Re: Validating input file

jeff wrote:
> On May 13, 11:07 am, "HeyBub" <hey...@[EMAIL PROTECTED]
> wrote:
>> jeff wrote:
>>> I have a input rec coming from a customer that is defined as
>>
>>> FD DISCFILE.
>>> 01 DISC-REC PIC X(129).
>>> 01 DISC-REC2 REDEFINES DISC-REC.
>>> 03 DISC-RECR OCCURS 129 PIC X.
>>
>>> It is a comma-delimited file with 16 fields (15 commas). I need to
>>> check and make sure that each line is only 129 characters and I
>>> assume 1 LF. What would be the best way to accomplish this?
>>
>> First, look at the file with a hex editor to validate your
>> assumptions.
>>
>> The usual purpose of a comma-delimited file is to dispense with
>> imbedded blanks (else the file would have been defined as fixed
>> length in the first place).
>>
>> I suspect the file, if really comma-delimited, will NOT be a uniform
>> 129 bytes.
>>
>> Going back to your question, what will you do if the record is NOT
>> 129 (or 130) bytes?
>>
>> Assuming a variable length record, made up of variable length
>> fields, your code should look like:
>>
>> READ DISCFILE
>> UNSTRING DISC-REC DELIMITED BY ',' INTO
>> FIELD-1
>> FIELD-2
>> FIELD-3
>> ...
>> FIELD-15
>
> Yeah this is the way I was leaning, Parse out to fields and have a
> count

No, no, no. The point I was making is that only under the weirdest cases 
would you even CARE how long the record might be.