home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!VM.CC.LATECH.EDU!SPJRG
- Message-ID: <REXXLIST%92082714224433@UGA.CC.UGA.EDU>
- Newsgroups: comp.lang.rexx
- Date: Thu, 27 Aug 1992 13:10:36 CST
- Sender: REXX Programming discussion list <REXXLIST@UGA.BITNET>
- From: Jim Glover <SPJRG@VM.CC.LATECH.EDU>
- Subject: Re: Blanks, REXX, and portability...
- Comments: To: rexxlist@uga.cc.uga.edu
- Lines: 16
-
- >However, it has been my experience (however limited you might perceive
- >that to be) that the format of the data is seldom constant, so, it seems
- >that *MOST* CMS REXX programmers parse data on the basis of white-space-
- >delimited words and/or the presence of known constants at least as often
- >as they parse data on the basis of hard-coded "column numbers".
-
- I'd like to add my "Amen!" I too am a CMS REXX programmer, and I can
- tell you that I *rarely* parse based on column position. I do make an
- exception sometimes when data will be located in a particular position
- without any delimiting white space at all, or when data will be located
- in a particular position and the preceeding fields may unpredictably
- be blank, thus preventing my knowing how many .'s to include in some-
- thing like PARSE VAR INPUT . . . Stuff_I_Want .
- I'd guess that such cases represent about 1% of what I do in REXX. The
- rest of the time, I'm using white space just like everyone else.
- --Jim
-