import json
import re
import subprocess

def get_existing_ids(file_path):
    with open(file_path, "r", encoding="utf-8") as f:
        content = f.read()
    ids = re.findall(r"- \*\*ID:\*\* `([^`]+)`", content)
    return set(ids)

def get_notebook_sources():
    cmd = ["/opt/ai-os/products/ceo/integrations/notebooklm-mcp-cli/.venv/bin/nlm", "source", "list", "a0f8646b-90b1-4a65-bddc-07280aa06a99", "--json"]
    res = subprocess.run(cmd, capture_output=True, text=True)
    if res.returncode != 0:
        print("Error listing sources:", res.stderr)
        return []
    try:
        return json.loads(res.stdout)
    except Exception as e:
        print("JSON parse error:", e)
        return []

existing = get_existing_ids("./content/research/domains/tourism/master_tourism_vn.md")
notebook_sources = get_notebook_sources()

notebook_ids = {s["id"] for s in notebook_sources}
new_ids = notebook_ids - existing

print(f"Existing IDs in file: {len(existing)}")
print(f"IDs in NotebookLM: {len(notebook_ids)}")
print(f"New IDs to sync: {len(new_ids)}")

for s in notebook_sources:
    if s["id"] in new_ids:
        print(f"New source: ID={s['id']}, Title={s.get('title', 'N/A')}")