home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.perl
- From: Sean@legastat.demon.co.uk (Sean Hughes)
- Path: sparky!uunet!pipex!demon!legastat.demo.co.uk!Sean
- Subject: DIFF
- Distribution: world
- Organization: Legastat
- Lines: 20
- Date: Tue, 15 Sep 1992 16:38:51 +0000
- Message-ID: <716575131snx@legastat.demo.co.uk>
- Sender: usenet@gate.demon.co.uk
-
- Hello happy perl people
-
- Has anyone written a diff-like routine?
-
- What I'm trying to do is to compare a target line of text against a
- whole document in order to find the text that matches the target most
- closely at a charcater level. The hamfisted approach would be to use
- split to break each line into one character per line lines and run
- system calls to diff - but this strikes me as pretty silly.
-
- I appreciate I should get into the diff algorithms(^) but before I start
- re-inventing square wheels has anybody done this kind of thing already ?
-
- (^ does anyone know an ftp or some such site where I can get hold of:
- "An O(ND) Difference Algorithm and its Variations", Eugene Myers,
- Algorithimica Vol. 1 No. 2, 1986, p251 )
-
- Many thanks
-
- Sean at the core dump
-
-