Re: [Axiom-developer] Nested functions in SPAD

[Gabriel Dos Reis]

address@hidden writes:

| > Waldek Hebisch <address@hidden> writes:
| >
| > | How to correctly define nested function in algebra file?  I changed
| > | beggining of goodCoef (in efstruc.spad.pamphlet) to the following:
| > |
| > |     goodCoef(v, l, s) ==
| > |       -- if k1 is part of k2 we should not express k1 in terms of k2
| > |       -- (othewise we would get infinite recursion)
| > |       -- below we impose a stronger condition: we require
| > |       -- height(k1) to be maximal
| > |       h:NonNegativeInteger := 0
| > |       j:Integer := 0
| > |       transcendental?(k:K):Boolean ==
| > |          is?(k, "log"::SY) => true
| > |          is?(k, "exp"::SY) => true
| > |          is?(k, "tan"::SY) => true
| > |          is?(k, "atan"::SY) => true
| > |          false
| > |       ll := [k for k in l | transcendental?(k)]
| > |
| > | that compiles OK, but at runtime I get:
| > |
| > |    >> System error:
| > |    The function |EFSTRUC;transcendental?| is undefined.
| >
| > Try:
| >     goodCoef(v, l, s) ==
| >       -- if k1 is part of k2 we should not express k1 in terms of k2
| >       -- (othewise we would get infinite recursion)
| >       -- below we impose a stronger condition: we require
| >       -- height(k1) to be maximal
| >       h:NonNegativeInteger := 0
| >       j:Integer := 0
| >       ll := [k for k in l | transcendental?(k)] where
| >          transcendental?(k:K):Boolean ==
| >             is?(k, "log"::SY) => true
| >             is?(k, "exp"::SY) => true
| >             is?(k, "tan"::SY) => true
| >             is?(k, "atan"::SY) => true
| >             false
| >       ll
| >
| > -- Gaby
| 
| Interesting. Both should work.

The second works for me, the first seems not; so what do we do when
the map disagrees with the terrain? Swiss Army says "trust the terrain" :-)

The One Way to resolve this is to fix the interpreter/compiler.

-- Gaby