RE: Free Form RPG Opinions Wanted

["Joe Pluta" <joepluta@xxxxxxxxxxxxxxxxx>]

> From: Douglas Handy
>
> The point was that it was not as simple as reusing the EXACT SAME parsing
> already in the compiler, as you suggested.

A statement oriented parser first reads and builds the entire line from
source following its standard continuation rules, then processes it.  In
both cases (the extended factor 2 of my technique and the semi-colon
terminated statement of free-form) the factor 2 section (the part not
including the opcode) would be EXACTLY THE SAME, and so could follow EXACTLY
THE SAME parsing rules.

Now, if Hans were to say that he processes the statement at the same time he
parses it, then I'd say there's a slight issue.  It could be overcome,
however, and I'd also raise an eyebrow at his design, because it would
definitely not be very flexible, but that's a different issue.


> I'll let Hans answer this if he cares too, or else just go back
> to the archives
> where he has explained it in the past.   It is not that simple to avoid
> ambiguities.

Hans' argument doesn't apply here, Doug, so you need to make your own
analysis.

See, Hans' ambiguity arises precisely BECAUSE of the free-form syntax.
There is no ambiguity in a fixed column environment, because you only need
to check for an opcode to determine whether a line is a continuation.

That's how the eval statement works.

Joe