>From [log in to unmask] Fri Oct 30 14:39:18 2015
>
>On 30/10/2015 13:45, Anton Shterenlikht wrote:
>> Sub-clause 9.10.2.26 RECL= specifier in the INQUIRE statement
>> of f2008, j3/10-007r1 has:
>
>statement. In other cases such as sequential access where different
>records vary in length, my reading of the Standard is that it only
>returns the maximum record length, assuming that the system imposes a limit.
Thanks Clive
The exact phrasing is:
*quote*
...the RECL= specifier is assigned the value ... for direct access,
or the value of the maximum record length of a connection for
sequential access.
*end quote*
There is nothing about cases where OS does not impose a limit.
My understanding of -1 was similar to yours. However, should not
the standard prescribe a specific value for such cases?
Or say it is processor dependent?
Also, the wording above can be understood as the
maximum record length of all records in the file,
which would suggest some sorting of the file records,
hence potentially impacting performance substantially.
Whereas your interpretation is about the OS limit
for the max. record length. Perhaps the wording should
be made more precise?
Anton
|