[m-rev.] for review: Announce subtypes and coerce.

[Peter Wang]

On Sun, 04 Apr 2021 16:36:10 +1000 Julien Fischer <jfischer at opturion.com> wrote:
> 
> Hi Peter,
> 
> On Thu, 1 Apr 2021, Peter Wang wrote:
> 
> > NEWS:
> >    As above.
> >
> > doc/reference_manual.texi:
> >    Uncomment documentation for coerce.
> >
> > diff --git a/NEWS b/NEWS
> > index 058e00662..ead462b39 100644
> > --- a/NEWS
> > +++ b/NEWS
> > @@ -341,6 +341,9 @@ Changes to the Mercury standard library
> > Changes to the Mercury language
> > -------------------------------
> > 
> > +* The type system now supports subtypes, which work in tandem with
> > +  type conversion expressions ("coerce").
> 
> I thinks this feature warrants a bit more of a description than one
> sentence. In particular, the NEWS file should definitely contain an
> example subtype.

How about:

diff --git a/NEWS b/NEWS
index ead462b39..7bd2a2582 100644
--- a/NEWS
+++ b/NEWS
@@ -342,7 +342,21 @@ Changes to the Mercury language
 -------------------------------

 * The type system now supports subtypes, which work in tandem with
-  type conversion expressions ("coerce").
+  type conversion expressions ("coerce"). For example, the following
+  defines a subtype `real_color` of a discriminated union type `color`:
+
+        :- type color
+            --->    rgb(float, float, float)
+            ;       cmyk(float, float, float, float)
+            ;       named(string).
+
+        :- type real_color =< color
+            --->    rgb(float, float, float)
+            ;       cmyk(float, float, float, float).
+
+  A term of type `real_color` can be converted to a term of type `color`
+  with `coerce(Term)`. A term of type `color` can be converted to a term of
+  type `real_color` if it has an appropriate inst.

 * The compiler can implement tabling only when generating C code.
   When compiling a predicate that has a `pragma memo` specified for it


Peter