feat: Checkboxes and Radio Button Mapping for PDFs (GSoC)

src/filler.py

@@ -39,8 +39,32 @@ def fill_form(self, pdf_form: str, llm: LLM):
                 for annot in sorted_annots:
                     if annot.Subtype == "/Widget" and annot.T:
                         if i < len(answers_list):
-                            annot.V = f"{answers_list[i]}"
-                            annot.AP = None
+                            answer = answers_list[i]
+
+                            # Check if the field type is a Button (Checkbox/Radio)
+                            field_type = annot.FT if annot.FT else (annot.Parent.FT if annot.Parent else None)
+                            if str(field_type) == "/Btn":
+                                is_truthy = str(answer).lower() in ["yes", "true", "1", "x", "on"]
+
+                                # Find the 'ON' state from the appearance dictionary
+                                on_state = "/Yes" # Default assumption
+                                if annot.AP and annot.AP.N:
+                                    keys = [k for k in annot.AP.N.keys() if k != "/Off"]
+                                    if keys:
+                                        on_state = keys[0]
+
+                                if is_truthy:
+                                    from pdfrw import PdfName
+                                    annot.V = PdfName(on_state.strip("/"))
+                                    annot.AS = PdfName(on_state.strip("/"))
+                                else:
+                                    from pdfrw import PdfName
+                                    annot.V = PdfName("Off")
+                                    annot.AS = PdfName("Off")
+                            else:
+                                annot.V = f"{answer}"
+                                annot.AP = None
+
                             i += 1
                         else:
                             # Stop if we run out of answers

src/main.py

@@ -1,4 +1,6 @@
+from typing import Union
 import os
+
 os.environ["CUDA_VISIBLE_DEVICES"] = ""
 
 # Monkey patch rfdetr to force CPU usage on Mac Silicon / Docker
@@ -12,10 +14,53 @@ def patched_ensure(model_ctx):
 except ImportError:
     pass
 
-from commonforms import prepare_form
+from commonforms import prepare_form 
 from pypdf import PdfReader
 from controller import Controller
 
+def input_fields(num_fields: int):
+    fields = []
+    for i in range(num_fields):
+        field = input(f"Enter description for field {i + 1}: ")
+        fields.append(field)
+    return fields
+
+def run_pdf_fill_process(user_input: str, definitions: list, pdf_form_path: Union[str, os.PathLike]):
+    """
+    This function is called by the frontend server.
+    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.")
+    print(f"[2] PDF template path: {pdf_form_path}")
+
+    # Normalize Path/PathLike to a plain string for downstream code
+    pdf_form_path = os.fspath(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...")
+    try:
+        controller = Controller()
+        output_name = controller.fill_form(
+            user_input=user_input,
+            fields=definitions,
+            pdf_form_path=pdf_form_path
+        )
+
+        print("\n----------------------------------")
+        print(f"✅ Process Complete.")
+        print(f"Output saved to: {output_name}")
+
+        return output_name
+
+    except Exception as e:
+        print(f"An error occurred during PDF generation: {e}")
+        # Re-raise the exception so the frontend can handle it
+        raise e
 if __name__ == "__main__":
     file = "./src/inputs/file.pdf"
     user_input = "Hi. The employee's name is John Doe. His job title is managing director. His department supervisor is Jane Doe. His phone number is 123456. His email is jdoe@ucsc.edu. The signature is <Mamañema>, and the date is 01/02/2005"