Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 10 Mar 2001 12:40:52 -0800
From:      Jordan Hubbard <jkh@osd.bsdi.com>
To:        phk@critter.freebsd.dk
Cc:        mwlucas@blackhelicopters.org, hackers@FreeBSD.ORG
Subject:   Re: context or unified diffs in PRs? 
Message-ID:  <20010310124052O.jkh@osd.bsdi.com>
In-Reply-To: <39587.984249413@critter>
References:  <20010310124249.A87381@blackhelicopters.org> <39587.984249413@critter>

next in thread | previous in thread | raw e-mail | index | archive | help
> Unified diffs are also context diffs.
> 
> Context diffs are named such because they contain undisturbed context
> around the changed lines, unlike normal diffs.

Erm, no. :)

Both context and unidiffs show surrounding context, it's simply the
meta-data format which changes.  In the case of unidiffs, the
presentation of multiple diffs in one file is rather more readable and
easy to edit without throwing off patch(1), otherwise they look almost
identical in their presentation of "undisturbed context" for one diff:

From the diff(1) man page:
       -c     Use the context output format.
       -u     Use the unified output format.

[and a list of ten numbers, the new version eliding the "five"]

% diff -c numbers.orig numbers
*** numbers.orig        Sat Mar 10 12:34:51 2001
--- numbers     Sat Mar 10 12:34:54 2001
***************
*** 2,8 ****
  two
  three
  four
- five
  six
  seven
  eight
--- 2,7 ----

% diff -u numbers.orig numbers
--- numbers.orig        Sat Mar 10 12:34:51 2001
+++ numbers     Sat Mar 10 12:34:54 2001
@@ -2,7 +2,6 @@
 two
 three
 four
-five
 six
 seven
 eight

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010310124052O.jkh>