[Aldor-l] [Axiom-developer] Unions in Spad

[Bill Page]

On 7/9/07, Ralf Hemmecke <ralf at hemmecke.de> wrote:
> ...
> Bill Page wrote:
> > Use quotes like this?
> >
> >  'MachineInteger'
> >
> > That makes in a member of an Enumeration. Maybe that is what Aldor expects?
>
> ...
> But, of course it must fail.
>
>    'MachineInteger'
>
> is nothing else than
>
>    Enumeration(MachineInteger: Type)
>
> but what I need is an element of this enumeration type.
>

Sorry. I should have written MachineInteger$'MachineInteger' I think
that is the correct explicit notation of the element. Or you can
import from 'MachineInteger'.

>
> It seems that MachineInteger is rather special. Look at the following
> session.
>
> Ralf
>
> woodpecker:~/scratch>aldor -gloop
>       AA  L      DDDD      OOO    RRRR
>      A A  L      D   D    O   O   R   R
>     A  A  L      D    D  O     O  R   R
>    AAAAA  L      D    D  O     O  RRRR
>   A    A  L      D   D    O   O   R  R
> A     A  LLLLL  DDDD      OOO    R   R
>
> (c) Numerical Algorithms Group Ltd 1995-2001
> Release: Aldor(C) version 1.0.3 for LINUX(glibc2.3) (debug version)
> Type "#int help" for more details.
> %1 >> #include "aldor"
>                                             Comp: 190 msec, Interp: 50 msec
> %2 >> #include "aldorinterp"
>                                             Comp: 80 msec, Interp: 0 msec
> %3 >> MyDom: with == add;
> Defined MyDom @
>   with
>      ==  add ()
>
>                                             Comp: 10 msec, Interp: 0 msec
> %4 >> MyDom
>    () @
>   with
>      ==  add ()
>
>                                             Comp: 0 msec, Interp: 0 msec
> %5 >> import from 'MyDom'
>                                             Comp: 0 msec, Interp: 0 msec
> %6 >> MyDom
>        ^
> [L6 C1] #1 (Error) Have determined 2 possible types for the expression.
>         Meaning 1:
>                   with
>                      ==  add ()
>         ...
>         Meaning 2: Enumeration(MyDom: Type)
>

This makes sense because of the import, right? Aldor cannot tell from
the context which one you mean.

> %7 >> MachineInteger
>    () @ Join(
> PrimitiveType with
>          0: %
>          1: %
>          -: (%, %) -> %
>          bytes: %
>          coerce: SInt -> %
>          coerce: % -> SInt
>          integer: Literal -> %
>          min: %
>          max: %
>          odd?: % -> Boolean
>          zero?: % -> Boolean
>      ==  add ()
> ,
> Join(CopyableType, IntegerType) with
>          mod+: (%, %, %) -> %
>          mod-: (%, %, %) -> %
>          mod*: (%, %, %) -> %
>          mod/: (%, %, %) -> %
>          mod^: (%, %, %) -> %
>          modInverse: (%, %) -> %
>          export to IntegerSegment(%)
>      ==  add ()
> ,
>   with
>      ==  add ()
> )
>                                             Comp: 0 msec, Interp: 0 msec
> %8 >> import from 'MachineInteger'
>                                             Comp: 10 msec, Interp: 0 msec
> %9 >> MachineInteger
>    () @ Join(
> PrimitiveType with
>          0: %
>          1: %
>          -: (%, %) -> %
>          bytes: %
>          coerce: SInt -> %
>          coerce: % -> SInt
>          integer: Literal -> %
>          min: %
>          max: %
>          odd?: % -> Boolean
>          zero?: % -> Boolean
>      ==  add ()
> ,
> Join(CopyableType, IntegerType) with
>          mod+: (%, %, %) -> %
>          mod-: (%, %, %) -> %
>          mod*: (%, %, %) -> %
>          mod/: (%, %, %) -> %
>          mod^: (%, %, %) -> %
>          modInverse: (%, %) -> %
>          export to IntegerSegment(%)
>      ==  add ()
> ,
>   with
>      ==  add ()
> )
>                                             Comp: 0 msec, Interp: 0 msec
>

Interesting. I would have expected the same error as above.

>
> But try to compile
>
> ---BEGIN aaa.as
> #include "aldor"
> #include "aldorio"
> macro I == MachineInteger;
> M1: with == add;
> main(): () == {
>         U == Union(M1, String);
>         import from String;
>         import from U;
>         import from 'M1';
>         s: String := "I am a string";
>         u: U := union s;
>         if u case M1 then {
>                 stdout << "Integer: " << newline;
>         } else {
>                 stdout << "String: " << newline;
>         }
> }
>
> main();
> ---END aaa.as
>
> woodpecker:~/scratch>aldor -laldor -fx aaa.as
> Compiler bug...Bug: gen0UnionCaseIndex: branch not in union
>
> Cool!
>

Cool? That is the same error that I get if I write

 U == Union(Integer, String);
 u:U := ...
 if u case  Integer$'Integer' then ...

http://wiki.axiom-developer.org/SandBoxAldorTesting#msg20070709122631-0500@wiki.axiom-developer.org

Any, I think you were right the first. It was the intention of the
Aldor developers that an untagged Union should be possible. It just
seems that there is an error in the compiler that currently prevents
this.

Regards,
Bill Page.