In a recent note, Arthur T. said: > Date: Wed, 31 May 2006 19:50:38 -0400 > > Actually, that's restriction of JCL and various > programs which access PDSes. Actual member names can be > (almost?) any characters. I've been in the position of > having to delete a member which had a lower case letter as > part of its name. I've also seen some with other strange > hex characters. > And that, in turn, is a consequence of a fragmented design which does not provide a common routine to check name syntax and abets the idiosyncratic definition of name restrictions by those various programs.
It has always been possible to create almost any name with STOW, and nowadays a considerably extended set with Binder with PARM=CASE(MIXED). Other components should not restrict manipulating such members. As a prime example of such inconsistency, I discovered many years ago that ISPF will allow processing an unusually named member by typing 'S' on the prefix area for that member, but not by "SELECT 'Member'" on the command line. Why not? Concerning the problem of entering some characters from some keyboards, well designed languages provide escaped forms, for example assembler's "X'nn'" construct. In others support could be added; sometimes easiy. Wouldn't it be a boon if the JCL translator predefined SET symbols &X00-&XFF for the code points of the corresponding hexadecimal values? This would facilitate entering those characters with no change in the basic JCL syntax. -- gil -- StorageTek INFORMATION made POWERFUL ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
