feat+refactor(engine): Add expression extractor + more restructuring and cleanup

tests/unit/test_executor.py

@@ -7,7 +7,7 @@
 from pydantic import SecretStr
 
 from tracecat.contexts import RunContext, ctx_role
-from tracecat.dsl.common import create_default_dsl_context
+from tracecat.dsl.common import create_default_execution_context
 from tracecat.dsl.models import ActionStatement, RunActionInput
 from tracecat.executor.service import run_action_from_input, sync_executor_entrypoint
 from tracecat.expressions.expectations import ExpectedField
@@ -118,7 +118,7 @@ async def test_executor_can_run_udf_with_secrets(
                 for_each=None,
                 args={"secret_key_name": "TEST_UDF_SECRET_KEY"},
             ),
-            exec_context=create_default_dsl_context(),
+            exec_context=create_default_execution_context(),
             run_context=mock_run_context,
         )
 
@@ -221,7 +221,7 @@ async def test_executor_can_run_template_action_with_secret(
                 for_each=None,
                 args={"secret_key_name": "TEST_TEMPLATE_SECRET_KEY"},
             ),
-            exec_context=create_default_dsl_context(),
+            exec_context=create_default_execution_context(),
             run_context=mock_run_context,
         )
 
@@ -260,7 +260,7 @@ def test_sync_executor_entrypoint(test_role, mock_run_context):
                     run_if=None,
                     for_each=None,
                 ),
-                exec_context=create_default_dsl_context(),
+                exec_context=create_default_execution_context(),
                 run_context=mock_run_context,
             )
             result = sync_executor_entrypoint(input, test_role)

tests/unit/test_parse.py

@@ -1,5 +1,5 @@
 from tracecat.expressions.functions import eval_jsonpath
-from tracecat.parse import traverse_leaves
+from tracecat.parse import traverse_expressions, traverse_leaves
 
 
 def test_iter_dict_leaves():
@@ -43,3 +43,44 @@ def test_more_iter_dict_leaves():
     obj4 = {"a": 1, "b": [2, 3], "c": "hello"}
     expected4 = [("a", 1), ("b[0]", 2), ("b[1]", 3), ("c", "hello")]
     assert list(traverse_leaves(obj4)) == expected4
+
+
+def test_traverse_expressions():
+    # Test case 1: Single expression in string
+    data = {
+        "test": "Hello, ${{ var.name }}",
+    }
+    assert list(traverse_expressions(data)) == ["var.name"]
+
+    # Test case 2: Multiple expressions in string
+    data = {
+        "test": "This is a ${{ 1 }} or ${{ 2 }}",
+    }
+    assert list(traverse_expressions(data)) == ["1", "2"]
+
+    # Test case 3: Nested expressions in objects and lists
+    data = {
+        "test": "This is a ${{ 1 }} or ${{ 2 }}",
+        "list": [
+            "This is a ${{ 3 }} or ${{ 4 }}",
+            "second",
+            {
+                "test": "This is a ${{ 5 }} or ${{ 6 }}",
+            },
+        ],
+        "data": "${{ 7 }}${{ 8 }}",
+    }
+    assert list(traverse_expressions(data)) == ["1", "2", "3", "4", "5", "6", "7", "8"]
+
+    # Test case 4: No expressions
+    data = {
+        "test": "Hello world",
+        "list": ["no expressions", {"test": 123}],
+    }
+    assert list(traverse_expressions(data)) == []
+
+    # Test case 5: Empty data structures
+    data = {}
+    assert list(traverse_expressions(data)) == []
+    data = {"test": {}, "list": []}
+    assert list(traverse_expressions(data)) == []