On Mon, 22 Nov 2010 12:39:58 -0600, Mike Schwab wrote:

>Can you do a CL64'initial text' for the definition and have the
>execute definition overlay the initial text as needed?
>
IIRC, the problem arose because the ROUTCDE and DESC must follow
snugly the message of unknown length.  Yes, I could fake it by
coding the expansion of the macro myself; I couldn't do it using
the WTO in SYS1.MACLIB.

>On Mon, Nov 22, 2010 at 12:16 PM, Edward Jaffe wrote:
>> On 11/22/2010 9:58 AM, Paul Gilmartin wrote:
>>>
>>> No, no, no!  I was referring not to the syntax of the macro instruction,
>>> but to the generated argument list to the SVC, which places options after
>>> the variable-length text string, requiring convolutions to supply a
>>> text string of unpredictable length in the MF=E form.
>>
>> An ugly design for sure. Those options were clearly an add-on to the
>> original design. I know that WTO can be a frustrating interface. The doc
>> can't possibly describe all of the nuances. Often, you have to experiment
>> with numerous compiles (and MNOTEs or assembly errors) until you finally get
>> what you want.
>>
Ugh.  I encountered the problem some decades ago when I was trying
to provide a WTO subroutine, written in assembler and callable from
Pascal (not IBM's Pascal; another).  I believe I succeeded, but
not by using the WTO macro.

>> Have you tried using the TEXT= keyword?
>>
That should have worked.  Was it available in days of yore?

Hmmm.  What happens if the programmer codes ROUTCDE= and DESC=
on the MF=L, and supplies (only) TEXT= on the MF=E?

Often, it's unduly difficult to encapsulate Assembler Services
macros as subroutines to be called from HHLs.  In retrospect,
every one should have generated CALL, so the service could be
invoked from other languages.

-- gil

----------------------------------------------------------------------
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

Reply via email to