Make deezify not pick up garbage albums

src/deezify.js

@@ -1,14 +1,12 @@
 import fs from "fs";
 import path from "path";
-import crypto from "crypto"
+import crypto from "crypto";
 
 const CACHE_DIR = "./deezer_cache";
 fs.mkdirSync(CACHE_DIR, { recursive: true });
 
-const artists = [
+const artists = [];
-];
+const album_links = [];
-const album_links = [
-]
 const BAD_KEYWORDS = [
   " remix",
   "remix ",
@@ -22,7 +20,8 @@ const BAD_KEYWORDS = [
   "anniversary",
   "(remixes)",
   "(remix)",
-  "(acustic)"
+  "(acustic)",
+  "mixes",
 ];
 
 function normalize(str) {
@@ -31,8 +30,9 @@ function normalize(str) {
 
 function hasBadKeyword(title) {
   const t = normalize(title);
-  return BAD_KEYWORDS.some(k => t.includes(k));
+  return BAD_KEYWORDS.some((k) => t.includes(k));
 }
+
 function cachePath(url) {
   const hash = crypto.createHash("sha1").update(url).digest("hex");
   return path.join(CACHE_DIR, `${hash}.json`);
@@ -53,13 +53,9 @@ async function cachedFetch(url) {
   return json;
 }
 
-
 async function getArtistId(name) {
   const q = encodeURIComponent(name);
-  const data = await cachedFetch(
+  const data = await cachedFetch(`https://api.deezer.com/search/artist?q=${q}`);
-    `https://api.deezer.com/search/artist?q=${q}`
-  );
-
   return data?.data?.[0]?.id ?? null;
 }
 
@@ -75,63 +71,84 @@ async function getAllAlbums(artistId) {
 
   return out;
 }
+
 async function getAlbum(albumId) {
   const url = `https://api.deezer.com/album/${albumId}`;
   return await cachedFetch(url);
 }
-function selectBestAlbumVersions(albums) {
-  const byTitle = new Map();
 
-  for (const a of albums) {
+async function selectUniqueReleases(albums) {
-    if (hasBadKeyword(a.title)) continue;
+  const seenTracks = new Set();
+  const finalReleases = [];
 
-    const key = normalize(a.title.replace(/\s*\(.*?\)\s*/g, ""));
+  const albumsOnly = albums.filter(
+    (a) => a.record_type === "album" && !hasBadKeyword(a.title),
+  );
+  const singlesEPs = albums.filter(
+    (a) =>
+      (a.record_type === "single" || a.record_type === "ep") &&
+      !hasBadKeyword(a.title),
+  );
 
-    if (!byTitle.has(key)) {
+  for (const a of albumsOnly) {
-      byTitle.set(key, a);
+    const fullAlbum = await getAlbum(a.id);
-      continue;
+    const tracks = (fullAlbum.tracks?.data ?? []).map((t) =>
+      normalize(t.title),
+    );
+
+    if (tracks.every((t) => seenTracks.has(t))) continue;
+    tracks.forEach((t) => seenTracks.add(t));
+    finalReleases.push(fullAlbum);
   }
 
-    const existing = byTitle.get(key);
+  for (const s of singlesEPs) {
+    const fullSingle = await getAlbum(s.id);
+    const tracks = (fullSingle.tracks?.data ?? []).map((t) =>
+      normalize(t.title),
+    );
 
-    // Prefer explicit
+    if (tracks.every((t) => seenTracks.has(t))) continue;
-    if (!existing.explicit_lyrics && a.explicit_lyrics) {
+    tracks.forEach((t) => seenTracks.add(t));
-      byTitle.set(key, a);
+    finalReleases.push(fullSingle);
-    }
   }
 
-  return [...byTitle.values()];
+  return finalReleases;
 }
 
 (async () => {
   const links = [];
-  const displays = []
+  const displays = [];
-  const album_links_raw = await Promise.all(album_links.map(async z => await getAlbum(z.replace(/[^\d]*(\d*)$/gm, "$1"))));
+  const album_links_raw = await Promise.all(
+    album_links.map(
+      async (z) => await getAlbum(z.replace(/[^\d]*(\d*)$/gm, "$1")),
+    ),
+  );
 
   for (const artist of artists) {
     console.log(`\n=== ${artist} ===`);
 
-    const artistId = typeof artist == "number" ? artist : await getArtistId(artist);
+    const artistId =
+      typeof artist === "number" ? artist : await getArtistId(artist);
     if (!artistId) {
       console.log("Artist not found");
       continue;
     }
 
-    const albums = (await getAllAlbums(artistId));
+    const albums = await getAllAlbums(artistId);
+    const cleanAlbums = await selectUniqueReleases(albums);
 
-    const cleanAlbums = selectBestAlbumVersions(albums);
     let artistLines = [];
     let extraCount = 0;
 
     for (const a of cleanAlbums) {
       console.log(
-        `${a.title} | ${a.explicit_lyrics ? "EXPLICIT" : "CLEAN"} | https://www.deezer.com/en/album/${a.id}`
+        `${a.title} | ${a.explicit_lyrics ? "EXPLICIT" : "CLEAN"} | https://www.deezer.com/en/album/${a.id}`,
       );
 
       links.push(`https://www.deezer.com/en/album/${a.id}`);
       if (artistLines.length < 100) {
         artistLines.push(
-          `+ ${typeof artist == "number" ? `No name (${artist})` : artist} - ${a.title} ${a.explicit_lyrics ? "[E]" : ""}`
+          `+ ${typeof artist === "number" ? `No name (${artist})` : artist} - ${a.title} ${a.explicit_lyrics ? "[E]" : ""}`,
         );
       } else {
         extraCount++;
@@ -144,16 +161,24 @@ function selectBestAlbumVersions(albums) {
 
     displays.push(...artistLines);
   }
-  album_links_raw.forEach(z => {
+
+  album_links_raw.forEach((z) => {
     links.push(`https://www.deezer.com/en/album/${z.id}`);
     displays.push(
-      `+ ${z.artist.name} - ${z.title} ${z.explicit_lyrics ? "[E]" : ""}`
+      `+ ${z.artist.name} - ${z.title} ${z.explicit_lyrics ? "[E]" : ""}`,
     );
-  })
+  });
-  fs.writeFileSync("orpheus_links.txt", links.join(" "))
-  fs.writeFileSync("discord_send.txt", displays.join(" \n").match(/.{1,2000}$/gms).map(z => z.trim()).join("\n\n"+'-'.repeat(15)+"\n\n"))
 
-})().catch(err => {
+  fs.writeFileSync("orpheus_links.txt", links.join(" "));
+  fs.writeFileSync(
+    "discord_send.txt",
+    displays
+      .join(" \n")
+      .match(/.{1,2000}$/gms)
+      .map((z) => z.trim())
+      .join("\n\n" + "-".repeat(15) + "\n\n"),
+  );
+})().catch((err) => {
   console.error(err);
   process.exit(1);
 });