import os import shutil import time import uuid from pathlib import Path import gradio as gr PERSIST_ROOT = Path(os.getenv("PERSIST_ROOT", "./data")) OUT_DIR = PERSIST_ROOT / "outputs" OUT_DIR.mkdir(parents=True, exist_ok=True) from commonforms import prepare_form # API documented in the repo README def run_commonforms( input_pdf_path: str, model_name: str, keep_existing_fields: bool, use_signature_fields: bool, image_size: int, confidence: float, ) -> tuple[str, str]: """ Runs CommonForms and returns (download_path, status_message). """ work_id = f"{int(time.time())}-{uuid.uuid4().hex[:8]}" work_dir = Path("/tmp") / f"cf-{work_id}" work_dir.mkdir(parents=True, exist_ok=True) src = Path(input_pdf_path) in_path = work_dir / f"input{src.suffix or '.pdf'}" shutil.copy2(src, in_path) out_name = f"fillable-{src.stem}-{work_id}.pdf" out_tmp = work_dir / out_name out_final = OUT_DIR / out_name try: import torch # optional, only for device detect device = "cuda" if torch.cuda.is_available() else "cpu" except Exception: device = "cpu" # Call CommonForms (kwargs mirror the CLI flags in README) prepare_form( in_path, out_tmp, model_or_path=model_name, # "FFDNet-L" or "FFDNet-S" or path to .pt keep_existing_fields=keep_existing_fields, use_signature_fields=use_signature_fields, device=device, # "cpu" | "cuda" | "0" image_size=image_size, # e.g. 1600 confidence=confidence, # e.g. 0.3 ) shutil.move(out_tmp, out_final) return str(out_final), f"✅ Generated {out_name} on {device}" def _process(pdf, model_name, keep_existing_fields, use_signature_fields, image_size, confidence ): if pdf is None: return None, None, "Please upload a PDF." in_path = pdf if isinstance(pdf, str) else getattr(pdf, "name", None) if not in_path or not os.path.exists(in_path): return None, None, "Could not locate uploaded file." try: path, msg = run_commonforms( in_path, model_name, keep_existing_fields, use_signature_fields, image_size, confidence, ) # DownloadButton expects a filepath; File can mirror the same path. return path, path, msg except Exception as e: return None, None, f"❌ Error: {e}" with gr.Blocks(title="CommonForms — PDF → Fillable PDF") as demo: gr.Markdown( """ # CommonForms — PDF → Fillable PDF Upload a PDF, select options, and generate a fillable PDF. [📄 CommonForms Paper](https://arxiv.org/abs/2509.16506) | [🤗 Dataset](https://huggingface.co/datasets/jbarrow/CommonForms) | [🤗 FFDNet-L](https://huggingface.co/jbarrow/FFDNet-L) | [🤗 FFDNet-S](https://huggingface.co/jbarrow/FFDNet-S) | [💾 Code](https://github.com/jbarrow/commonforms) """ ) with gr.Row(): with gr.Column(scale=2): pdf_in = gr.File(label="Upload PDF", file_types=[".pdf"], type="filepath") with gr.Row(): model_name = gr.Dropdown( label="Detector model", choices=["FFDNet-L", "FFDNet-S"], value="FFDNet-L", ) with gr.Row(): keep_existing_fields = gr.Checkbox(label="Keep existing fields", value=False) use_signature_fields = gr.Checkbox(label="Use signature widgets for signatures", value=True) with gr.Row(): image_size = gr.Slider(800, 2048, value=1600, step=32, label="Image size") confidence = gr.Slider(0.05, 0.9, value=0.30, step=0.05, label="Confidence threshold") run_btn = gr.Button("Run CommonForms", variant="primary") status = gr.Textbox(label="Status", interactive=False) with gr.Column(scale=1): download_btn = gr.DownloadButton(label="Download Fillable PDF", value=None) file_out = gr.File(label="Output (preview / alt download)") run_btn.click( _process, inputs=[pdf_in, model_name, keep_existing_fields, use_signature_fields, image_size, confidence ], outputs=[download_btn, file_out, status], queue=True, api_name="run", ) # Limit default concurrency to 1 (safe for ZeroGPU / GPU-bound ops) demo.launch()