Diario YouTube Transcript Search y Index Pipeline

import requests, os, json
from datetime import datetime

H = {"x-api-key": os.environ["SCAVIO_API_KEY"]}

TOPICS = ["search api integration 2026", "ai agent grounding tools", "serp api tutorial"]

def search_youtube_topics(topic):
    """Search YouTube for a topic and return video metadata."""
    r = requests.post("https://api.scavio.dev/api/v1/search", headers=H,
        json={"platform": "youtube", "query": topic}, timeout=10).json()
    videos = []
    for item in r.get("organic", [])[:5]:
        videos.append({
            "video_id": item.get("link", "").split("v=")[-1] if "v=" in item.get("link", "") else "",
            "title": item.get("title", ""),
            "description": item.get("snippet", ""),
            "link": item.get("link", ""),
            "topic": topic,
            "indexed_at": datetime.utcnow().isoformat()
        })
    return videos

# Collect all videos
all_videos = []
for topic in TOPICS:
    videos = search_youtube_topics(topic)
    all_videos.extend(videos)
    print(f"[YOUTUBE] {topic}: {len(videos)} videos found")

# MongoDB insert (pseudo-code - replace with your pymongo connection)
# from pymongo import MongoClient
# db = MongoClient(os.environ["MONGO_URI"]).transcripts
# db.videos.create_index([("title", "text"), ("description", "text")])
# for v in all_videos:
#     db.videos.update_one({"video_id": v["video_id"]}, {"$set": v}, upsert=True)

print(f"\nTotal videos to index: {len(all_videos)}")
for v in all_videos[:3]:
    print(f"  {v['title'][:80]} | {v['link']}")

const H = {"x-api-key": process.env.SCAVIO_API_KEY, "Content-Type": "application/json"};

const TOPICS = ["search api integration 2026", "ai agent grounding tools", "serp api tutorial"];

async function searchYoutubeTopics(topic) {
  const r = await fetch("https://api.scavio.dev/api/v1/search", {
    method: "POST", headers: H,
    body: JSON.stringify({platform: "youtube", query: topic})
  }).then(r => r.json());
  return (r.organic || []).slice(0, 5).map(item => ({
    videoId: (item.link || "").includes("v=") ? item.link.split("v=").pop() : "",
    title: item.title || "",
    description: item.snippet || "",
    link: item.link || "",
    topic,
    indexedAt: new Date().toISOString()
  }));
}

(async () => {
  const allVideos = [];
  for (const topic of TOPICS) {
    const videos = await searchYoutubeTopics(topic);
    allVideos.push(...videos);
    console.log(`[YOUTUBE] ${topic}: ${videos.length} videos found`);
  }
  // MongoDB insert (pseudo-code - replace with your mongodb connection)
  // const { MongoClient } = require("mongodb");
  // const db = (await MongoClient.connect(process.env.MONGO_URI)).db("transcripts");
  // await db.collection("videos").createIndex({title: "text", description: "text"});
  // for (const v of allVideos) {
  //   await db.collection("videos").updateOne({videoId: v.videoId}, {$set: v}, {upsert: true});
  // }
  console.log(`\nTotal videos to index: ${allVideos.length}`);
  allVideos.slice(0, 3).forEach(v => console.log(`  ${v.title.slice(0, 80)} | ${v.link}`));
})();