Re: Compile error - clarification

[Hans Boldt <boldt@xxxxxxxxxx>]

Jon Paris wrote:
>  >> Also - error refers to the READ statement.
> 
> I no longer have your original message, but are you reading the file name?
> There is currently a limitation in this support such that it is only
> accepted when you read the record format not the file.  I hope that in the
> case of single format files they'll drop this restriction as I find it a
> pain to have to go change all the Reads to use the format name.

Yes, we were quite overly protective in the design for result data structure 
on externally described I/O. For various reasons, including those mentioned 
in this thread, we will likely throw off all restrictions regarding the 
type of result data structure. Perhaps the only rule will be a minimum size 
on the data structure, but then again, there's probably not much reason to 
be even that restrictive. We might even go so far as to allow even a 
stand-alone field as the result.

> 
> There was also an issue at one time which required the DS to be defined
> with LIKEREC (not just an externally described DS) - I think this was
> corrected with a PTF.

Sorry, no PTF yet. Once we fully understand all the implications of relaxing 
the rules, then a PTF might be a possibility. 

Cheers! Hans