In general I like the idea of using a system symbol. And for a couple things it will work just fine. It will breakdown when large number of things are being kept track of. There was a thread not long ago about hanging a vendor area off a special ( started out as user TCB ) control block that is registered with IBM.
Rob Schramm On Sat, Sep 30, 2017, 11:37 AM Peter Relson <[email protected]> wrote: > >>Does that symbol substitution service indicate, via a return code, that > it > >>failed to find a definition for the symbol? > >No > > Somewhat incorrect. But you do have to ask for it. And keep in mind that > there could be any number of symbols if you don't control the input. > So "the symbol" is a valid thought only if you know that there is exactly > one symbol in the input. > > In ASASYMBT: > SYMBTWARNNOSUB EQU X'20' When no substitution at all has occurred, * > produce a warning return code. > similarly SYMBT1WARNNOSUB. > > *03* 16 > *03* Meaning: Warning. When WarnNoSub was specified, the > * substitution process encountered no symbols for > * which to substitute. > * The substitution processing completed normally. > *03* Action: None required. > > >I don't like ASASYMBM much either :-) > > Care to provide a helpful comment about in what ways it does not meet your > needs? > You are welcome to bash all you like, but without specifics doing so is > unproductive. > The service is admittedly primitive. It meets our needs. > > >I wonder why would someone pass a symbol table to this service > >to have it substitute values for you? > > Not everything to be substituted for is "system symbol". If you have > additional or even alternate symbols then you'd have to inform the service > what those other symbols are so that it can do the substitution. If you're > wondering why not just do the substitution yourself, I doubt that you'd > bother dealing with all the cases that ASASYMBM handles. > You can also provide this symbol table operand to provide flags to direct > the processing (such as WarnNoSub) even if there are no additional > symbols. > > But I still would like to know why/when/if it is important to a program if > a particular symbol is (or is not) defined. Obviously to someone creating > a string for evaluation it is important to know what symbols are available > to be used. Customers don't seem to have a problem knowing that. > > Peter Relson > z/OS Core Technology Design > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- Rob Schramm ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
