Re: [Gnu-arch-users] Some issues

[Cameron Patrick]

Andrew Suffield wrote:

| > We could do better than we do now, though.  Something like rdiff or
| > xdelta would be more efficient for some binary files.  We could grok
| > gzip/zip/bzip2 files and store some kind of patch based on the
| > compressed data.  Having said that, I'm not sure that any of this is
| > worth the extra complexity.
| 
| And you're just talking compression - that's really not very
| interesting. Absent a useful patch equivalent, there's no semantic
| difference, so a smarter server or filesystem could do it in the
| background without affecting the changeset format.

That's not quite true.  If arch looked inside zipped files, say, it
could compute patches to the text data inside Openoffice documents.
This /is/ a semantic difference.  While it isn't possible to write a
clever diff+patch algorithm for arbitrary binary files, it is
certainly possible to write diff/patch equivalents that understand
/specific/ types of files.

IIRC Subversion allows you to define external commands to run when
applying patches (actually 3-way merging I suppose) to specific types
of (non-text) files.  I don't know if anyone actually uses this
functionality.

Cameron.