[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Diff against a file that doesn't exist in HEAD?

From: Greg Hudson <ghudson_at_MIT.EDU>
Date: 2006-03-14 07:54:22 CET

On Mon, 2006-03-13 at 18:30 -0800, Justin Erenkrantz wrote:
> Why is 'diff -rX:Y URL' not equivalent to 'diff URL@X URL@Y'?

svn diff -r 9000:10000
http://svn.collab.net/repos/svn/trunk/subversion/libsvn_fs_fs/dag.c

dag.c was copied from libsvn_fs_base between 9000 and 10000. This
command would fail if it were equivalent to --old=URL@X --old=URL@Y.

It's pretty clear that -r 9000:10000 should be a pegged diff; there's no
way you'd want to be comparing two completely different objects, or
having the comparison fail, if you're using the -r X:Y syntax across a
copy operation. It's arguable whether the peg rev for URLs should
default to HEAD or to Y (or perhaps X), but I think HEAD makes the most
sense.

> This is completely counter-intuitive.

There is nothing that's intuitive in all cases once you introduce
directory versioning. People didn't like the old way either; we
wouldn't have implemented something as mind-twisting as peg-revs if we
weren't getting lots of complaints.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Mar 14 07:55:45 2006

This is an archived mail posted to the Subversion Dev mailing list.