home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.std.internat
- Path: sparky!uunet!think.com!paperboy.osf.org!down.osf.org!lehotsky
- From: lehotsky@down.osf.org (Alan Lehotsky)
- Subject: Re: Dumb Americans (was INTERNATIONALIZATION: JAPAN, FAR EAST)
- Message-ID: <1992Dec31.165126.22124@osf.org>
- Keywords: Han Kanji Katakana Hirugana ISO10646 Unicode Codepages
- Sender: news@osf.org (USENET News System)
- Organization: Open Software Foundation
- References: <8490@charon.cwi.nl> <1992Dec31.110735.17551@infodev.cam.ac.uk> <C04py2.8C5@world.std.com>
- Date: Thu, 31 Dec 1992 16:51:26 GMT
- Lines: 22
-
-
- Regarding Peter's observation that dumb software incorrectly expands
- "St. Botolph St." as "Saint Botolph Saint", I should point out that
- the earlier drafts of POSIX.2 addressed this problem by supporting a
- syntax in the LC_COLLATE category for context-sensitive symbol
- substitution. This appears to have been withdrawn from the final
- drafts (perhaps due to a belief that it was not arbitrarily powerful).
-
- Still, in OSF/1, we implemented the 'substitute' construct. This
- permits BRE's in the string expressions so that you can say something
- like:
-
- substitute "\(.*\)St\.$" with "\1Street"
-
- substutute "St\([e]\)\. \(.*\)" with "Saint\1 \2"
-
- and (if I haven't made some stupid mistake with the syntax) get
- strcoll() to do the right things. [The second case even handles
- female saints correctly!]
-
- -- Al Lehotsky
-
-