On Wed, 26 Apr 2017 22:04:18 -0400, Tony Harminc  wrote:
>
>> SVC 99, I'd be less certain [that allocation of a socket to a DDNAME is 
>> unsupported]:
>> o A socket can have a descriptor.
>> o DYNALLOC can allocate to PATH('/dev/fd/'descriptor).  I've done this
>>   with unnamed pipes created after step initiation; never tried with sockets
>
>It's not just the allocation, but the access methods. Are you going to open
>your socket with QSAM? VSAM? EXCP? The  $$$$ Media Manager?
> 
QSAM and BSAM work pretty well with allocated UNIX files.  Surprisingly
well, I'd say:
o For FILEDATA=TEXT The data are broken into records at each newline.
  - For RECFM=V(B) 
    The access method synthesizes RDWs with COUNT=data length+4.
  - For RECFM=F(B) records are padded to LRECL with x'40'.  The
    programmer can not choose a different padding character.
  If a record (plus possible RDW) exceeds LRECL, the access method
  invokes SYNAD.
o For FILEDATA=BINARY:
  - For RECFM=F(B) the data are broken into LRECL-sized records.
    There is no special treatment of the newline character.  The
    last record might be padded if necessary.
  - For RECFM=V(B) the data are broken into LRECL-4-sized records.
    The access method synthesizes RDWs.  The last record may be
    short.

Don't know about VSAM or $$$$ Media Manager.  BPAM is read-only.
I'd bet against EXCP.

Decades ago, before any customer dreamed of Unix System Services,
I was first struggling with BPAM and QSAM in assembler.  I wished that
the access methods had been as compliant with legacy data sets as
they are today with UNIX files:  If the attributes in the DCB disagree
with those in the F1 DSCB, tranform the data on input or output to what
the program says it expects.  QSAM for UNIX files shows that such a
transformation is feasible.  I can write a UNIX file with FILEDATA=TEXT,
RECFM=VB then read it back with FILEDATA=TEXT,RECFM=FB with no
problem.  I must know the rules and accept the very rational behavior.

>There are some DYNALLOC keys missing for things that are inherently batch
>JCLy; DD DDNAME= and of course DD * or DATA, and one or two others.
> 
Hmmm.  If I DYNALLOC with the key for DD *, I wonder where the data
come from.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Reply via email to