#!/usr/bin/env python3
"""
Docling PDF Converter for Canadian Building Codes
PDF → MD + JSON (with metadata for BYOD)
"""
import os
import sys
import json
import hashlib
import argparse
from pathlib import Path
from datetime import datetime
from docling.document_converter import DocumentConverter
def calculate_md5(file_path: str) -> str:
"""Calculate MD5 hash of a file."""
hash_md5 = hashlib.md5()
with open(file_path, "rb") as f:
for chunk in iter(lambda: f.read(4096), b""):
hash_md5.update(chunk)
return hash_md5.hexdigest()
def get_pdf_metadata(pdf_path: str) -> dict:
"""Extract PDF metadata for version verification."""
import fitz # PyMuPDF
doc = fitz.open(pdf_path)
metadata = {
"filename": Path(pdf_path).name,
"md5": calculate_md5(pdf_path),
"page_count": len(doc),
"file_size_bytes": os.path.getsize(pdf_path),
"modified_date": datetime.fromtimestamp(
os.path.getmtime(pdf_path)
).isoformat(),
}
# Get PDF internal metadata if available
pdf_meta = doc.metadata
if pdf_meta:
if pdf_meta.get("title"):
metadata["pdf_title"] = pdf_meta["title"]
if pdf_meta.get("author"):
metadata["pdf_author"] = pdf_meta["author"]
if pdf_meta.get("creationDate"):
metadata["pdf_creation_date"] = pdf_meta["creationDate"]
doc.close()
return metadata
def convert_pdf(pdf_path: str, output_dir: str = None) -> dict:
"""
Convert PDF to Markdown + JSON using Docling.
Args:
pdf_path: Path to the PDF file
output_dir: Output directory (default: docling_output/<pdf_name>/)
Returns:
dict with paths to output files and metadata
"""
pdf_path = Path(pdf_path).resolve()
if not pdf_path.exists():
raise FileNotFoundError(f"PDF not found: {pdf_path}")
# Determine output directory
if output_dir is None:
base_name = pdf_path.stem.replace(" ", "_").lower()
output_dir = Path("docling_output") / base_name
else:
output_dir = Path(output_dir)
output_dir.mkdir(parents=True, exist_ok=True)
print(f"Converting: {pdf_path.name}")
print(f"Output dir: {output_dir}")
# Get PDF metadata first (for BYOD verification)
print("Extracting PDF metadata...")
pdf_metadata = get_pdf_metadata(str(pdf_path))
print(f" Pages: {pdf_metadata['page_count']}")
print(f" MD5: {pdf_metadata['md5'][:16]}...")
# Convert with Docling
print("Running Docling conversion...")
converter = DocumentConverter()
result = converter.convert(str(pdf_path))
# Export Markdown
md_content = result.document.export_to_markdown()
md_path = output_dir / f"{pdf_path.stem}.md"
with open(md_path, "w", encoding="utf-8") as f:
f.write(md_content)
print(f" Saved: {md_path.name} ({len(md_content):,} chars)")
# Export JSON (document structure with coordinates)
doc_dict = result.document.export_to_dict()
# Add our metadata
doc_dict["_source_pdf"] = pdf_metadata
doc_dict["_conversion"] = {
"tool": "docling",
"timestamp": datetime.now().isoformat(),
}
json_path = output_dir / f"{pdf_path.stem}.json"
with open(json_path, "w", encoding="utf-8") as f:
json.dump(doc_dict, f, ensure_ascii=False, indent=2)
print(f" Saved: {json_path.name}")
# Save metadata separately for quick access
meta_path = output_dir / f"{pdf_path.stem}_meta.json"
with open(meta_path, "w", encoding="utf-8") as f:
json.dump({
"source_pdf": pdf_metadata,
"conversion": doc_dict["_conversion"],
"md_file": md_path.name,
"json_file": json_path.name,
}, f, ensure_ascii=False, indent=2)
print(f" Saved: {meta_path.name}")
print("Done!")
return {
"pdf_path": str(pdf_path),
"output_dir": str(output_dir),
"md_path": str(md_path),
"json_path": str(json_path),
"meta_path": str(meta_path),
"pdf_metadata": pdf_metadata,
}
def convert_all(source_dir: str = "sources", output_base: str = "docling_output"):
"""Convert all PDFs in a directory."""
source_dir = Path(source_dir)
pdfs = list(source_dir.glob("*.pdf")) + list(source_dir.glob("*.PDF"))
print(f"Found {len(pdfs)} PDFs in {source_dir}")
print("=" * 50)
results = []
for i, pdf in enumerate(pdfs, 1):
print(f"\n[{i}/{len(pdfs)}] {pdf.name}")
print("-" * 40)
try:
result = convert_pdf(str(pdf), None)
results.append({"status": "success", **result})
except Exception as e:
print(f"ERROR: {e}")
results.append({"status": "error", "pdf": str(pdf), "error": str(e)})
# Save summary
summary_path = Path(output_base) / "conversion_summary.json"
with open(summary_path, "w", encoding="utf-8") as f:
json.dump({
"timestamp": datetime.now().isoformat(),
"total": len(pdfs),
"success": sum(1 for r in results if r["status"] == "success"),
"failed": sum(1 for r in results if r["status"] == "error"),
"results": results,
}, f, ensure_ascii=False, indent=2)
print("\n" + "=" * 50)
print(f"Conversion complete!")
print(f" Success: {sum(1 for r in results if r['status'] == 'success')}")
print(f" Failed: {sum(1 for r in results if r['status'] == 'error')}")
print(f" Summary: {summary_path}")
def main():
parser = argparse.ArgumentParser(
description="Convert Building Code PDFs with Docling"
)
parser.add_argument(
"pdf_path",
nargs="?",
help="Path to a single PDF file (or 'all' to convert all in sources/)"
)
parser.add_argument(
"--output", "-o",
help="Output directory (default: docling_output/<pdf_name>/)"
)
parser.add_argument(
"--source-dir", "-s",
default="sources",
help="Source directory for batch conversion (default: sources/)"
)
args = parser.parse_args()
if args.pdf_path is None or args.pdf_path.lower() == "all":
convert_all(args.source_dir)
else:
convert_pdf(args.pdf_path, args.output)
if __name__ == "__main__":
main()
