See attached patch.
This one is also fairly straight forward.
Regression tested on x86_64.
OK for mainline and then backport to 16?
Regards,
Jerry
---
In gfc_match_varspec, when parsing component references on an
inferred-type ASSOCIATE name, the parser incorrectly matched the
component name as a type-bound procedure .
For inferred-type ASSOCIATE names the parse-time candidate type may
differ from the final resolved type. If gfc_find_component fails with the
default access check, retry with noaccess=true; the resolution pass
will substitute the correct type.
Assisted by: Claude Sonnet 4.6
PR fortran/125531
gcc/fortran/ChangeLog:
* primary.cc (gfc_match_varspec): Before erroring on a zero-argument
COMPCALL, check for a same-named data component and fall back to the
data-component path. For inferred-type ASSOCIATE names, retry
gfc_find_component with noaccess=true when the normal search fails.
gcc/testsuite/ChangeLog:
* gfortran.dg/associate_infer_program_type.f90: New test.
---From 8d04ddd05a114776389203a4ce92bb42d389f3c8 Mon Sep 17 00:00:00 2001
From: Jerry DeLisle <[email protected]>
Date: Fri, 5 Jun 2026 11:07:07 -0700
Subject: [PATCH] fortran: inferred-type ASSOCIATE name giving spurious
"Expected argument list"
In gfc_match_varspec, when parsing component references on an
inferred-type ASSOCIATE name, the parser incorrectly matched the
component name as a type-bound procedure .
For inferred-type ASSOCIATE names the parse-time candidate type may
differ from the final resolved type. If gfc_find_component fails with the
default access check, retry with noaccess=true; the resolution pass
will substitute the correct type.
Assisted by: Claude Sonnet 4.6
PR fortran/125531
gcc/fortran/ChangeLog:
* primary.cc (gfc_match_varspec): Before erroring on a zero-argument
COMPCALL, check for a same-named data component and fall back to the
data-component path. For inferred-type ASSOCIATE names, retry
gfc_find_component with noaccess=true when the normal search fails.
gcc/testsuite/ChangeLog:
* gfortran.dg/associate_infer_program_type.f90: New test.
---
gcc/fortran/primary.cc | 27 ++++++++++++-
.../associate_infer_program_type.f90 | 39 +++++++++++++++++++
2 files changed, 65 insertions(+), 1 deletion(-)
create mode 100644 gcc/testsuite/gfortran.dg/associate_infer_program_type.f90
diff --git a/gcc/fortran/primary.cc b/gcc/fortran/primary.cc
index 2ca2c4744bb..da517f8394f 100644
--- a/gcc/fortran/primary.cc
+++ b/gcc/fortran/primary.cc
@@ -2715,6 +2715,19 @@ gfc_match_varspec (gfc_expr *primary, int equiv_flag, bool sub_flag,
primary->value.compcall.actual = NULL;
else
{
+ /* Before erroring, check whether there is also a data
+ component with this name. Use noaccess=true so
+ that private components are also found. */
+ if (sym && gfc_find_component (sym, name, true, true, NULL))
+ {
+ /* Restore expr to EXPR_VARIABLE and let the data
+ component path below handle it. */
+ primary->expr_type = EXPR_VARIABLE;
+ gfc_free_actual_arglist (primary->value.compcall.actual);
+ primary->value.compcall.actual = NULL;
+ tbp = NULL;
+ goto try_data_component;
+ }
gfc_error ("Expected argument list at %C");
return MATCH_ERROR;
}
@@ -2723,10 +2736,22 @@ gfc_match_varspec (gfc_expr *primary, int equiv_flag, bool sub_flag,
break;
}
+ try_data_component:
+
previous = component;
if (!inquiry && !intrinsic)
- component = gfc_find_component (sym, name, false, false, &tmp);
+ {
+ component = gfc_find_component (sym, name, false, false, &tmp);
+ /* For inferred-type ASSOCIATE names the parse-time candidate type
+ may not be the final type; a private component in the candidate
+ type may correspond to a public component in the correct type.
+ Accept it tentatively so that resolution can fix up the type. */
+ if (!component && !tbp
+ && primary->symtree && primary->symtree->n.sym->assoc
+ && primary->symtree->n.sym->assoc->inferred_type)
+ component = gfc_find_component (sym, name, true, false, &tmp);
+ }
else
component = NULL;
diff --git a/gcc/testsuite/gfortran.dg/associate_infer_program_type.f90 b/gcc/testsuite/gfortran.dg/associate_infer_program_type.f90
new file mode 100644
index 00000000000..28517013b57
--- /dev/null
+++ b/gcc/testsuite/gfortran.dg/associate_infer_program_type.f90
@@ -0,0 +1,39 @@
+! { dg-do compile }
+! { dg-options "" }
+!
+! PR fortran/125531
+! ASSOCIATE with a contained-function selector, using the result's data
+! components as arguments to another function (also from CONTAINS), where
+! the result type is defined in program scope. This used to fail with
+! "Invalid association target" because:
+!
+program test
+ implicit none
+
+ type :: args_t
+ integer :: order_ = 4
+ integer :: cells_ = 20
+ double precision :: x_min_ = 0d0, x_max_ = 1d0
+ end type
+
+ associate(args => get_args())
+ associate(result => compute(args%order_, args%cells_))
+ print *, result
+ end associate
+ end associate
+
+contains
+
+ function get_args() result(r)
+ type(args_t) :: r
+ r%order_ = 2
+ r%cells_ = 10
+ end function
+
+ function compute(order, cells) result(r)
+ integer, intent(in) :: order, cells
+ integer :: r
+ r = order * cells
+ end function
+
+end program
--
2.54.0
