+2019-12-12 Gary Dismukes <dismukes@adacore.com>
+
+ * sem_ch4.adb, sem_res.adb: Minor reformatting.
+
2019-12-12 Justin Squirek <squirek@adacore.com>
* sem_ch4.adb (Analyze_One_Call): Add condition to check for
-- Due to our current model of controlled type expansion we may
-- have resolved a user call to a non-visible controlled primitive
-- since these inherited subprograms may be generated in the current
- -- scope. This is a side-effect of the need for the expander to be
+ -- scope. This is a side effect of the need for the expander to be
-- able to resolve internally generated calls.
-- Specifically, the issue appears when predefined controlled
Get_First_Interp (N, I, It);
-- If the equality is user-defined, the type of the operands
- -- matches that of the formals. For a predefined operqtor,
+ -- matches that of the formals. For a predefined operator,
-- it is the scope that matters, given that the predefined
-- equality has Any_Type formals. In either case the result
- -- type (most often Booleam) must match the context .The
- -- scope is either that of the type if there is a generated
- -- equality (when there is an equality for the component type)
- -- or else Standard otherwise.
+ -- type (most often Boolean) must match the context. The scope
+ -- is either that of the type, if there is a generated equality
+ -- (when there is an equality for the component type), or else
+ -- Standard otherwise.
while Present (It.Typ) loop
if Etype (It.Nam) = Typ