Re: Fwd: Hdup feature suggestion

[On 25 Oct, @ 22:00, Robert wrote in "Re: [hdup-user] Fwd: Hdup feat ..."]
> this, if you have some sort of a specification for how you want the 
> .nobackup functionality to work.

couldn't resist, so I started on making some sort of directory crawl
tool, which is going to be folded in hdup. And maybe released as a
seperate tool - dunno...

Some stuff to think about:
0.
All include/exclude usage will result in 2 passes over the disk, one
to collect the list for tar and then the tarring itself.

1.
The include/exclude stuff is going to be regexp based. Not shell
globbing.  Should these patterns only apply to files, or also to
directories?

1.5
So 'exclude = .*\.html' will exclude all *.html files, should it
also skip dirs with the same name?

2. 
When including/excluding hdup will look for .nobackup files. No matter
what, if a .nobackup is found the directory is not backupped. And
neither does hdup descend the current dir.

3.
include/exclude clashes. One could conceive conflicting configuration
options. What to do? Make an explicit include weigh heavier?


My (not final answers):
0. 
So what?!

1. 
Globbing is ancient and not consistent. Regexps are the way to go
and more powerfull - but maybe also slower. I opt for regexps.

1.5 
Dunno - I think consistency is important here - so the regexp
applies to everything; dir + files.

2:
I think what I said, is what hdup should do.

3: 
explicit include are more important than exclude. 

grtz 
  -- Miek

PS the .nobackup files should be named .hdup_nobackup (and empty??)