[lug] File Compression Question

[Tkil]

>>>>> "David" == David Morris <boru at frii.com> writes:

David> I have a few directories filled mostly with text files.
David> Combined, I have about 800 MB to 900 MB of plain text files
David> that can be compressed down to 200 MB or so at best.

David> I was just wondering if LINUX offers any way to to compress
David> files, but still allow read access to the files...though at a
David> slower rate, of course.  Sort of like the native compression
David> available NTFS...not the best compression in the world, but it
David> can at times save a huge amount of space on the hard disk.

there have been various compressed filesystems made for linux.  do a
search for "e2compr", for instance; also, "cramfs" has some of these
attributes, but it's for a very different purpose.  it doesn't look
like e2compr has been worked on much lately, tho.  :(

my personal tactic for these is to simply put them into tar.gz files,
and then use emacs (which can uncompress and un-tar on the fly) to
browse them when needed.

it becomes a little more cumbersome to do the equivalent of a "grep"
in this situation, but it's much more lightweight than any filesystem
could be.  also, you get much better compression on a single tar file
than you would on block-sized chunks (as with any sane compressed FS),
and it will definitely be higher than most small files could compress
on their own.

t.