From: "Dick Hendrickson" <[log in to unmask]>
Sent: Sunday, 14 October 2012 5:58 AM
> On Sat, Oct 13, 2012 at 1:51 PM, Dick Hendrickson
> <[log in to unmask]>wrote:
>
>> I'm still thinking about this.
>>
>> But one thing I'm almost absolutely sure of is that "the statement that
>> initiated the transfer" is there to cover the use of derived type I/O where
>> child i/o statements might be far from the originating statement.
>>
>>
> After rereading 9.11 again I'm now almost absolutely sure that what I said
> above is wrong. "The statement that initiated the transfer" is intended to
> cover an ASYNCHRONOUS READ where the EOF is detected by a subsequent WAIT
> statement.
The WAIT statement didn't initiate the transfer.
In any case, the EOF is not permitted to occur during a successful READ.
|