Backfilling Cross-Platform Track IDs for a Spotify-Only Library

import fs from "node:fs";

const API_KEY = process.env.SONOVAULT_API_KEY!;
const BASE    = "https://api.sonovault.now/v1";
const BATCH   = 100; // /v1/tracks/resolve max

interface SpotifyRow { isrc: string; spotify_id: string; title: string; }

interface Link   { source: string; external_id: string; url: string; }
interface Result { input: string; status: "matched" | "not_found" | "skipped_no_credits"; links: Link[]; }

async function resolveBatch(isrcs: string[]): Promise<Result[]> {
  const res = await fetch(`${BASE}/tracks/resolve`, {
    method: "POST",
    headers: {
      "x-api-key":     API_KEY,
      "content-type":  "application/json",
    },
    body: JSON.stringify({ input_type: "isrc", items: isrcs }),
  });
  if (!res.ok) throw new Error(`${res.status} ${res.statusText}`);
  const data = await res.json();
  return data.results;
}

// Load Spotify export: isrc,spotify_id,title
const rows: SpotifyRow[] = fs.readFileSync("./spotify.csv", "utf-8")
  .trim().split("\n").slice(1)
  .map(line => {
    const [isrc, spotify_id, title] = line.split(",");
    return { isrc, spotify_id, title };
  });

// Only resolve rows that actually have an ISRC. Stragglers go to fallback.
const resolvable = rows.filter(r => r.isrc);
const missing    = rows.filter(r => !r.isrc);

const out: string[] = ["spotify_id,isrc,title,beatport_id,applemusic_id,tidal_id,discogs_id,musicbrainz_id"];

for (let i = 0; i < resolvable.length; i += BATCH) {
  const chunk   = resolvable.slice(i, i + BATCH);
  const results = await resolveBatch(chunk.map(r => r.isrc));

  for (let j = 0; j < chunk.length; j++) {
    const row = chunk[j];
    const { links = [] } = results[j] ?? {};
    const id = (s: string) => links.find(l => l.source === s)?.external_id ?? "";

    out.push([
      row.spotify_id, row.isrc, row.title,
      id("beatport"), id("applemusic"),
      id("tidal"),    id("discogs"),
      id("musicbrainz"),
    ].join(","));
  }
  console.log(`Resolved ${Math.min(i + BATCH, resolvable.length)} / ${resolvable.length}`);
}

fs.writeFileSync("./library.enriched.csv", out.join("\n"));
console.log(`Wrote ${out.length - 1} resolved rows. ${missing.length} rows have no ISRC — run the fallback next.`);

// For rows with no ISRC: fall back to /v1/tracks/search and adopt
// the canonical ISRC from the top result, then re-resolve.

async function findIsrcByName(artist: string, title: string): Promise<string | null> {
  const qs  = new URLSearchParams({ artist, title, limit: "1" });
  const res = await fetch(`${BASE}/tracks/search?${qs}`, {
    headers: { "x-api-key": API_KEY },
  });
  if (!res.ok) return null;
  const { results } = await res.json();
  return results?.[0]?.isrc ?? null;
}

const recovered: { spotify_id: string; isrc: string }[] = [];

for (const row of missing) {
  // row.title is "Track — Artist" in this example; split as needed.
  const [title, artist] = row.title.split(" — ");
  const isrc = await findIsrcByName(artist ?? "", title ?? "");
  if (isrc) recovered.push({ spotify_id: row.spotify_id, isrc });
}

console.log(`Recovered ${recovered.length} / ${missing.length} via fuzzy search`);
// Feed the recovered array back through resolveBatch() the same way as the ISRC path.