In a recent note, Thomas Berg said: > Date: Fri, 18 Aug 2006 15:43:12 +0200 > > Yes, but at least regarding REXX I haven't found any mentionable > "wartifications" > . > Do You think of any specific points ? > o Absence of construct to pass arguments by reference.
o Absence of facility to enumerate members of a compound variable. o Absence of construct to use expressions as compound tails ... I think those are the top three from the Rexx fora. o Absence of instream data sets. o Absence of facility to perform concurrent ATTCHPGM ... but sometimes "address 'SYSCALL' 'spawn ...'" suffices. o Absence of facility to perform molecular ENQs, a la JCL, in order to avoid deadlocks. o Inability to WAIT for SYSDSN ENQs. o Absence of tape-specific support, such as RETAIN and VOL=REF. I recognize the last several are environmental rather than intrinsic to Rexx, and might be implemented in a function package. But some of them need to be addressed in order to make Rexx a viable alternative to JCL, or in any alternative to JCL. But I still prefer Rexx to JCL, and use it to build JCL when some of the points above preclude an entirely-Rexx solution. -- 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
