Add requires_review flag to detect incomplete LLM extraction

README.md

@@ -55,3 +55,34 @@ __Contributors:__
 - Vincent Harkins (@vharkins1)
 - Marc Vergés (@marcvergees) 
 - Jan Sans
+
+
+## Local Development Setup (Beginner Friendly)
+
+1. Clone your fork and enter project folder:
+
+ - git clone <your-fork-url>
+   cd FireForm (Terminal)
+
+2. Create virtual environment:
+
+ - python3 -m venv venv
+   source venv/bin/activate 
+
+3. Install dependencies:
+
+4. Initialize database tables:
+
+5. Run backend server:
+
+6. Open Swagger UI in browser: (http://127.0.0.1:8000/docs)
+
+### Common Errors
+
+- `sqlite3.OperationalError: no such table`
+→ Run database initialization step.
+
+- `Could not connect to Ollama`
+→ Ensure Ollama server is running locally.
+
+

api/db/models.py

@@ -15,4 +15,5 @@ class FormSubmission(SQLModel, table=True):
     template_id: int
     input_text: str
     output_pdf_path: str
-    created_at: datetime = Field(default_factory=datetime.utcnow)
+    requires_review: bool = False
+    created_at: datetime = Field(default_factory=datetime.utcnow)

api/routes/forms.py

@@ -1,27 +1,23 @@
-from fastapi import APIRouter, Depends
-from sqlmodel import Session
-from api.deps import get_db
-from api.schemas.forms import FormFill, FormFillResponse
-from api.db.repositories import create_form, get_template
-from api.db.models import FormSubmission
-from api.errors.base import AppError
-from src.controller import Controller
+from pydantic import BaseModel, Field, field_validator
 
-router = APIRouter(prefix="/forms", tags=["forms"])
 
+class FormFill(BaseModel):
+    template_id: int
+    input_text: str = Field(..., min_length=1, max_length=50000)
 
-@router.post("/fill", response_model=FormFillResponse)
-def fill_form(form: FormFill, db: Session = Depends(get_db)):
-    fetched_template = get_template(db, form.template_id)
-    if not fetched_template:
-        raise AppError("Template not found", status_code=404)
+    @field_validator("input_text")
+    @classmethod
+    def validate_input_text(cls, v):
+        stripped = v.strip()
+        if not stripped:
+            raise ValueError("Input text cannot be empty or only whitespace")
+        return stripped
 
-    controller = Controller()
-    path = controller.fill_form(
-        user_input=form.input_text,
-        fields=fetched_template.fields,
-        pdf_form_path=fetched_template.pdf_path,
-    )
 
-    submission = FormSubmission(**form.model_dump(), output_pdf_path=path)
-    return create_form(db, submission)
+class FormFillResponse(BaseModel):
+    id: int
+    template_id: int
+    input_text: str
+    output_pdf_path: str
+
+    model_config = {"from_attributes": True}

src/controller.py

@@ -5,7 +5,14 @@ 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)
+        path, review_flag = self.file_manipulator.fill_form(
+        user_input=user_input,
+        fields=fields,
+        pdf_form_path=pdf_form_path
+    )
+        return path, review_flag
+
+
 
     def create_template(self, pdf_path: str):
         return self.file_manipulator.create_template(pdf_path)

src/file_manipulator.py

@@ -1,7 +1,10 @@
 import os
+import logging
 from src.filler import Filler
 from src.llm import LLM
 
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
 class FileManipulator:
     def __init__(self):
@@ -28,24 +31,33 @@ def fill_form(self, user_input: str, fields: list, pdf_form_path: str):
         It receives the raw data, runs the PDF filling logic,
         and returns the path to the newly created file.
         """
-        print("[1] Received request from frontend.")
+        logger.info("[1] Received request from frontend.")
         print(f"[2] PDF template path: {pdf_form_path}")
 
         if not os.path.exists(pdf_form_path):
             print(f"Error: PDF template not found at {pdf_form_path}")
             return None  # Or raise an exception
 
-        print("[3] Starting extraction and PDF filling process...")
+        logger.info("[3] Starting extraction...")
         try:
             self.llm._target_fields = fields
             self.llm._transcript_text = user_input
             output_name = self.filler.fill_form(pdf_form=pdf_form_path, llm=self.llm)
 
+            from src.utils.validation import requires_review
+
+            extracted_data = self.llm.get_data()
+
+            review_flag = requires_review(
+                 extracted_data,
+                  fields.keys()
+         )
+
             print("\n----------------------------------")
             print("✅ Process Complete.")
             print(f"Output saved to: {output_name}")
 
-            return output_name
+            return output_name , review_flag
 
         except Exception as e:
             print(f"An error occurred during PDF generation: {e}")