From: Jeff King <peff@peff.net>
Date: Fri, 06 May 2011 22:13:00 GMT
Subject: Re: [PATCH 1/2] tests: eliminate unnecessary setup test assertions
Message-ID: <20110506221300.GB17848@sigill.intra.peff.net>
In-Reply-To: <20110506214801.GA17848@sigill.intra.peff.net>

On Fri, May 06, 2011 at 05:48:01PM -0400, Jeff King wrote:

>   pattern_authors() {
>     git grep -n "$@" |
>     while IFS=: read file line match; do
>       git blame -L "$line,$line" "$file"
>     done |
>     perl -lpe '/\((.*?) \d+-\d+-\d+/; $_=$1'
>   }

Two minor complaints on git-blame; maybe somebody can point out
something clever I've missed.

  1. blame's "-L" understands patterns already. So in theory I could
     tell it "blame all lines that match pattern X". But I don't think
     there is a way to do that (it tries looking to for _one_ range to
     blame for each -L, not a set of ranges).

  2. Parsing the human-readable output blame output sucks. But parsing
     --porcelain is annoyingly complex for quick-and-dirty things like
     this. It doesn't repeat the commit information per-line.

     I guess we could have an inefficient --line-porcelain format that
     breaks down ranges into single lines and repeats the commit info
     for each one.

     The clever among you may notice that in this particular case,
     though, I could have gotten away with regular --porcelain as I
     blame a single line at a time.

-Peff