On Mon, 1 Dec 2014 18:29:40 +0000, Pommier, Rex wrote:

>Or possibly when the precursor to the binder was developed that long ago, 
>whoever wrote this section figured with a 44 character DSN being the longest 
>available, they padded it with 20 bytes to make sure it was plenty long.  Then 
>when Unix-style path names came along later, whoever added the path capability 
>decided to not increase the length in order to not take a chance at breaking 
>anything?
>
And yet that plausibly putative "whoever" didn't flinch at adding code to do the
wrap, a probably more risky operation.  (Increasing the length alone wouldn't
suffice; wrapping would still be necessary for long paths.)

Might 54 characters sometimes be needed for "Data.Set(Member)"?

Something else I think I'm noticing (but there are uncontrolled variables).  
When
the DDNAME refers to a PDS, as in "INCLUDE SMP00001(member)", the DDNAME
is listed only once in the DATA SET SUMMARY even though there are scores of
INCLUDEs.  When it refers to a UNIX directory, as in "INCLUDE SYS00011(member)",
it is listed once for each of the scores of INCLUDEd members, with the 
aggravation
doubled by the needless (in this case) wrap.  Does anyone else observe this?

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to