Index new documents nightly and log each one in a sheet

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

SHEET_ID, Q = "your-spreadsheet-id", "'your-folder-id' in parents and mimeType='application/vnd.google-apps.document'"
HOST, NS = "https://kb-abc123.svc.aped-1234.pinecone.io", "library"  # 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()

@on_trigger
def run(event):
    creds = Credentials(credentials.connection("google", "work"))
    drive = build("drive", "v3", credentials=creds, cache_discovery=False).files()
    new = drive.list(q=f"{Q} and createdTime > '{state.get('since') or '2026-01-01T00:00:00Z'}'",
                     fields="files(id,name,createdTime)").execute()["files"]
    if not new:
        return {"indexed": 0}
    rows = []
    for f in new:
        text = drive.export(fileId=f["id"], mimeType="text/plain").execute().decode("utf-8", "ignore")
        chunks = [text[i:i + 1500] for i in range(0, len(text), 1500)]
        upsert([{"_id": f"{f['id']}#{n}", "chunk_text": c, "file_name": f["name"]} for n, c in enumerate(chunks)])
        rows.append([f["name"], len(chunks), (event.scheduled_time or "")[:10], f["id"]])
    build("sheets", "v4", credentials=creds, cache_discovery=False).spreadsheets().values().append(
        spreadsheetId=SHEET_ID, range="Ingestion log!A:D", valueInputOption="RAW", body={"values": rows}).execute()
    state.set("since", max(f["createdTime"] for f in new))
    return {"indexed": len(new)}