On Sat, 20 Oct 2012 09:19:24 -0400, John Gilmore wrote:

>COBOL's 88-level machinery is an artefact of lacunæ elsewhere.  In
>particular, boolean variables, while they have made their way into the
>COBOL standard, have not yet made it into IBM COBOL implementations.
> 
IBM's adherence to standards is frequently moderated by an NIH
attitude.  IBM fails to understand that nowadays it is a tail (in
a couple senses) that can no longer wag the dog.


On Sat, 20 Oct 2012 07:32:24 -0500, Mike Schwab wrote:

>Most levels are for breaking an area of memory into various fields.
>Not an 88 level.
> 
Ah!  I quite misunderstood, taking '88' to be a cardinal number
(not Jesuit).

>DATA DIVISION.
>...
>10 FIELD-NAME PIC X(5).
>    88  FIELD-NAME-TRUE VALUE 'TRUE ';.
>    88  FIELD-NAME-FALSE VALUE 'FALSE'.
>...
>PROCEDURE DIVISION
>...
>IF FIELD-NAME-TRUE THEN
>
>is equivalent to
>IF FIELD-NAME = 'TRUE ' THEN
>
So it's almost but not quite entirely unlike the use of "enum"
in other languages.  But still:

>On Fri, Oct 19, 2012 at 6:18 PM, Paul Gilmartin wrote:
>>>
>> ...  How does this (help to) provide the
>> function of strcasecmp()?  (Perhaps a schematic example?)

Thanks,
gil

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

Reply via email to