fix: register AppError handler, PDF validation, duplicate field crash, Pydantic V2, tests init, Dockerfile PYTHONPATH

Dockerfile

@@ -16,7 +16,7 @@ RUN pip install --no-cache-dir -r requirements.txt
 COPY . .
 
 # Set Python path so imports work correctly
-ENV PYTHONPATH=/app/src
+ENV PYTHONPATH=/app
 
 # Keep container running for interactive use
 CMD ["tail", "-f", "/dev/null"]

api/main.py

@@ -1,7 +1,10 @@
 from fastapi import FastAPI
 from api.routes import templates, forms
+from api.errors.handlers import register_exception_handlers
 
 app = FastAPI()
 
+register_exception_handlers(app)
+
 app.include_router(templates.router)
 app.include_router(forms.router)

api/routes/forms.py

@@ -17,7 +17,10 @@ def fill_form(form: FormFill, db: Session = Depends(get_db)):
     fetched_template = get_template(db, form.template_id)
 
     controller = Controller()
-    path = controller.fill_form(user_input=form.input_text, fields=fetched_template.fields, pdf_form_path=fetched_template.pdf_path)
+    try:
+        path = controller.fill_form(user_input=form.input_text, fields=fetched_template.fields, pdf_form_path=fetched_template.pdf_path)
+    except FileNotFoundError:
+        raise AppError("Template PDF file not found on disk", status_code=404)
 
     submission = FormSubmission(**form.model_dump(), output_pdf_path=path)
     return create_form(db, submission)

api/schemas/forms.py

@@ -1,15 +1,14 @@
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 
 class FormFill(BaseModel):
     template_id: int
     input_text: str
 
 
 class FormFillResponse(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
     id: int
     template_id: int
     input_text: str
-    output_pdf_path: str
-
-    class Config:
-        from_attributes = True
+    output_pdf_path: str

api/schemas/templates.py

@@ -1,15 +1,14 @@
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 
 class TemplateCreate(BaseModel):
     name: str
     pdf_path: str
     fields: dict
 
 class TemplateResponse(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
     id: int
     name: str
     pdf_path: str
-    fields: dict
-
-    class Config:
-        from_attributes = True
+    fields: dict

src/file_manipulator.py

@@ -26,8 +26,7 @@ def fill_form(self, user_input: str, fields: list, pdf_form_path: str):
         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
+            raise FileNotFoundError(f"PDF template not found at {pdf_form_path}")
 
         print("[3] Starting extraction and PDF filling process...")
         try:

src/filler.py

@@ -1,3 +1,4 @@
+import os
 from pdfrw import PdfReader, PdfWriter
 from src.llm import LLM
 from datetime import datetime
@@ -12,12 +13,8 @@ def fill_form(self, pdf_form: str, llm: LLM):
         Fill a PDF form with values from user_input using LLM.
         Fields are filled in the visual order (top-to-bottom, left-to-right).
         """
-        output_pdf = (
-            pdf_form[:-4]
-            + "_"
-            + datetime.now().strftime("%Y%m%d_%H%M%S")
-            + "_filled.pdf"
-        )
+        base, _ = os.path.splitext(pdf_form)
+        output_pdf = base + "_" + datetime.now().strftime("%Y%m%d_%H%M%S") + "_filled.pdf"
 
         # Generate dictionary of answers from your original function
         t2j = llm.main_loop()
@@ -29,13 +26,13 @@ def fill_form(self, pdf_form: str, llm: LLM):
         pdf = PdfReader(pdf_form)
 
         # Loop through pages
+        i = 0
         for page in pdf.pages:
             if page.Annots:
                 sorted_annots = sorted(
                     page.Annots, key=lambda a: (-float(a.Rect[1]), float(a.Rect[0]))
                 )
 
-                i = 0
                 for annot in sorted_annots:
                     if annot.Subtype == "/Widget" and annot.T:
                         if i < len(answers_list):

src/llm.py

@@ -60,13 +60,18 @@ def main_loop(self):
             }
 
             try:
-                response = requests.post(ollama_url, json=payload)
+                response = requests.post(ollama_url, json=payload, timeout=60)
                 response.raise_for_status()
             except requests.exceptions.ConnectionError:
                 raise ConnectionError(
                     f"Could not connect to Ollama at {ollama_url}. "
                     "Please ensure Ollama is running and accessible."
                 )
+            except requests.exceptions.Timeout:
+                raise TimeoutError(
+                    f"Ollama request timed out after 60 seconds at {ollama_url}. "
+                    "The model may be overloaded or unavailable."
+                )
             except requests.exceptions.HTTPError as e:
                 raise RuntimeError(f"Ollama returned an error: {e}")
 
@@ -97,8 +102,11 @@ def add_response_to_json(self, field, value):
         if ";" in value:
             parsed_value = self.handle_plural_values(value)
 
-        if field in self._json.keys():
-            self._json[field].append(parsed_value)
+        if field in self._json:
+            existing = self._json[field]
+            if not isinstance(existing, list):
+                existing = [existing]
+            self._json[field] = existing + [parsed_value]
         else:
             self._json[field] = parsed_value
 

src/main.py

@@ -1,6 +1,7 @@
 import os
-# from backend import Fill  
-from commonforms import prepare_form 
+from typing import Union
+# from backend import Fill
+from commonforms import prepare_form
 from pypdf import PdfReader
 from controller import Controller