[FEAT]: Implement schema validation for extracted data and integrate into controller pipeline with unit tests

src/controller.py

@@ -1,11 +1,17 @@
+from src.validation import validate_extracted_data
 from src.file_manipulator import FileManipulator
 
 class Controller:
     def __init__(self):
         self.file_manipulator = FileManipulator()
 
     def fill_form(self, user_input: str, fields: list, pdf_form_path: str):
-        return self.file_manipulator.fill_form(user_input, fields, pdf_form_path)
+    data = self.file_manipulator.fill_form(user_input, fields, pdf_form_path)
+
+    if not validate_extracted_data(data):
+        raise ValueError("Invalid extracted data")
+
+    return data
 
     def create_template(self, pdf_path: str):
-        return self.file_manipulator.create_template(pdf_path)
+        return self.file_manipulator.create_template(pdf_path)

src/validation.py

@@ -0,0 +1,13 @@
+def validate_extracted_data(data: dict) -> bool:
+    """
+    Basic validation for extracted form data.
+    Ensures required fields are present and non-empty.
+    """
+
+    required_fields = ["patient_name", "age", "diagnosis"]
+
+    for field in required_fields:
+        if field not in data or not data[field]:
+            return False
+
+    return True

tests/test_controller.py

@@ -0,0 +1,16 @@
+from src.controller import Controller
+
+
+def test_fill_form_validation_fail(monkeypatch):
+    controller = Controller()
+
+    def mock_fill_form(user_input, fields, pdf_form_path):
+        return {"patient_name": "", "age": 30, "diagnosis": "Flu"}
+
+    monkeypatch.setattr(controller.file_manipulator, "fill_form", mock_fill_form)
+
+    try:
+        controller.fill_form("input", [], "file.pdf")
+        assert False  # Should not reach here
+    except ValueError:
+        assert True

tests/test_validation.py

@@ -0,0 +1,28 @@
+import pytest
+from src.validation import validate_extracted_data
+
+
+def test_valid_data():
+    data = {
+        "patient_name": "John Doe",
+        "age": 30,
+        "diagnosis": "Flu"
+    }
+    assert validate_extracted_data(data) == True
+
+
+def test_missing_field():
+    data = {
+        "patient_name": "John Doe",
+        "age": 30
+    }
+    assert validate_extracted_data(data) == False
+
+
+def test_empty_field():
+    data = {
+        "patient_name": "",
+        "age": 30,
+        "diagnosis": "Flu"
+    }
+    assert validate_extracted_data(data) == False