Re: select vs. if-elseif

Mark,
The difference is intentional ...
When you code the form you show below, that generates the equivalent of a "case" statement in Pascal, or a "switch" in C ... or a FORTRAN computed GOTO ...  whereas, if you code the equivalent with:
    if ...    elseif ...    elseif ...        ...    else    endif
that will always generate that many "tests"... and branches ...

A computed goto usually uses a "branch table" so it has an array of addresses or labels, and does a "look-up" or index operation into that array to get the address of where to go, so it is generally much faster than having to perform all of those tests, especially if there are hundreds of "cases"...

However, if you code something like:
    select    when a = b;
        ...    when c = d;        ...
    etc;
        ...    otherwise;        ...
    end-sel;

this format will also generate multiple tests and branches, just like the " if-elseif-elseif-else-endif " scenario.

Hope that helps,
Mark S. Waterbury

On Friday, January 25, 2019, 4:01:55 PM EST, Mark Murphy <jmarkmurphy@xxxxxxxxx> wrote:

Does anyone else find the redundancy between the select block and the
if-elseif-else block a little disconcerting? Seems to me that there should
be a clear distinction between the two. There never can be at this point,
so I tend to use a relatively inconsistent mix of the two in my code based
on what mood I am in. But the select syntax has a potentially game changing
syntax.

select <variable>;
  when <value>;
      ...
  when <value>;
      ...
  when <value>;
      ...
  other;
      ...
endsl;

This syntax reduces redundancy in a significant amount of  code, by
reducing the need to retype code variable names over and over again.
Wouldn't it be nice?