Index new documents into Pinecone nightly and email what changed

import base64, json
from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build
from clickruntime import on_trigger, credentials, http, state

ME, Q = "me@example.com", "'your-folder-id' in parents and trashed=false and mimeType='application/vnd.google-apps.document'"
HOST, NS = "https://kb-abc123.svc.aped-1234.pinecone.io", "company"  # your index host and namespace
def upsert(records):
    http.post(f"{HOST}/records/namespaces/{NS}/upsert", data="\n".join(json.dumps(r) for r in records[:96]), headers={
        "Api-Key": credentials.secret("PINECONE_API_KEY"), "X-Pinecone-API-Version": "2025-04", "Content-Type": "application/x-ndjson"}).raise_for_status()
def send(to, subject, body):
    raw = f"To: {to}\r\nSubject: {subject}\r\nContent-Type: text/plain; charset=utf-8\r\n\r\n{body}"
    build("gmail", "v1", credentials=Credentials(credentials.connection("google", "work")), cache_discovery=False).users() \
        .messages().send(userId="me", body={"raw": base64.urlsafe_b64encode(raw.encode()).decode()}).execute()

@on_trigger
def run(event):
    drive = build("drive", "v3", credentials=Credentials(credentials.connection("google", "work")), cache_discovery=False).files()
    files, known = drive.list(q=Q, fields="files(id,name,modifiedTime)", pageSize=1000).execute()["files"], state.get("known") or {}
    changed = [f for f in files if known.get(f["id"]) != f["modifiedTime"]]
    for f in changed:
        text = drive.export(fileId=f["id"], mimeType="text/plain").execute().decode("utf-8", "ignore")
        upsert([{"_id": f"{f['id']}#{n}", "chunk_text": text[i:i + 1500], "file_id": f["id"], "file_name": f["name"]}
                for n, i in enumerate(range(0, len(text), 1500))])
    if changed:
        send(ME, f"Knowledge base: {len(changed)} files indexed", "\n".join(
            f"{'Updated' if f['id'] in known else 'Added'}: {f['name']}" for f in changed))
    state.set("known", {f["id"]: f["modifiedTime"] for f in files})
    return {"indexed": len(changed)}