[lug] sym link via patch

D. Stimits stimits at idcomm.com
Thu Apr 19 00:43:26 MDT 2001


Nate Duehr wrote:
> 
> Would CVS keep the links properly in binary mode and then be able to
> provide a `cvs diff`?  Just a thought, didn't try it.

cvs and sym links don't get along well without a patch. But even if cvs
can do diffs (and it can quite well), it is the patch program itself I
am having problems with. If I have a new file, diff can create patch
files that will then add the new file when running patch. However, what
I want is not a new file, but a sym link to a new file. One file, but
aliased as a second name in a second directory, where changing one
alters the other.

As for the need for sym links, it is an old and sad tail of inherited
code.

D. Stimits, stimits at idcomm.com

> 
> On Wed, Apr 18, 2001 at 06:45:20PM -0600, D. Stimits wrote:
> > I'm trying to find out if there is a way to run a diff between two
> > directory trees to create a patch, but such that symbolic links are
> > preserved. For example:
> >
> > newest_version/hard_link.h
> > newest_version/sym_link.h (points at hard_link.h)
> >
> > older_version/hard_link.h
> >
> > My goal is a recursive diff such as:
> > diff -c -r -N oldest_version newest_version > some.patch
> >
> > The problem I'm running into is that the sym link file, after running
> > patch to update an old tree, will be added as a hard link rather than a
> > sym link. Does anyone here know if it is possible to preserve sym links
> > with a patch?
> >
> > D. Stimits, stimits at idcomm.com
> > _______________________________________________
> > Web Page:  http://lug.boulder.co.us
> > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> 
> --
> Nate Duehr <nate at natetech.com>
> 
> GPG Key fingerprint = DCAF 2B9D CC9B 96FA 7A6D AAF4 2D61 77C5 7ECE C1D2
> Public Key available upon request, or at wwwkeys.pgp.net and others.
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug



More information about the LUG mailing list