Dicken/dickendock
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 1 Wiki Activity

revert: Netflix XLSX entfernt - überlastet Container (CPU 90%, RAM voll). Netflix läuft wieder über TMDb-Näherung

Browse Source
This commit is contained in:
Dicken
2026-06-19 12:01:18 +02:00
parent 3ebf99637d
commit 4cec47b083
3 changed files with 11 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files

103
backend/src/tools/media/routes.js
View File

@@ -2,13 +2,11 @@ const express = require('express');
const router = express.Router();
const db = require('../../db');
const { authenticate, requireAdmin } = require('../../middleware/auth');
const XLSX = require('xlsx');
const TMDB_BASE = 'https://api.themoviedb.org/3';
const XREL_BASE = 'https://api.xrel.to/v2';
const THIS_YEAR = new Date().getFullYear();
const MIN_YEAR = THIS_YEAR - 1;
const NETFLIX_TOP10_URL = 'https://www.netflix.com/tudum/top10/data/all-weeks-countries.xlsx';
// Sprachen die wir NICHT wollen
const BLOCKED_LANGS = new Set(['tr', 'hi', 'te', 'ta', 'ml', 'kn', 'bn', 'mr', 'pa']);
@@ -491,69 +489,12 @@ router.get('/xrel-cached', authenticate, (req, res) => {
// ── Streaming Top 10 pro Anbieter ──────────────────────────────────────────────
// TMDb Watch-Provider-IDs für Deutschland (watch_region=DE)
// ── Netflix offizielle Top-10-Daten (keine Näherung – echte Zahlen von Netflix selbst) ──
// Quelle: https://www.netflix.com/tudum/top10/data/all-weeks-countries.xlsx
// Wird wöchentlich (dienstags) von Netflix aktualisiert → 20h Cache reicht völlig
async function getNetflixTop10(category /* 'Films' | 'TV' */) {
const cacheKey = `netflix-top10:${category}`;
const cached = cacheGet(cacheKey);
if (cached !== undefined) return cached;
const resp = await fetch(NETFLIX_TOP10_URL, {
headers: { 'User-Agent': 'Mozilla/5.0 (compatible; DickenDock/1.0)' },
});
if (!resp.ok) throw new Error(`Netflix-Daten nicht erreichbar (HTTP ${resp.status})`);
const buf = await resp.arrayBuffer();
const wb = XLSX.read(buf, { type: 'array' });
const sheet = wb.Sheets[wb.SheetNames[0]];
const rows = XLSX.utils.sheet_to_json(sheet);
// Spalten (Stand der Netflix-Datei): country_name, country_iso2, category, week, weekly_rank,
// show_title, season_title, cumulative_weeks_in_top10, ...
const deRows = rows.filter(r => r.country_iso2 === 'DE' && r.category === category);
if (!deRows.length) { cacheSet(cacheKey, [], 60*60*1000); return []; }
// Neueste Woche finden
const latestWeek = deRows.reduce((max, r) => (r.week > max ? r.week : max), deRows[0].week);
const top10 = deRows
.filter(r => r.week === latestWeek)
.sort((a, b) => a.weekly_rank - b.weekly_rank)
.slice(0, 10)
.map(r => ({
rank: r.weekly_rank,
title: r.show_title,
seasonTitle: r.season_title || '',
weeksInTop10: r.cumulative_weeks_in_top10 || null,
week: r.week,
}));
cacheSet(cacheKey, top10, 20 * 60 * 60 * 1000); // 20h Cache
return top10;
}
// Netflix-Titel mit TMDb abgleichen (für Poster, Genres, Modal-Infos)
async function matchNetflixTitleToTmdb(title, mediaType, genreMap) {
try {
const searchPath = mediaType === 'tv' ? '/search/tv' : '/search/movie';
const data = await tmdb(searchPath, { query: title });
const hit = (data.results || [])[0];
if (!hit) return null;
if (mediaType === 'tv') {
return {
id: hit.id, media_type: 'tv',
title: hit.name || hit.original_name,
original_title: hit.original_name,
release_date: hit.first_air_date || '',
poster_path: hit.poster_path || '',
genres: (hit.genre_ids||[]).map(gid => genreMap[gid]).filter(Boolean),
vote_average: hit.vote_average,
original_language: hit.original_language,
origin_country: hit.origin_country || [],
};
}
return { ...enrichMovie(hit, genreMap, null, null), media_type: 'movie' };
} catch { return null; }
}
// HINWEIS: Eine frühere Version hat hier versucht die offizielle Netflix-XLSX-Datei
// (komplette Historie seit 2021, alle ~90 Länder) direkt zu laden und zu parsen.
// Das hat den Container überlastet (CPU ~90%, RAM voll, Container kurzzeitig nicht
// erreichbar) – die Datei ist für ein synchrones Parsing im selben Prozess zu groß.
// Netflix läuft daher wie alle anderen Anbieter über die leichtgewichtige
// TMDb-Näherung weiter unten.
const STREAMING_PROVIDERS = [
{ id: 'netflix', label: 'Netflix', tmdbId: '8' },
@@ -593,37 +534,6 @@ router.get('/streaming/:providerId', authenticate, async (req, res) => {
const provider = STREAMING_PROVIDERS.find(p => p.id === req.params.providerId);
if (!provider) return res.status(404).json({ error: 'Unbekannter Anbieter' });
// ── NETFLIX: echte offizielle Daten, KEIN Filter ─────────────────────────
if (provider.id === 'netflix') {
const genreMap = await getGenreMap();
const [nfFilms, nfShows] = await Promise.all([
getNetflixTop10('Films'),
getNetflixTop10('TV'),
]);
const movies = (await Promise.all(
nfFilms.map(async nf => {
const matched = await matchNetflixTitleToTmdb(nf.title, 'movie', genreMap);
return matched ? { ...matched, rank: nf.rank, weeksInTop10: nf.weeksInTop10, netflixTitle: nf.title } : null;
})
)).filter(Boolean);
const tvShows = (await Promise.all(
nfShows.map(async nf => {
const matched = await matchNetflixTitleToTmdb(nf.title, 'tv', genreMap);
return matched ? { ...matched, rank: nf.rank, weeksInTop10: nf.weeksInTop10, netflixTitle: nf.title } : null;
})
)).filter(Boolean);
return res.json({
provider: { id: provider.id, label: provider.label },
movies, tvShows,
isOfficialData: true,
filterInfo: null, // keine Filter – Original Netflix Top 10
});
}
// ── ENDE NETFLIX-Sonderfall ───────────────────────────────────────────────
// Nur Titel der letzten 2 Jahre (aktuelles Jahr - 1)
const minYear = new Date().getFullYear() - 1;
const minDate = `${minYear}-01-01`;
@@ -721,7 +631,6 @@ router.get('/streaming/:providerId', authenticate, async (req, res) => {
res.json({
provider: { id: provider.id, label: provider.label },
movies, tvShows,
isOfficialData: false,
filterInfo: {
minYear,
minVotes: MIN_VOTES,