threads / discuss / 35885

Is there something like a blamed diff?

Subject: Is there something like a blamed diff?

## tl;dr

4 messages between Feb 15, 2014 and Feb 17, 2014.

replies: 3people: 2as markdown or json

David Kastrup· Feb 15, 2014, 18:38 UTC · lore

When comparing two branches, decorating the flat diff with the respectively responsible commits seems like it would be nice to do/have (the blame on the identical parts, in contrast, is not really interesting). Is there any tool that provides something like that?

-- 
David Kastrup
Thomas Rast· Feb 17, 2014, 10:24 UTC · re: David Kastrup · lore

Re: Is there something like a blamed diff?

David Kastrup <dak@gnu.org> writes:
> When comparing two branches, decorating the flat diff with the
> respectively responsible commits seems like it would be nice to do/have
> (the blame on the identical parts, in contrast, is not really
> interesting).  Is there any tool that provides something like that?
This seems to come up every year or so:
http://thread.gmane.org/gmane.comp.version-control.git/110369/focus=110383
-- 
Thomas Rast
tr@thomasrast.ch
David Kastrup· Feb 17, 2014, 11:02 UTC · re: Thomas Rast · lore

Re: Is there something like a blamed diff?

Thomas Rast <tr@thomasrast.ch> writes:
Show 10 quoted lines
> David Kastrup <dak@gnu.org> writes:
>
>> When comparing two branches, decorating the flat diff with the
>> respectively responsible commits seems like it would be nice to do/have
>> (the blame on the identical parts, in contrast, is not really
>> interesting).  Is there any tool that provides something like that?
>
> This seems to come up every year or so:
>
> http://thread.gmane.org/gmane.comp.version-control.git/110369/focus=110383

Nice. That one could likely be sped up by calling git-blame just once on each file with multiple -L options.

Though actually with the current implementation of git-blame you are likely faster for a large number of differences by doing each region separately anyway since then the cost for digging through the history is obviously O(n) for n consecutive runs on n chunks while the administrative overhead for several chunks in a single run is O(n^2) at the moment.

-- 
David Kastrup
Thomas Rast· Feb 17, 2014, 17:41 UTC · re: David Kastrup · lore

Re: Is there something like a blamed diff?

David Kastrup <dak@gnu.org> writes:
Show 15 quoted lines
> Thomas Rast <tr@thomasrast.ch> writes:
>
>> David Kastrup <dak@gnu.org> writes:
>>
>>> When comparing two branches, decorating the flat diff with the
>>> respectively responsible commits seems like it would be nice to do/have
>>> (the blame on the identical parts, in contrast, is not really
>>> interesting).  Is there any tool that provides something like that?
>>
>> This seems to come up every year or so:
>>
>> http://thread.gmane.org/gmane.comp.version-control.git/110369/focus=110383
>
> Nice.  That one could likely be sped up by calling git-blame just once
> on each file with multiple -L options.

Yes. You'll note that the email predates the support for multiple -L options by a few years :-)

I never had a need for such a script, I just wrote that in response to someone asking about it. If you find it useful, please clean it up for inclusion in contrib/.

-- 
Thomas Rast
tr@thomasrast.ch

← back to recent threads