From greger@iuk Mon Apr 22 17:27:16 1991 Received: from mcsun.EU.net by dkuug.dk via EUnet with SMTP (5.64+/8+bit/IDA-1.2.8) id AA21564; Mon, 22 Apr 91 17:27:16 +0200 Received: from ism.isc.com by mcsun.EU.net with SMTP; id AA18594 (5.65a/CWI-2.83); Mon, 22 Apr 91 13:23:11 +0200 Received: by ism.isc.com (Sendmail5.65/1.35) id AA18514; Mon, 22 Apr 91 05:23:12 -0700 Received: from friherr by iuk.isc.com (5.61/smail2.2/11-14-88) id AA01450; Mon, 22 Apr 91 09:34:10 GMT Received: by (5.61/1.35/jcb-s) id AA00296; Mon, 22 Apr 91 11:10:58 GMT Date: Mon, 22 Apr 91 11:10:58 GMT Message-Id: <9104221110.AA00296@> To: erik@sra.co.jp Cc: wg15rin@dkuug.dk From: greger@ism.isc.com ("greger@ism.isc.com (Greger Leijonhufvud, ISC, High Wycombe, U.K.)") Subject: Re: (wg15rin 99) symbolic names in localedef X-Charset: ASCII X-Char-Esc: 29 In reply to your message of Mon Apr 22 08:27:04 1991 ------- >Dear wg15rinners: >Some of the localedefs that I have seen use instead of >%. One might think that the portable characters don't need to be >surrounded by <>. Is there a reason for using the symbolic names for >absolutely everything? >Erik ------- While it might be true that most of the portable characters will be the same for most implementations, that is not always the case. The POSIX standards have been written so as not to be code set dependent (except in areas of information interchange). We certainly expect most POSIX systems to be ASCII based, but at least two major vendors (IBM and Unisys comes to mind) which have a vested interest in EBCDIC. In addition, we might in the future see systems utilizing e.g. Unicode, 10646 or even X212. Consequently, it would be prudent to use the symbolic names also for these. (It means that you can port an "ASCII" localedef source or charmap to e.g. EBCDIC rather easily). When you use a character in "itself", localedef can assume that the encoding is the default one; with a symbolic name that can never be a problem. "Cross-compilations" could also be done using the charmap... Another reson is to avoid some of the ambiguities Keld points out, such as the . -greger-