Re: [PATCH/RFC 0/4] Perl rewrite of Ruby git-related

Eric Sunshine [off-list ref] writes:

In this submission, the command name has changed to git-contacts since
git-related felt too generic. (git-contacts seemed best of several
possibilities I surveyed: git-people, git-interested, git-mentioned,
git-blame-us.)

I admit I am pretty bad at naming, but "contacts" sounds like the
most sensible name for what it wants to do (blame-us sounds cute to
my ears, though ;-).

No attempt is made to answer Junio's v9 review[5], as I lack sufficient
insight with '-C' options to be able to respond properly.

I just wanted to see if we want to allow the end user of this script
to specify what -C level they want the underlying blame to use, or
just a hardcoded one should suffice (and if so an explanation why).

My Perl may be rusty and idiomatic usage may be absent.

That is OK.  We need to start somewhere.

Thanks.  Folks, please discuss ;-). 

On Mon, Jul 1, 2013 at 1:00 PM, Junio C Hamano [off-list ref] wrote:

Eric Sunshine [off-list ref] writes:

quoted

No attempt is made to answer Junio's v9 review[5], as I lack sufficient
insight with '-C' options to be able to respond properly.

I just wanted to see if we want to allow the end user of this script
to specify what -C level they want the underlying blame to use, or
just a hardcoded one should suffice (and if so an explanation why).

It might indeed make sense to allow the user control over the -C
level, just as the user likely should have control over other
assumptions made by the script (to wit: $since = '5-years-ago';
$min_percent = 10). More fodder for future patches.

Now that I've read up more carefully on the -C option, I too wonder
why git-blame and git-show are invoked with different -C levels. At
the very least, I think they should be unified in this initial patch
to either one or two -C's. Since the script already is quite slow, I'm
leaning toward just one -C.