refactor(workflow): add Jinja2 renderer abstraction for template transform

[tomerqodo]

Benchmark PR from agentic-review-benchmarks#3

[qodo-code-review]

Code Review by Qodo

🐞 Bugs (1) 📘 Rule violations (2) 📎 Requirement gaps (0)

1. Test file exceeds 800 lines 📘 Rule violation ⛯ Reliability

Description

• The updated test module under api/ is now well over the 800-line limit, reducing readability and
  maintainability.
• This violates the repository constraint requiring backend Python files under api/ to remain
  below 800 lines.

Code

api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[R1120-1151]

+        assert result.status == WorkflowNodeExecutionStatus.SUCCEEDED
+        assert result.outputs["output"] == "abc"
+
+    @patch(
+        "core.workflow.nodes.template_transform.template_transform_node.CodeExecutorJinja2TemplateRenderer.render_template"
+    )
+    def test_run_with_escape_filter(self, mock_execute, mock_graph, mock_graph_runtime_state, graph_init_params):
+        """Test _run with escape filter for HTML."""
+        node_data = {
+            "title": "Escape Filter Template",
+            "variables": [{"variable": "html", "value_selector": ["sys", "html_content"]}],
+            "template": "{{ html | escape }}",
+        }
+
+        mock_html = MagicMock()
+        mock_html.to_object.return_value = "<script>alert('xss')</script>"
+
+        mock_graph_runtime_state.variable_pool.get.return_value = mock_html
+        mock_execute.return_value = "&lt;script&gt;alert(&#39;xss&#39;)&lt;/script&gt;"
+
+        node = TemplateTransformNode(
+            id="test_node",
+            config=node_data,
+            graph_init_params=graph_init_params,
+            graph=mock_graph,
+            graph_runtime_state=mock_graph_runtime_state,
+        )
+
+        result = node._run()
+
+        assert result.status == WorkflowNodeExecutionStatus.SUCCEEDED
+        assert "&lt;script&gt;" in result.outputs["output"]

Evidence

The compliance rule requires Python files under api/ to be fewer than 800 lines. The test file
contains code at line 1151, proving the file is at least 1151 lines long.

AGENTS.md
api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[1120-1151]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
`api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py` exceeds the 800-line maximum for Python files under `api/`.

## Issue Context
The file now reaches line ~1151, which violates the repository compliance constraint and makes the tests harder to maintain.

## Fix Focus Areas
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[412-1151]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

2. render_template() may return None 📘 Rule violation ⛯ Reliability

Description

• CodeExecutorJinja2TemplateRenderer.render_template() is annotated to return str but can return
  None when result.get("result") is missing.
• TemplateTransformNode._run() immediately calls len(rendered), which will raise a TypeError
  if rendered is None, causing an unhandled crash instead of a controlled failure result.

Code

api/core/workflow/nodes/template_transform/template_renderer.py[R37-40]

+        rendered = result.get("result")
+        if rendered is not None and not isinstance(rendered, str):
+            raise TemplateRenderError("Template render result must be a string.")
+        return rendered

Evidence

The renderer returns rendered without handling the None case, and the node run path assumes
rendered is always a string and calls len(rendered). This is a missing null/edge-case handling
at a clear failure point.

Rule 3: Generic: Robust Error Handling and Edge Case Management
api/core/workflow/nodes/template_transform/template_renderer.py[37-40]
api/core/workflow/nodes/template_transform/template_transform_node.py[68-70]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
`CodeExecutorJinja2TemplateRenderer.render_template()` can return `None` even though its return type is `str`. This leads to `TemplateTransformNode._run()` calling `len(rendered)` and crashing with `TypeError`.

## Issue Context
`result.get("result")` may return `None` if the executor returns an unexpected payload; this should be handled as a controlled failure with a helpful error.

## Fix Focus Areas
- api/core/workflow/nodes/template_transform/template_renderer.py[29-40]
- api/core/workflow/nodes/template_transform/template_transform_node.py[67-80]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

3. Tests use wrong ctor 🐞 Bug ✓ Correctness

Description

• Newly added unit tests instantiate TemplateTransformNode with graph=..., but
  TemplateTransformNode.__init__ does not accept a graph keyword argument, so these tests will
  raise TypeError and fail CI.
• These tests also pass a raw node_data dict as config, but the base Node initializer requires
  a wrapped config containing id and data; this will raise ValueError("Node ID is required.")
  even if the graph kwarg is removed.

Code

api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[R451-457]

+        node = TemplateTransformNode(
+            id="test_node",
+            config=node_data,
+            graph_init_params=graph_init_params,
+            graph=mock_graph,
+            graph_runtime_state=mock_graph_runtime_state,
+        )

Evidence

The new tests pass graph=mock_graph and a raw node_data dict into TemplateTransformNode. The
production constructor signature has no graph parameter, and the base Node constructor enforces
that config must include an id and data mapping; therefore the test code will error at
construction time.

api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[434-457]
api/core/workflow/nodes/template_transform/template_transform_node.py[26-41]
api/core/workflow/nodes/base/node.py[223-235]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

### Issue description
The unit tests in `template_transform_node_spec.py` instantiate `TemplateTransformNode` with an unsupported `graph` keyword argument and with a `config` dict that is not shaped like the production node config (missing `id`/`data`). This will raise during test execution.

### Issue Context
`TemplateTransformNode` subclasses `Node`, and `Node.__init__` requires `config["id"]` and `config["data"]`.

### Fix Focus Areas
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[434-470]
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[58-66]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

ⓘ The new review experience is currently in Beta. Learn more

[qodo-code-review]

1. Test file exceeds 800 lines 📘 Rule violation ⛯ Reliability

• The updated test module under api/ is now well over the 800-line limit, reducing readability and
  maintainability.
• This violates the repository constraint requiring backend Python files under api/ to remain
  below 800 lines.

Agent prompt

## Issue description
`api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py` exceeds the 800-line maximum for Python files under `api/`.

## Issue Context
The file now reaches line ~1151, which violates the repository compliance constraint and makes the tests harder to maintain.

## Fix Focus Areas
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[412-1151]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

[qodo-code-review]

2. render_template() may return none 📘 Rule violation ⛯ Reliability

• CodeExecutorJinja2TemplateRenderer.render_template() is annotated to return str but can return
  None when result.get("result") is missing.
• TemplateTransformNode._run() immediately calls len(rendered), which will raise a TypeError
  if rendered is None, causing an unhandled crash instead of a controlled failure result.

Agent prompt

## Issue description
`CodeExecutorJinja2TemplateRenderer.render_template()` can return `None` even though its return type is `str`. This leads to `TemplateTransformNode._run()` calling `len(rendered)` and crashing with `TypeError`.

## Issue Context
`result.get("result")` may return `None` if the executor returns an unexpected payload; this should be handled as a controlled failure with a helpful error.

## Fix Focus Areas
- api/core/workflow/nodes/template_transform/template_renderer.py[29-40]
- api/core/workflow/nodes/template_transform/template_transform_node.py[67-80]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

[qodo-code-review]

3. Tests use wrong ctor 🐞 Bug ✓ Correctness

• Newly added unit tests instantiate TemplateTransformNode with graph=..., but
  TemplateTransformNode.__init__ does not accept a graph keyword argument, so these tests will
  raise TypeError and fail CI.
• These tests also pass a raw node_data dict as config, but the base Node initializer requires
  a wrapped config containing id and data; this will raise ValueError("Node ID is required.")
  even if the graph kwarg is removed.

Agent prompt

### Issue description
The unit tests in `template_transform_node_spec.py` instantiate `TemplateTransformNode` with an unsupported `graph` keyword argument and with a `config` dict that is not shaped like the production node config (missing `id`/`data`). This will raise during test execution.

### Issue Context
`TemplateTransformNode` subclasses `Node`, and `Node.__init__` requires `config["id"]` and `config["data"]`.

### Fix Focus Areas
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[434-470]
- api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py[58-66]

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools