Miles Bader wrote:
> Junio C Hamano <gitster@xxxxxxxxx> writes:
> > How readable can you make this for human consumption while still keeping
> > it machine readable? The answer could be it already is human readble.
> >
> > Two reasons I ask the above question are that I find the feature quite
> > interesting, and would want to see if it can be also fed to humans, and
> > that the combination of this new option and the existing --color-words is
> > misnamed.
>
> There's the format used by the "wdiff" program, which is more like
> traditional diff output in that it doesn't use color, but is human
> friendly, and also seems to be somewhat machine-parseable:
>
> $ echo 'This is a test' > /tmp/a
> $ echo 'This is funky test' > /tmp/b
> $ wdiff /tmp/a /tmp/b
> This is [-a-] {+funky+} test
>
> [I say "somewhat" because wdiff itself doesn't appear to escape potentially
> ambiguous content, e.g., if there's actually a "{+" in the file....]
Junio C Hamano wrote:
> If you call this --word-diff, then it would become more clear that
> --color-words perhaps should have been called --word-diff=color or
> something.
Excellent ideas! I don't have anything to add ;-)
This makes [1/2] a rather new patch though, I moved the whole
prefix/suffix handing further out to accommodate different styles.
There's a little change in [2/2] apart from the obvious option
renaming, too: it interprets --color-words and --word-diff as an
initial setting for the checkbox.
Thomas Rast (2):
diff: add --word-diff option that generalizes --color-words
gitk: add the equivalent of diff --color-words
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html