IBM does not yet have any eight-character opcodes.  It seems unlikely
that any will be defined soon.  Macro definitions that have such
'long' names are therefore unlikely to conflict with opcode mnemonics.

As I suppose most of you know such construction as

|&macname   setc  'gubbins'             --character value of macro name
|            &macname  . . .
|

or

|&macnames(1) setc 'MCGUFFIN','MCGOFFIN','GUBBINS'
| . . .
|            &macnames(&i) . . .            --character value of macro
name

is already supported by the HLASM.  The value of a character set
symbol may appear in the operation field where a macro's name usually
appears.

This suggests that without much trouble the assembler could be induced
to support such an idiom as

|&macname  setc 'gubbins'              --character value of macro name
|&xmacname setc  C2X(&macname)  --its hexadecimal value using available BIF
|             &xmacname . . .

This could be routinized, even encapsulated, readily to make it less tedious.

It is my uninformed guess that it would also be easy to implement
since BPAM would not find it problematic, but John Ehrman or one of
the Perth Laboratory people who monitor this list may see
insurmountable difficulties in doing so.
--
John Gilmore, Ashland, MA 01721 - USA

Reply via email to