Differences

This shows you the differences between two versions of the page.

--- language:impl [2023/04/09 22:06] – [Overrides] rajit
+++ language:impl [2026/08/05 15:30] (current) – [Variant overrides] rajit
@@ Line 259: / Line 259: @@
 <code act>
 +{
-    type<param>+ id;   // single ID only
+    type<param1,param2,...,paramN>+ id;   // single ID only
     ...
  }
 </code>
-The ''+'' indicates that this is an extra template parameter that should be added to the existing ones that have already been specified in the instance for ''id''.
+The ''+'' indicates that these are //extra// template parameter(s) that should be added to the existing ones that have already been specified for the instance ''id'' in the original type definition.
+==== Variant overrides ====
+If a process has non-strict template parameters, then variants of the process with different non-strict parameter values may need different overrides. While this is straightforward for simple instances, if an array contains a collection of process variants, then this creates a challenge in declaring an override.
+To support this, consider the process below:
+<code act>
+template<pint W | pint VAL>
+defproc addbuf(chan?(int<W>) L; chan!(int<W>) R)
+{
+   int<W> x;
+   chp {
+      *[ L?x; R!(x+VAL) ]
+   }
+}
+</code>
+If we have different implementations of ''addbuf'' depending on the value of ''V'', we can describe this as follows:
+<code act>
+defproc impl_addbuf <: addbuf()
++{
+     ... any overrides for ports here ...
+}
+{
+    /* the default implementation */
+    ...
+}
+| addbuf<W,4> => {
+   /* different implementation when the non-strict parameter has value 4 */
+}
+| addbuf<W,8> => {
+   /* different implementation when the non-strict parameter has value 8 */
+}
+</code>
+This variant override syntax permits variations of the implementation relation based on the specific values of the non-strict template parameters.