I used to work for a company that stored their electronic designs in CVS.
They stored them in binary mode in the repository, and modified CVS and RCS
to support compressed RCS files.
You do not want to compress files before committing them to CVS. CVS has
a binary diff algorithm that's reasonably efficient provided differences are
small. Compression tends to convert small changes into very large diffs,
thus making the RCS files grow faster than they would by compressing RCS
files containing the raw data.
This was a long time ago, in the CVS v1.3 days and RCS v5.6. But the work
is available in alpha quality on the web at
http://www.wakawaka.com/source.html
--- Forwarded mail from [EMAIL PROTECTED]
has anybody tried to use CVS for storing/versioning data for electronic
design? If yes, how did you deal with binary databases? Did you use
filters which convert binary files into a compressed archive and stored
the archive? Did you use two systems, one for ASCII data, one for binary
data?
--- End of forwarded message from [EMAIL PROTECTED]
_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
