Interesting. That was one of my original ideas (to just stuff the tag name in the meta-data of the file and have the getLabel code do all the work by searching all the files with that tag name in the meta-data). Just did not seem to scale very well.
But, now that I revisit the issue, it would be the best of both worlds to have everything in the archive and have the current version point to the archive as you say CVS does. But, unlike CVS, still have labels exist as entities with the pointers. This would avoid the issues of my case 1, but without the duplications of case 2. Thanks for giving me a fresh view of things. Any further suggestions, warnings, etc are appreciated. dave _______________________________________________ Info-cvs mailing list Info-cvs@nongnu.org http://lists.nongnu.org/mailman/listinfo/info-cvs
