In Algol 68, is it legal to say this?
proc even = (int x) bool: (x = 0 | true | odd(x – 1));
proc odd = (int x) bool: (x = 0 | false | even(x – 1));
Forward references in structures are used in the Revised Report (10.3.5), but this works because, I think, the syntax “expands” the right-hand si...
@LeoB. I'm more interested in whether and how it's handled by the standard's VW-grammar than in whether the feature was implemented or how to achieve it.
Still, the answer is in the positive. See 4.2.2.3 of the Informal Introduction. A question formulated as "How were the calls to yet undeclared procedures handled in Algol-68 VW-grammar?" posted to CS.SE may be in order.
Absolutely the code as shown was legal Algol 68, but how the grammar permits that is a question I can't answer. I never really got to grips with it in any serious way.
I ran the code from the question (after converting to uppercase stropping) with the online genie compiler.
Upvoted for ALGOL 68 question and would have upvoted several more times for a W-grammar question! (Though that might properly go elsewhere, perhaps CS stack?)
(Unfortunately in "Grammars For Programming Languages" (Cleaveland, Uzgalis, 1977), which is about the only thing I know of that really describes w-grammars, the sample language ASPLE they use to given an example of a full static semantics of a language requires declare before use.)
@davidbak Frank Pagan's Formal specification of programming languages is another source, and it purports to use a w-grammar to describe a language that allows forward references, but I don't understand how.
@texdr.aft - great! I'll look for it! by the way, do you have his "A Practical Guide To ALGOL 68"? if so look at middle of page 101 ... I'll post a (quasi-)answer soon.
@texdr.aft - no, what I (elliptically) meant was: I've had that book in my library for decades - but don't think I've ever cracked it open. Now I will! Thanks much for pointing me to it!
Discussion on question by texdr.aft: …
Discussion on question by texdr.aft: …