lkml.org 
[lkml]   [2010]   [Sep]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: RFC: get_maintainer.pl: append reason for cc to the name by default
On Mon, 27 Sep 2010 11:44:41 -0400
Ted Ts'o <tytso@mit.edu> wrote:

> On Mon, Sep 27, 2010 at 04:57:48PM +0200, Florian Mickler wrote:
> >
> > Would in essence have the same effect, but I think it's slightly better
> > to have some shorter tags in the mail addresses, as I expect them to
> > actually show up on lkml quite a bit.
>
> What if there are no tags on mail addresses that come from the
> MAINTAINERS file, but only tags on the mail addresses that come from
> guessing wildly based on git sign-offs? People should just be looking
> in the MAINTAINERS file, after all, and I don't think that's something
> that needs an explanation.
>
> The thing that needs explanation is when someone like Steve Hemminger
> gets cc'ed on a patch for fs/ext4/acl.c, which really makes no sense
> at all, where you desperately need some kind of tag:
>
> shemminger@vyatta.com (Wild guess using get_maintainer.pl --git)
>
> - Ted

Might make sense to omit the tags on the MAINTAINER-source. I can
agree to your reasoning there.


As far as I can see, the use and the use cases for the git-part of
get_maintainer.pl are mostly to get patches to not-so-well-maintained
parts of the tree upstream. What I envision for the git part is a
scoring based classification scheme that uses all readily available
information of the git-history to determine relevant people for patch
review and patch routing.

I already have implemented a small parser that extracts that information
out of git-log and makes them available to the script, but didn't have
time to wire it up yet.

Another improvement (beyond finding a decent heuristic based on the
artifacts 'authorship', 'signed-off-by', 'reviewed-by', 'acked-by',
'committer' and nr-of-lines-changed.. and maybe time) is probably to
not make an arbitrarily 1-Year-Back cut-off, but to check the last N
commits on that region of the tree. (I'm thinking of the more
"settled down" areas of the tree here)
But let's see what I come up with...

Regards,
Flo


\
 
 \ /
  Last update: 2010-09-27 20:11    [from the cache]
©2003-2011 Jasper Spaans