EvalEvaluator is called only if in root context. Added action and action_context to ExecutionContext

src/core/builtin_helpers.py

@@ -25,7 +25,9 @@ def is_same_success(context, return_values):
 
         if isinstance(ret_val.body, Concept):
             if not ret_val.body.metadata.is_evaluated:
-                with context.push(desc=f"Evaluating concept '{ret_val.body}'") as sub_context:
+                with context.push(BuiltinConcepts.EVALUATE_CONCEPT,
+                                  ret_val.body,
+                                  desc=f"Evaluating concept '{ret_val.body}'") as sub_context:
                     sub_context.local_hints.add(BuiltinConcepts.EVAL_BODY_REQUESTED)
                     evaluated = context.sheerka.evaluate_concept(sub_context, ret_val.body)
                     if evaluated.key != ret_val.body.key:
@@ -237,7 +239,7 @@ def parse_unrecognized(context, source, parsers):
     steps = [BuiltinConcepts.BEFORE_PARSING, BuiltinConcepts.PARSING, BuiltinConcepts.AFTER_PARSING]
     sheerka = context.sheerka
 
-    with context.push(desc=f"Parsing unrecognized '{source}'") as sub_context:
+    with context.push(BuiltinConcepts.PARSING, source, desc=f"Parsing unrecognized '{source}'") as sub_context:
         # disable all parsers but the following ones
         sub_context.add_preprocess(BaseParser.PREFIX + "*", enabled=False)
         for parser in parsers:
@@ -322,7 +324,7 @@ def ensure_evaluated(context, concept):
     if concept.metadata.is_evaluated:
         return concept
 
-    with context.push(desc=f"Evaluating concept {concept}") as sub_context:
+    with context.push(BuiltinConcepts.EVALUATE_CONCEPT, concept, desc=f"Evaluating concept {concept}") as sub_context:
         sub_context.local_hints.add(BuiltinConcepts.EVAL_BODY_REQUESTED)
         evaluated = context.sheerka.evaluate_concept(sub_context, concept)
         sub_context.add_values(return_values=evaluated)