Hi, Mark: Mark Schlegel wrote:
> > I would like to cvs some binary files with the -kb option, but they are > so large that I > would like to only retain the current/latest copy. So if I get newer > copies of these > database files from the source I get them from, I would like the cvs > commit to > replace the existing files with the new files to keep the cvs repository > from getting > too large. So basically the cvs would only be one file deep for these > files > (I'm assuming that since cvs doesn't do binary files in diff'ed format > like it does > with non-binary files, it just stores the entire file) > Hey, Greg! it seems this one is for you. > Is there a way to do this? > Well, even if there's a way, why you should want to do this? (well, I know you *want* it, but surely you reasons are not so "reasonable"). Don't put the files under CVS at all. Disregarding the 'C', what remains stands for Versioning System. You don't want "Versioning", then you don't want CVS. What I would suggest is well, have that file under the same box your repository is, and get it by whatever means that fits. You can write a little silly script that recovers/uploads that file at leisure and that's all (probably you will find sooner or later that "uuuuups, too fast handed. That file shouldn't go there... let's recover... uuuuuups, I just keep the last file revision"). -- SALUD, Jes�s *** [EMAIL PROTECTED] *** _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
