getting this information.
I'm pretty sure the bugs list is still the primary spooled notification mechanism: http://mail.python.org/mailman/listinfo/python-bugs-list
There are also the weekly tracker activity summaries that are posted here to python-dev.
Cheers, Nick.
Nick Coghlan wrote:
I'm pretty sure the bugs list is still the primary spooled notification mechanism: http://mail.python.org/mailman/listinfo/python-bugs-list
That's what I was after, thanks!
Chris
On Wed, 06 Jan 2010 11:41:28 +0000, Chris Withers chris@simplistix.co.uk wrote:
Nick Coghlan wrote:
I'm pretty sure the bugs list is still the primary spooled notification mechanism: http://mail.python.org/mailman/listinfo/python-bugs-list
That's what I was after, thanks!
Just for completeness, there's also new-bugs-announce if you want just *new* bug notification. That's more for people who want to watch for bugs they want to become nosy on, though; if you are doing triage python-bugs-list is what you want.
Please also read http://www.python.org/dev/workflow/ if you haven't already. Thanks for being willing to chip in!
--David
"Nick" == Nick Coghlan ncoghlan@gmail.com writes:
Nick> I'm pretty sure the bugs list is still the primary spooled Nick> notification mechanism: Nick> http://mail.python.org/mailman/listinfo/python-bugs-list
Actually, there is a new-bugs-announce list:
http://mail.python.org/mailman/listinfo/new-bugs-announce
Skip
In article 4B4475F3.5040406@gmail.com, Nick Coghlan ncoghlan@gmail.com wrote:
Michael Foord wrote:
I assumed there would be RSS feeds for bug tracker activity but can't easily find these on the tracker. There is a bot that posts activity to #python-dev, so there must be some way of getting this information.
I'm pretty sure the bugs list is still the primary spooled notification mechanism: http://mail.python.org/mailman/listinfo/python-bugs-list
Also, that mailing list (along with most python development related mailing lists) is mirrored at gmane.org which means it can also be obtained via a newsreader (NNTP) or various RSS feeds.
http://dir.gmane.org/gmane.comp.python.bugs
On Wed, Jan 6, 2010 at 12:24 PM, Michael Foord fuzzyman@voidspace.org.uk wrote:
On 06/01/2010 11:19, Chris Withers wrote:
Hi All,
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
How do I do that?
Bug triaging is one of Python's "big needs" and anything you do to help on this score would be much appreciated. Particularly reviewing new and outstanding issues.
Another useful triage I think, is to review the oldest bugs (some of them are > 5 years) and remove the ones that are not relevant anymore, or duplicate with newer entries.
Tarek
Tarek Ziadé wrote:
Another useful triage I think, is to review the oldest bugs (some of them are > 5 years) and remove the ones that are not relevant anymore, or duplicate with newer entries.
I'm sprinting for 2 days at PyCon, I'd verymuch be up for doing this with someone as a paired task for those 2 days...
Chris
On Wed, Jan 6, 2010 at 1:31 PM, Chris Withers chris@simplistix.co.uk wrote:
Tarek Ziadé wrote:
Another useful triage I think, is to review the oldest bugs (some of them are > 5 years) and remove the ones that are not relevant anymore, or duplicate with newer entries.
I'm sprinting for 2 days at PyCon, I'd verymuch be up for doing this with someone as a paired task for those 2 days...
I'll be doing Distutils stuff but I can probably help around a bit in that task: Distutils have quite a few old issues so I can tackle those
Tarek Ziadé wrote:
Another useful triage I think, is to review the oldest bugs (some of them are > 5 years) and remove the ones that are not relevant anymore, or duplicate with newer entries.
I believe someone (Daniel Diniz, maybe?) did do a pass over those some time in the last 12 months, so most of the obviously irrelevant ones that are that old should already be gone. Not to say it isn't worth doing another pass, just saying not to get disheartened if there aren't many that can be readily closed.
There are at least a few still kicking around just because they're difficult to deal with (there's an ancient one to do with one of the ways circular imports can fail that I occasionally go back and reread before moving on to something more tractable).
Cheers, Nick.
On Wed, Jan 6, 2010 at 06:57, Nick Coghlan ncoghlan@gmail.com wrote:
I believe someone (Daniel Diniz, maybe?) did do a pass over those some time in the last 12 months, so most of the obviously irrelevant ones that are that old should already be gone. Not to say it isn't worth doing another pass, just saying not to get disheartened if there aren't many that can be readily closed.
There are at least a few still kicking around just because they're difficult to deal with (there's an ancient one to do with one of the ways circular imports can fail that I occasionally go back and reread before moving on to something more tractable).
Cheers, Nick.
On the topic of bugs that can be readily closed (literally), I've recently come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue remains open. What is the best course of action there? I'd just go ahead and close the issue myself but I don't have tracker privileges.
I'm willing to help out with another Daniel Diniz-esque triage sweep if that would help.
Brian
Le Wed, 06 Jan 2010 08:57:42 -0600, Brian Curtin a écrit :
On the topic of bugs that can be readily closed (literally), I've recently come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue remains open. What is the best course of action there?
Post a message on the issue asking for info.
Antoine Pitrou <solipsis <at> pitrou.net> writes:
Le Wed, 06 Jan 2010 08:57:42 -0600, Brian Curtin a écrit :
On the topic of bugs that can be readily closed (literally), I've recently come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue remains open. What is the best course of action there?
Post a message on the issue asking for info.
Ok, I realize my answer might have been a bit terse :-) The patch might be waiting to be merged in all development branches, or it may not totally resolve the issue, or perhaps documentation needs to be updated, or perhaps it is pending a verdict from the buildbots, etc. You can't deduce that the issue is completely fixed from the simple fact that something has been committed.
Regards
Antoine Pitrou.
On Wed, Jan 6, 2010 at 06:57, Brian Curtin brian.curtin@gmail.com wrote:
On Wed, Jan 6, 2010 at 06:57, Nick Coghlan ncoghlan@gmail.com wrote:
I believe someone (Daniel Diniz, maybe?) did do a pass over those some time in the last 12 months, so most of the obviously irrelevant ones that are that old should already be gone. Not to say it isn't worth doing another pass, just saying not to get disheartened if there aren't many that can be readily closed.
There are at least a few still kicking around just because they're difficult to deal with (there's an ancient one to do with one of the ways circular imports can fail that I occasionally go back and reread before moving on to something more tractable).
Cheers, Nick.
On the topic of bugs that can be readily closed (literally), I've recently come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue remains open. What is the best course of action there? I'd just go ahead and close the issue myself but I don't have tracker privileges.
If a core developer is willing to step forward and vouch for you to get tracker privileges then I will give them to you. We are trying to give out tracker privs w/ less time than required to get commit privileges. So as long as you have helped out on a few issues in a positive and correct way that should be enough to get one of the regulars who perform triage to notice.
-Brett
I'm willing to help out with another Daniel Diniz-esque triage sweep if that
would help.
Brian
Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/brett%40python.org
On Wed, 06 Jan 2010 11:03:32 -0800, Brett Cannon wrote:
On Wed, Jan 6, 2010 at 06:57, Brian Curtin brian.curtin@gmail.com wrote:
On the topic of bugs that can be readily closed (literally), I've recently come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue remains open. What is the best course of action there? I'd just go ahead and close the issue myself but I don't have tracker privileges.
If a core developer is willing to step forward and vouch for you to get tracker privileges then I will give them to you. We are trying to give out tracker privs w/ less time than required to get commit privileges. So as long as you have helped out on a few issues in a positive and correct way that should be enough to get one of the regulars who perform triage to notice.
-Brett
I've done a quick scan of issues Brian is nosy on to refresh my memory, and I'd say he's definitely been making positive contributions. I'm willing to volunteer to keep an eye on his triage work for a while if you grant him tracker privs.
Brian, I assume you'll be cognizant of Antoine's advice about making sure a bug really should be closed before closing it :) Hanging out in #python-dev on freenode while working on issues can be helpful, as well, since you can quickly ask whoever is there for second opinions on particular bugs.
-- R. David Murray www.bitdance.com Business Process Automation - Network/Server Management - Routers/Firewalls
On Wed, Jan 6, 2010 at 17:22, R. David Murray rdmurray@bitdance.com wrote:
On Wed, 06 Jan 2010 11:03:32 -0800, Brett Cannon wrote:
On Wed, Jan 6, 2010 at 06:57, Brian Curtin brian.curtin@gmail.com
wrote:
On the topic of bugs that can be readily closed (literally), I've
recently
come across a number of issues which appear to be sitting in a patch or review stage, but their patches have been committed and the issue
remains
open. What is the best course of action there? I'd just go ahead and
close
the issue myself but I don't have tracker privileges.
If a core developer is willing to step forward and vouch for you to get tracker privileges then I will give them to you. We are trying to give
out
tracker privs w/ less time than required to get commit privileges. So as long as you have helped out on a few issues in a positive and correct way that should be enough to get one of the regulars who perform triage to notice.
-Brett
I've done a quick scan of issues Brian is nosy on to refresh my memory, and I'd say he's definitely been making positive contributions. I'm willing to volunteer to keep an eye on his triage work for a while if you grant him tracker privs.
Done for the username brian.curtin (email doesn't match the one Brian emailed from so do let me know, Brian if this is the right username). Welcome aboard!
-Brett
Brian, I assume you'll be cognizant of Antoine's advice about making sure a bug really should be closed before closing it :) Hanging out in #python-dev on freenode while working on issues can be helpful, as well, since you can quickly ask whoever is there for second opinions on particular bugs.
-- R. David Murray www.bitdance.com Business Process Automation - Network/Server Management - Routers/Firewalls
On Wed, Jan 6, 2010 at 19:28, Brett Cannon brett@python.org wrote:
On Wed, Jan 6, 2010 at 17:22, R. David Murray rdmurray@bitdance.comwrote:
On Wed, 06 Jan 2010 11:03:32 -0800, Brett Cannon wrote:
On Wed, Jan 6, 2010 at 06:57, Brian Curtin brian.curtin@gmail.com
wrote:
On the topic of bugs that can be readily closed (literally), I've
recently
come across a number of issues which appear to be sitting in a patch
or
review stage, but their patches have been committed and the issue
remains
open. What is the best course of action there? I'd just go ahead and
close
the issue myself but I don't have tracker privileges.
If a core developer is willing to step forward and vouch for you to get tracker privileges then I will give them to you. We are trying to give
out
tracker privs w/ less time than required to get commit privileges. So as long as you have helped out on a few issues in a positive and correct
way
that should be enough to get one of the regulars who perform triage to notice.
-Brett
I've done a quick scan of issues Brian is nosy on to refresh my memory, and I'd say he's definitely been making positive contributions. I'm willing to volunteer to keep an eye on his triage work for a while if you grant him tracker privs.
Done for the username brian.curtin (email doesn't match the one Brian emailed from so do let me know, Brian if this is the right username). Welcome aboard!
Yep, that's the one. Thanks!
On Jan 6, 2010, at 7:29 AM, Tarek Ziadé wrote:
On Wed, Jan 6, 2010 at 12:24 PM, Michael Foord fuzzyman@voidspace.org.uk wrote:
On 06/01/2010 11:19, Chris Withers wrote:
Hi All,
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
How do I do that?
Bug triaging is one of Python's "big needs" and anything you do to help on this score would be much appreciated. Particularly reviewing new and outstanding issues.
Another useful triage I think, is to review the oldest bugs (some of them are > 5 years) and remove the ones that are not relevant anymore, or duplicate with newer entries.
I was actually thinking about that the other day when I saw that the average age of bugs on the Python tracker was at some hideously large 3 digit number.
The 'success' statistic would be to bring that down below, say, 100.
S
Hi Chris,
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
Do you really want to get such notifications? There may be a lot of them. If you want however, you can join #python-dev on IRC (irc.freenode.net) where there's a bot which posts updates of all bugs on the tracker. There's usually not a lot of discussion going on so you probably won't feel flooded.
In addition to bug triage, what is needed is reviewing of existing patches, as well as writing patches for issues which haven't been addressed yet :-)
Regards
Antoine.
On Wed, Jan 6, 2010 at 8:19 AM, Chris Withers chris@simplistix.co.uk wrote:
Is there a high volume of incoming bugs to the Python tracker? If so, I'd like to help with triaging. I think I have all the necessary access, what I'm missing is the knowledge of how to set myself up to get notifications of new bugs...
Not notifications, but maybe a way to have a higher look of them for easy selection:
http://www.taniquetil.com.ar/cgi-bin/pytickets.py
Regards,
-
Antoine Pitrou -
Brett Cannon -
Brian Curtin -
Chris Withers -
Facundo Batista -
Michael Foord -
Ned Deily -
Nick Coghlan -
R. David Murray -
Roché Compaan -
skip＠pobox.com -
ssteinerX＠gmail.com -
Tarek Ziadé