song-popularity/get_popularity.js

const http = require("http");
const { load } = require("cheerio");
const Anthropic = require("@anthropic-ai/sdk");
const OpenAI = require("openai");

const OPENAI_API_KEY="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";
const CLIENT_ID = "44e13265ab164d209a8665e7d0ae2237";
const CLIENT_SECRET = "cbd639db2c6b4c01981e8dd0cb00569b";
const REDIRECT_URI = "http://127.0.0.1:8888/callback";

// --- Rate Limiter ---

function createRateLimiter(maxPerSecond) {
  const minInterval = 1000 / maxPerSecond;
  let lastCall = 0;

  return async function throttle() {
    const now = Date.now();
    const elapsed = now - lastCall;
    if (elapsed < minInterval) {
      await new Promise((resolve) => setTimeout(resolve, minInterval - elapsed));
    }
    lastCall = Date.now();
  };
}

const spotifyThrottle = createRateLimiter(2);

async function spotifyFetch(url, options = {}) {
  await spotifyThrottle();
  return fetch(url, options);
}

// --- LLM Providers ---

const EXTRACTION_PROMPT = `Extract all music artist or band names from this web page text. The page is about concerts, festivals, or music venues.

Return ONLY a JSON array of artist names, nothing else. Example: ["Artist 1", "Artist 2"]

If no artist names are found, return an empty array: []

Page text:
`;

async function extractWithAnthropic(text) {
  const client = new Anthropic();
  const message = await client.messages.create({
    model: "claude-haiku-4-5-20251001",
    max_tokens: 1024,
    messages: [{ role: "user", content: EXTRACTION_PROMPT + text }],
  });
  return message.content[0].text.trim();
}

async function extractWithOpenAI(text) {
  const client = new OpenAI({
    apiKey: process.env['OPENAI_API_KEY'] || OPENAI_API_KEY, // This is the default and can be omitted
  });  
  const completion = await client.chat.completions.create({
    model: "gpt-4o-mini",
    max_tokens: 1024,
    messages: [{ role: "user", content: EXTRACTION_PROMPT + text }],
  });
  return completion.choices[0].message.content.trim();
}

const LLM_PROVIDERS = {
  anthropic: extractWithAnthropic,
  openai: extractWithOpenAI,
};

// --- Spotify Auth ---

async function getAccessToken() {
  const response = await fetch("https://accounts.spotify.com/api/token", {
    method: "POST",
    headers: {
      "Content-Type": "application/x-www-form-urlencoded",
      Authorization: "Basic " + btoa(`${CLIENT_ID}:${CLIENT_SECRET}`),
    },
    body: "grant_type=client_credentials",
  });

  const data = await response.json();
  return data.access_token;
}

async function getUserAccessToken() {
  const scopes = "playlist-modify-public playlist-modify-private";
  const authUrl =
    `https://accounts.spotify.com/authorize?response_type=code` +
    `&client_id=${CLIENT_ID}` +
    `&scope=${encodeURIComponent(scopes)}` +
    `&redirect_uri=${encodeURIComponent(REDIRECT_URI)}`;

  console.log("\nOpen this URL in your browser to authorize:\n");
  console.log(authUrl + "\n");

  const code = await new Promise((resolve, reject) => {
    const server = http.createServer((req, res) => {
      const url = new URL(req.url, `http://127.0.0.1:8888`);
      const code = url.searchParams.get("code");
      const error = url.searchParams.get("error");

      if (error) {
        res.writeHead(200, { "Content-Type": "text/html" });
        res.end("<h1>Authorization denied.</h1><p>You can close this tab.</p>");
        server.close();
        reject(new Error(`Authorization denied: ${error}`));
        return;
      }

      if (code) {
        res.writeHead(200, { "Content-Type": "text/html" });
        res.end("<h1>Authorized!</h1><p>You can close this tab.</p>");
        server.close();
        resolve(code);
      }
    });

    server.listen(8888, () => {
      console.log("Waiting for authorization...");
    });
  });

  const response = await fetch("https://accounts.spotify.com/api/token", {
    method: "POST",
    headers: {
      "Content-Type": "application/x-www-form-urlencoded",
      Authorization: "Basic " + btoa(`${CLIENT_ID}:${CLIENT_SECRET}`),
    },
    body: new URLSearchParams({
      grant_type: "authorization_code",
      code,
      redirect_uri: REDIRECT_URI,
    }),
  });

  const data = await response.json();
  if (!response.ok) {
    throw new Error(`Token exchange failed: ${JSON.stringify(data)}`);
  }
  return data.access_token;
}

// --- Spotify API ---

async function searchArtist(token, artistName) {
  const response = await spotifyFetch(
    `https://api.spotify.com/v1/search?q=${encodeURIComponent(artistName)}&type=artist&limit=1`,
    { headers: { Authorization: `Bearer ${token}` } }
  );

  //console.log(`Server response for "${artistName}" on Spotify:`, response.headers, response.status, response.statusText, await response.text());
  const data = await response.json();
  const artist = data.artists.items[0];
  if (!artist) {
    return null;
  }
  return artist;
}

async function getTracksByArtist(token, artistName) {
  const response = await spotifyFetch(
    `https://api.spotify.com/v1/search?q=${encodeURIComponent(`artist:${artistName}`)}&type=track%2Cshow&limit=10`,
    { headers: { Authorization: `Bearer ${token}` } }
  );

  const data = await response.json();
  if (!response.ok) {
    throw new Error(`Spotify API error: ${JSON.stringify(data)}`);
  }
  return data.tracks.items;
}

async function getTopTracks(token, artistName, topN = 3) {
  const artist = await searchArtist(token, artistName);
  if (!artist) {
    return { artist: null, tracks: [] };
  }

  const tracks = await getTracksByArtist(token, artistName);
  const topTracks = tracks
    .filter((t) => t.artists.some((a) => a.id === artist.id))
    .slice(0, topN)
    .map((t) => ({ name: t.name, album: t.album.name, uri: t.uri }));

  return { artist: artist.name, tracks: topTracks };
}

// --- Playlist Creation ---

function playlistNameFromUrl(url) {
  return url
    .replace(/^https?:\/\//, "")
    .replace(/[<>:"\/\\|?*]+/g, "_")
    .replace(/_+/g, "_")
    .replace(/^_|_$/g, "");
}

async function createPlaylist(token, name, trackUris) {
  // Use POST /me/playlists (Feb 2026: /users/{id}/playlists removed)
  const response = await spotifyFetch(
    `https://api.spotify.com/v1/me/playlists`,
    {
      method: "POST",
      headers: {
        Authorization: `Bearer ${token}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        name,
        description: `Generated from ${name}`,
        public: false,
      }),
    }
  );

  const playlist = await response.json();
  if (!response.ok) {
    throw new Error(`Failed to create playlist: ${JSON.stringify(playlist)}`);
  }

  // Use POST /playlists/{id}/items (Feb 2026: /playlists/{id}/tracks removed)
  // Spotify allows max 100 tracks per request
  for (let i = 0; i < trackUris.length; i += 100) {
    const batch = trackUris.slice(i, i + 100);
    const addResponse = await spotifyFetch(
      `https://api.spotify.com/v1/playlists/${playlist.id}/items`,
      {
        method: "POST",
        headers: {
          Authorization: `Bearer ${token}`,
          "Content-Type": "application/json",
        },
        body: JSON.stringify({ uris: batch }),
      }
    );

    if (!addResponse.ok) {
      const err = await addResponse.json();
      throw new Error(`Failed to add tracks: ${JSON.stringify(err)}`);
    }
  }

  return playlist;
}

// --- Web Scraping with LLM ---

async function scrapeArtists(url, provider, log = console.log) {
  log(`Fetching ${url}...`);
  const response = await fetch(url);
  if (!response.ok) {
    throw new Error(`Failed to fetch ${url}: ${response.status}`);
  }

  const html = await response.text();
  const $ = load(html);

  $("script, style, nav, footer, header, noscript, svg").remove();
  const textContent = $("body").text().replace(/\s+/g, " ").trim();

  const maxChars = 20000;
  const truncated =
    textContent.length > maxChars
      ? textContent.slice(0, maxChars) + "..."
      : textContent;

  log(`Extracting artist names with ${provider}...`);

  const extract = LLM_PROVIDERS[provider];
  const responseText = await extract(truncated);

  const jsonString = responseText.replace(/^```(?:json)?\s*/, "").replace(/\s*```$/, "");
  return JSON.parse(jsonString);
}

// --- Main ---

function parseArgs() {
  const args = process.argv.slice(2);
  const parsed = {
    provider: "anthropic",
    url: null,
    artist: null,
    topN: 3,
    createPlaylist: false,
  };

  const llmIndex = args.indexOf("--llm");
  if (llmIndex !== -1) {
    parsed.provider = args[llmIndex + 1];
    if (!LLM_PROVIDERS[parsed.provider]) {
      console.error(
        `Unknown provider "${parsed.provider}". Available: ${Object.keys(LLM_PROVIDERS).join(", ")}`
      );
      process.exit(1);
    }
  }

  const urlIndex = args.indexOf("--url");
  if (urlIndex !== -1) {
    parsed.url = args[urlIndex + 1];
  }

  const topIndex = args.indexOf("--top");
  if (topIndex !== -1) {
    parsed.topN = parseInt(args[topIndex + 1], 10);
  }

  if (args.includes("--create_playlist")) {
    if (!parsed.url) {
      console.error("--create_playlist can only be used with --url");
      process.exit(1);
    }
    parsed.createPlaylist = true;
  }

  // Artist is any arg that isn't a flag or flag value
  const flagIndices = new Set();
  for (const flag of ["--llm", "--url", "--top"]) {
    const idx = args.indexOf(flag);
    if (idx !== -1) {
      flagIndices.add(idx);
      flagIndices.add(idx + 1);
    }
  }
  const flagsWithoutValue = ["--create_playlist"];
  for (const flag of flagsWithoutValue) {
    const idx = args.indexOf(flag);
    if (idx !== -1) flagIndices.add(idx);
  }
  const positional = args.filter((_, i) => !flagIndices.has(i));
  if (positional.length > 0) {
    parsed.artist = positional[0];
  }

  return parsed;
}

async function main() {
  const { provider, url, artist, topN, createPlaylist: shouldCreatePlaylist } = parseArgs();

  if (url) {
    const artistNames = await scrapeArtists(url, provider, console.error);
    if (artistNames.length === 0) {
      console.log(JSON.stringify({ artists: [] }));
      return;
    }

    console.error(`Found ${artistNames.length} artist(s): ${artistNames.join(", ")}`);
    console.error("Looking up top tracks on Spotify...");

    const token = await getAccessToken();
    const artists = [];
    const allTrackUris = [];

    for (const name of artistNames) {
      const result = await getTopTracks(token, name, topN);
      if (result.artist) {
        artists.push(result);
        allTrackUris.push(...result.tracks.map((t) => t.uri));
      }
    }

    const output = { artists };

    if (shouldCreatePlaylist && allTrackUris.length > 0) {
      console.error(`Creating playlist with ${allTrackUris.length} tracks...`);
      const userToken = await getUserAccessToken();
      const playlistName = playlistNameFromUrl(url);
      const playlist = await createPlaylist(userToken, playlistName, allTrackUris);
      output.playlist = {
        name: playlist.name,
        url: playlist.external_urls.spotify,
      };
    }

    console.log(JSON.stringify(output, null, 2));
  } else {
    const artistName = artist || "Radiohead";
    const token = await getAccessToken();
    const result = await getTopTracks(token, artistName, topN);
    console.log(JSON.stringify(result, null, 2));
  }
}

if (require.main === module) {
  main().catch(console.error);
}

module.exports = {
  getAccessToken,
  searchArtist,
  getTracksByArtist,
  getTopTracks,
  scrapeArtists,
  createPlaylist,
  playlistNameFromUrl,
  LLM_PROVIDERS,
};