[ANNOUNCE] git-related-0.1

[ANNOUNCE] git-related-0.1

[Felipe Contreras]

Hi,

This is basically the same as my original series[1] which was dropped
for no good reasons, plus all the other features I also sent patches
for.

This is way superior to the current git-contacts that sits on the contrib area.

For the uninitiated, this tool finds people that might be interested
in a patch, by going back through the history for each single hunk
modified, and finding people that reviewed, acknowledged, signed, or
authored the code the patch is modifying.

It does this by running `git blame` incrementally on each hunk, and
then parsing the commit message. After gathering all the relevant
people, it groups them to show what exactly was their role when the
participated in the development of the relevant commit, and on how
many relevant commits they participated. They are only displayed if
they pass a minimum threshold of participation.

For example:

  % git related master..fc/transport/improv
  Junio C Hamano <gitster@pobox.com> (signer: 90%, author: 5%)
  Felipe Contreras <felipe.contreras@gmail.com> (author: 25%, reviewer: 2%)
  Sverre Rabbelier <srabbelier@gmail.com> (author: 17%, acker: 2%, signer: 7%)
  Jeff King <peff@peff.net> (acker: 17%, author: 10%)
  Shawn O. Pearce <spearce@spearce.org> (author: 5%, signer: 2%, cced: 2%)
  Elijah Newren <newren@gmail.com> (author: 10%)

In addition, it has an option to output the list of commits, instead
of the contributors, which allows you to easily find out the previous
changes to the lines your patches modify.

  % git related -c master..fc/transport/improv
  99d9ec0 Merge branch 'fc/transport-helper-no-refspec'
  67c9c78 transport-helper: barf when user tries old:new
  0460ed2 documentation: trivial style cleanups
  126aac5 transport-helper: fix remote helper namespace regression
  21610d8 transport-helper: clarify pushing without refspecs
  a93b4a0 transport-helper: warn when refspec is not used
  664059f transport-helper: update remote helper namespace
  c4458ec fast-export: Allow pruned-references in mark file
  ...

Moreover, when sending patches for review, you can configure `git
send-email` to use `git related` to find relevant people that should
be Cc'ed:

  % git send-email --cc-cmd='git related' *.patch

It has way many more features than git-contacts which is a rewrite in
Perl of this script, but only the initial bare minimum stage.

I've also added tests, and this is how git-contacts fares in comparison:

--- expected 2013-10-10 21:13:45.938291298 +0000
+++ actual 2013-10-10 21:13:45.937291298 +0000
@@ -1,2 +1,2 @@
-Jon Stewart <jon@stewart.com> (author: 50%)
-Pablo Escobar <pablo@escobar.com> (author: 50%)
+Jon Stewart <jon@stewart.com>
+Pablo Escobar <pablo@escobar.com>
not ok 1 - basic

git-contacts doesn't show the amount of involvement, but that's easy to fix.

--- expected 2013-10-10 21:13:45.975291299 +0000
+++ actual 2013-10-10 21:13:45.974291299 +0000
@@ -1,2 +1,3 @@
-John Poppins <john@doe.com> (author: 66%)
-Jon Stewart <jon@stewart.com> (reviewer: 33%, author: 33%)
+John Doe <john@doe.com>
+John Poppins <john@doe.com>
+Jon Stewart <jon@stewart.com>
not ok 2 - others

git-contacts doesn't show the kind of involvement (reviewer, author)
nor does it group people by their email address.

(skip other failures)

expecting success:
git related -craw -1 master | git log --format='%s' --no-walk --stdin
> actual &&
cat > expected <<-EOF &&
four
three
one
EOF
test_cmp expected actual

not ok 7 - commits

Also, it doesn't have the option to show the commits themselves, which
is useful to investigate the diffs. This feature alone is what makes
the make "git-contacts" not appropriate, because if this feature is
implemented, 'git contacts --commits' wouldn't have anything to do
with "contacts".

For that matter it doesn't support any options at all (e.g.
--min-percent, --since), and doesn't have a usage helper (--help).

Just put git-related in your PATH, and enjoy :)

Cheers.

[1] http://article.gmane.org/gmane.comp.version-control.git/226058

-- 
Felipe Contreras

Re: [ANNOUNCE] git-related-0.1

[Felipe Contreras]

On Thu, Oct 10, 2013 at 4:43 PM, Felipe Contreras
<felipe.contreras@gmail.com> wrote:

> Just put git-related in your PATH, and enjoy :)

And of course I forgot the link:

https://github.com/felipec/git-related

-- 
Felipe Contreras

Re: [ANNOUNCE] git-related-0.1

[Andrew Ardill]

On 11 October 2013 08:43, Felipe Contreras <felipe.contreras@gmail.com> wrote:
> After gathering all the relevant
> people, it groups them to show what exactly was their role when the
> participated in the development of the relevant commit, and on how
> many relevant commits they participated.

It looks like you group by emails, any idea if anyone purposefully
uses the same email but a different name? I don't know if the email is
assumed to be the unique identifier, or if the name/email pair is,
however I think the typical use case is definitely to group by email
so it's a useful default behaviour.

> They are only displayed if
> they pass a minimum threshold of participation.

Out of interest, how is the threshold determined and is it configurable?

Regards,

Andrew Ardill

Re: [ANNOUNCE] git-related-0.1

[Felipe Contreras]

On Thu, Oct 10, 2013 at 5:24 PM, Andrew Ardill <andrew.ardill@gmail.com> wrote:
> On 11 October 2013 08:43, Felipe Contreras <felipe.contreras@gmail.com> wrote:
>> After gathering all the relevant
>> people, it groups them to show what exactly was their role when the
>> participated in the development of the relevant commit, and on how
>> many relevant commits they participated.
>
> It looks like you group by emails, any idea if anyone purposefully
> uses the same email but a different name? I don't know if the email is
> assumed to be the unique identifier, or if the name/email pair is,
> however I think the typical use case is definitely to group by email
> so it's a useful default behaviour.

Yes, I think I described that in the announcement. It was a feature
somebody requested, and it was easy to implement.

>> They are only displayed if
>> they pass a minimum threshold of participation.
>
> Out of interest, how is the threshold determined and is it configurable?

The threshold by default is 10%, but it's configurable with --min-percent.

-- 
Felipe Contreras