localscrapers caching

This commit is contained in:
tapframe 2025-10-04 16:51:51 +05:30
parent 7a64851256
commit 90f99985a0
5 changed files with 1417 additions and 11 deletions

View file

@ -107,6 +107,28 @@ interface UseMetadataReturn {
imdbId: string | null;
scraperStatuses: ScraperStatus[];
activeFetchingScrapers: string[];
clearScraperCache: () => Promise<void>;
invalidateScraperCache: (scraperId: string) => Promise<void>;
invalidateContentCache: (type: string, tmdbId: string, season?: number, episode?: number) => Promise<void>;
getScraperCacheStats: () => Promise<{
local: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
};
global: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
hitRate: number;
};
combined: {
totalEntries: number;
hitRate: number;
};
}>;
}
export const useMetadata = ({ id, type, addonId }: UseMetadataProps): UseMetadataReturn => {
@ -1484,6 +1506,23 @@ export const useMetadata = ({ id, type, addonId }: UseMetadataProps): UseMetadat
};
}, [cleanupStreams]);
// Cache management methods
const clearScraperCache = useCallback(async () => {
await localScraperService.clearScraperCache();
}, []);
const invalidateScraperCache = useCallback(async (scraperId: string) => {
await localScraperService.invalidateScraperCache(scraperId);
}, []);
const invalidateContentCache = useCallback(async (type: string, tmdbId: string, season?: number, episode?: number) => {
await localScraperService.invalidateContentCache(type, tmdbId, season, episode);
}, []);
const getScraperCacheStats = useCallback(async () => {
return await localScraperService.getCacheStats();
}, []);
return {
metadata,
loading,
@ -1517,5 +1556,9 @@ export const useMetadata = ({ id, type, addonId }: UseMetadataProps): UseMetadat
imdbId,
scraperStatuses,
activeFetchingScrapers,
clearScraperCache,
invalidateScraperCache,
invalidateContentCache,
getScraperCacheStats,
};
};

View file

@ -0,0 +1,379 @@
import { localScraperCacheService, CachedScraperResult } from './localScraperCacheService';
import { supabaseGlobalCacheService, GlobalCachedScraperResult } from './supabaseGlobalCacheService';
import { logger } from '../utils/logger';
import { Stream } from '../types/streams';
export interface HybridCacheResult {
validResults: Array<CachedScraperResult | GlobalCachedScraperResult>;
expiredScrapers: string[];
allExpired: boolean;
source: 'local' | 'global' | 'hybrid';
}
export interface HybridCacheStats {
local: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
};
global: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
hitRate: number;
};
combined: {
totalEntries: number;
hitRate: number;
};
}
class HybridCacheService {
private static instance: HybridCacheService;
private readonly ENABLE_GLOBAL_CACHE = true; // Can be made configurable
private readonly FALLBACK_TO_LOCAL = true; // Fallback to local if global fails
private constructor() {}
public static getInstance(): HybridCacheService {
if (!HybridCacheService.instance) {
HybridCacheService.instance = new HybridCacheService();
}
return HybridCacheService.instance;
}
/**
* Get cached results with hybrid approach (global first, then local)
*/
async getCachedResults(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<HybridCacheResult> {
try {
// Try global cache first if enabled
if (this.ENABLE_GLOBAL_CACHE) {
try {
const globalResults = await supabaseGlobalCacheService.getCachedResults(type, tmdbId, season, episode);
if (globalResults.validResults.length > 0) {
logger.log(`[HybridCache] Using global cache: ${globalResults.validResults.length} results`);
return {
...globalResults,
source: 'global'
};
}
} catch (error) {
logger.warn('[HybridCache] Global cache failed, falling back to local:', error);
}
}
// Fallback to local cache
if (this.FALLBACK_TO_LOCAL) {
const localResults = await localScraperCacheService.getCachedResults(type, tmdbId, season, episode);
if (localResults.validResults.length > 0) {
logger.log(`[HybridCache] Using local cache: ${localResults.validResults.length} results`);
return {
...localResults,
source: 'local'
};
}
}
// No valid results found
return {
validResults: [],
expiredScrapers: [],
allExpired: true,
source: 'hybrid'
};
} catch (error) {
logger.error('[HybridCache] Error getting cached results:', error);
return {
validResults: [],
expiredScrapers: [],
allExpired: true,
source: 'hybrid'
};
}
}
/**
* Cache results in both local and global cache
*/
async cacheResults(
type: string,
tmdbId: string,
results: Array<{
scraperId: string;
scraperName: string;
streams: Stream[] | null;
error: Error | null;
}>,
season?: number,
episode?: number
): Promise<void> {
try {
// Cache in local storage first (fastest)
const localPromises = results.map(result =>
localScraperCacheService.cacheScraperResult(
type, tmdbId, result.scraperId, result.scraperName,
result.streams, result.error, season, episode
)
);
await Promise.all(localPromises);
// Cache in global storage (shared across users)
if (this.ENABLE_GLOBAL_CACHE) {
try {
await supabaseGlobalCacheService.cacheResults(type, tmdbId, results, season, episode);
logger.log(`[HybridCache] Cached ${results.length} results in both local and global cache`);
} catch (error) {
logger.warn('[HybridCache] Failed to cache in global storage:', error);
// Local cache succeeded, so we continue
}
}
} catch (error) {
logger.error('[HybridCache] Error caching results:', error);
}
}
/**
* Cache a single scraper result
*/
async cacheScraperResult(
type: string,
tmdbId: string,
scraperId: string,
scraperName: string,
streams: Stream[] | null,
error: Error | null,
season?: number,
episode?: number
): Promise<void> {
await this.cacheResults(type, tmdbId, [{
scraperId,
scraperName,
streams,
error
}], season, episode);
}
/**
* Get list of scrapers that need to be re-run
*/
async getScrapersToRerun(
type: string,
tmdbId: string,
availableScrapers: Array<{ id: string; name: string }>,
season?: number,
episode?: number
): Promise<string[]> {
const { validResults, expiredScrapers } = await this.getCachedResults(type, tmdbId, season, episode);
const validScraperIds = new Set(validResults.map(r => r.scraperId));
const expiredScraperIds = new Set(expiredScrapers);
// Return scrapers that are either expired or not cached
const scrapersToRerun = availableScrapers
.filter(scraper =>
!validScraperIds.has(scraper.id) || expiredScraperIds.has(scraper.id)
)
.map(scraper => scraper.id);
logger.log(`[HybridCache] Scrapers to re-run: ${scrapersToRerun.join(', ')}`);
return scrapersToRerun;
}
/**
* Get all valid cached streams
*/
async getCachedStreams(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<Stream[]> {
const { validResults } = await this.getCachedResults(type, tmdbId, season, episode);
// Flatten all valid streams
const allStreams: Stream[] = [];
for (const result of validResults) {
if (result.success && result.streams) {
allStreams.push(...result.streams);
}
}
return allStreams;
}
/**
* Invalidate cache for specific content
*/
async invalidateContent(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<void> {
try {
// Invalidate both local and global cache
const promises = [
localScraperCacheService.invalidateContent(type, tmdbId, season, episode)
];
if (this.ENABLE_GLOBAL_CACHE) {
promises.push(
supabaseGlobalCacheService.invalidateContent(type, tmdbId, season, episode)
);
}
await Promise.all(promises);
logger.log(`[HybridCache] Invalidated cache for ${type}:${tmdbId}`);
} catch (error) {
logger.error('[HybridCache] Error invalidating cache:', error);
}
}
/**
* Invalidate cache for specific scraper
*/
async invalidateScraper(scraperId: string): Promise<void> {
try {
// Invalidate both local and global cache
const promises = [
localScraperCacheService.invalidateScraper(scraperId)
];
if (this.ENABLE_GLOBAL_CACHE) {
promises.push(
supabaseGlobalCacheService.invalidateScraper(scraperId)
);
}
await Promise.all(promises);
logger.log(`[HybridCache] Invalidated cache for scraper ${scraperId}`);
} catch (error) {
logger.error('[HybridCache] Error invalidating scraper cache:', error);
}
}
/**
* Clear all cached results
*/
async clearAllCache(): Promise<void> {
try {
// Clear both local and global cache
const promises = [
localScraperCacheService.clearAllCache()
];
if (this.ENABLE_GLOBAL_CACHE) {
promises.push(
supabaseGlobalCacheService.clearAllCache()
);
}
await Promise.all(promises);
logger.log('[HybridCache] Cleared all cache (local and global)');
} catch (error) {
logger.error('[HybridCache] Error clearing cache:', error);
}
}
/**
* Get combined cache statistics
*/
async getCacheStats(): Promise<HybridCacheStats> {
try {
const [localStats, globalStats] = await Promise.all([
localScraperCacheService.getCacheStats(),
this.ENABLE_GLOBAL_CACHE ? supabaseGlobalCacheService.getCacheStats() : Promise.resolve({
totalEntries: 0,
totalSize: 0,
oldestEntry: null,
newestEntry: null,
hitRate: 0
})
]);
return {
local: localStats,
global: globalStats,
combined: {
totalEntries: localStats.totalEntries + globalStats.totalEntries,
hitRate: globalStats.hitRate // Global cache hit rate is more meaningful
}
};
} catch (error) {
logger.error('[HybridCache] Error getting cache stats:', error);
return {
local: { totalEntries: 0, totalSize: 0, oldestEntry: null, newestEntry: null },
global: { totalEntries: 0, totalSize: 0, oldestEntry: null, newestEntry: null, hitRate: 0 },
combined: { totalEntries: 0, hitRate: 0 }
};
}
}
/**
* Clean up old entries in both caches
*/
async cleanupOldEntries(): Promise<void> {
try {
const promises = [
localScraperCacheService.clearAllCache() // Local cache handles cleanup automatically
];
if (this.ENABLE_GLOBAL_CACHE) {
promises.push(
supabaseGlobalCacheService.cleanupOldEntries()
);
}
await Promise.all(promises);
logger.log('[HybridCache] Cleaned up old entries');
} catch (error) {
logger.error('[HybridCache] Error cleaning up old entries:', error);
}
}
/**
* Get cache configuration
*/
getConfig(): {
enableGlobalCache: boolean;
fallbackToLocal: boolean;
} {
return {
enableGlobalCache: this.ENABLE_GLOBAL_CACHE,
fallbackToLocal: this.FALLBACK_TO_LOCAL
};
}
/**
* Update cache configuration
*/
updateConfig(config: {
enableGlobalCache?: boolean;
fallbackToLocal?: boolean;
}): void {
if (config.enableGlobalCache !== undefined) {
(this as any).ENABLE_GLOBAL_CACHE = config.enableGlobalCache;
}
if (config.fallbackToLocal !== undefined) {
(this as any).FALLBACK_TO_LOCAL = config.fallbackToLocal;
}
logger.log('[HybridCache] Configuration updated:', this.getConfig());
}
}
export const hybridCacheService = HybridCacheService.getInstance();
export default hybridCacheService;

View file

@ -0,0 +1,425 @@
import AsyncStorage from '@react-native-async-storage/async-storage';
import { logger } from '../utils/logger';
import { Stream } from '../types/streams';
export interface CachedScraperResult {
streams: Stream[];
timestamp: number;
success: boolean;
error?: string;
scraperId: string;
scraperName: string;
}
export interface CachedContentResult {
contentKey: string; // e.g., "movie:123" or "tv:123:1:2"
results: CachedScraperResult[];
timestamp: number;
ttl: number;
}
class LocalScraperCacheService {
private static instance: LocalScraperCacheService;
private readonly CACHE_KEY_PREFIX = 'local-scraper-cache';
private readonly DEFAULT_TTL_MS = 30 * 60 * 1000; // 30 minutes default TTL
private readonly MAX_CACHE_SIZE = 200; // Maximum number of cached content items
private readonly FAILED_RETRY_TTL_MS = 5 * 60 * 1000; // 5 minutes for failed scrapers
private readonly SUCCESS_TTL_MS = 60 * 60 * 1000; // 1 hour for successful scrapers
private constructor() {}
public static getInstance(): LocalScraperCacheService {
if (!LocalScraperCacheService.instance) {
LocalScraperCacheService.instance = new LocalScraperCacheService();
}
return LocalScraperCacheService.instance;
}
/**
* Generate cache key for content
*/
private getContentKey(type: string, tmdbId: string, season?: number, episode?: number): string {
if (season !== undefined && episode !== undefined) {
return `${type}:${tmdbId}:${season}:${episode}`;
}
return `${type}:${tmdbId}`;
}
/**
* Generate AsyncStorage key for cached content
*/
private getStorageKey(contentKey: string): string {
return `${this.CACHE_KEY_PREFIX}:${contentKey}`;
}
/**
* Check if cached result is still valid based on TTL
*/
private isCacheValid(timestamp: number, ttl: number): boolean {
return Date.now() - timestamp < ttl;
}
/**
* Get cached results for content, filtering out expired results
*/
async getCachedResults(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<{
validResults: CachedScraperResult[];
expiredScrapers: string[];
allExpired: boolean;
}> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const storageKey = this.getStorageKey(contentKey);
const cachedData = await AsyncStorage.getItem(storageKey);
if (!cachedData) {
return {
validResults: [],
expiredScrapers: [],
allExpired: true
};
}
const parsed: CachedContentResult = JSON.parse(cachedData);
// Check if the entire cache entry is expired
if (!this.isCacheValid(parsed.timestamp, parsed.ttl)) {
// Remove expired entry
await AsyncStorage.removeItem(storageKey);
return {
validResults: [],
expiredScrapers: parsed.results.map(r => r.scraperId),
allExpired: true
};
}
// Filter valid results and identify expired scrapers
const validResults: CachedScraperResult[] = [];
const expiredScrapers: string[] = [];
for (const result of parsed.results) {
// Use different TTL based on success/failure
const ttl = result.success ? this.SUCCESS_TTL_MS : this.FAILED_RETRY_TTL_MS;
if (this.isCacheValid(result.timestamp, ttl)) {
validResults.push(result);
} else {
expiredScrapers.push(result.scraperId);
}
}
logger.log(`[LocalScraperCache] Retrieved ${validResults.length} valid results, ${expiredScrapers.length} expired scrapers for ${contentKey}`);
return {
validResults,
expiredScrapers,
allExpired: validResults.length === 0
};
} catch (error) {
logger.error('[LocalScraperCache] Error getting cached results:', error);
return {
validResults: [],
expiredScrapers: [],
allExpired: true
};
}
}
/**
* Cache results for specific scrapers
*/
async cacheResults(
type: string,
tmdbId: string,
results: CachedScraperResult[],
season?: number,
episode?: number
): Promise<void> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const storageKey = this.getStorageKey(contentKey);
// Get existing cached data
const existingData = await AsyncStorage.getItem(storageKey);
let cachedContent: CachedContentResult;
if (existingData) {
cachedContent = JSON.parse(existingData);
// Update existing results or add new ones
for (const newResult of results) {
const existingIndex = cachedContent.results.findIndex(r => r.scraperId === newResult.scraperId);
if (existingIndex >= 0) {
// Update existing result
cachedContent.results[existingIndex] = newResult;
} else {
// Add new result
cachedContent.results.push(newResult);
}
}
} else {
// Create new cache entry
cachedContent = {
contentKey,
results,
timestamp: Date.now(),
ttl: this.DEFAULT_TTL_MS
};
}
// Update timestamp
cachedContent.timestamp = Date.now();
// Store updated cache
await AsyncStorage.setItem(storageKey, JSON.stringify(cachedContent));
// Clean up old cache entries if we exceed the limit
await this.cleanupOldEntries();
logger.log(`[LocalScraperCache] Cached ${results.length} results for ${contentKey}`);
} catch (error) {
logger.error('[LocalScraperCache] Error caching results:', error);
}
}
/**
* Cache a single scraper result
*/
async cacheScraperResult(
type: string,
tmdbId: string,
scraperId: string,
scraperName: string,
streams: Stream[] | null,
error: Error | null,
season?: number,
episode?: number
): Promise<void> {
const result: CachedScraperResult = {
streams: streams || [],
timestamp: Date.now(),
success: !error && streams !== null,
error: error?.message,
scraperId,
scraperName
};
await this.cacheResults(type, tmdbId, [result], season, episode);
}
/**
* Get list of scrapers that need to be re-run (expired or failed)
*/
async getScrapersToRerun(
type: string,
tmdbId: string,
availableScrapers: Array<{ id: string; name: string }>,
season?: number,
episode?: number
): Promise<string[]> {
const { validResults, expiredScrapers } = await this.getCachedResults(type, tmdbId, season, episode);
const validScraperIds = new Set(validResults.map(r => r.scraperId));
const expiredScraperIds = new Set(expiredScrapers);
// Return scrapers that are either expired or not cached at all
const scrapersToRerun = availableScrapers
.filter(scraper =>
!validScraperIds.has(scraper.id) || expiredScraperIds.has(scraper.id)
)
.map(scraper => scraper.id);
logger.log(`[LocalScraperCache] Scrapers to re-run: ${scrapersToRerun.join(', ')}`);
return scrapersToRerun;
}
/**
* Get all valid cached streams for content
*/
async getCachedStreams(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<Stream[]> {
const { validResults } = await this.getCachedResults(type, tmdbId, season, episode);
// Flatten all valid streams
const allStreams: Stream[] = [];
for (const result of validResults) {
if (result.success && result.streams) {
allStreams.push(...result.streams);
}
}
return allStreams;
}
/**
* Invalidate cache for specific content
*/
async invalidateContent(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<void> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const storageKey = this.getStorageKey(contentKey);
await AsyncStorage.removeItem(storageKey);
logger.log(`[LocalScraperCache] Invalidated cache for ${contentKey}`);
} catch (error) {
logger.error('[LocalScraperCache] Error invalidating cache:', error);
}
}
/**
* Invalidate cache for specific scraper across all content
*/
async invalidateScraper(scraperId: string): Promise<void> {
try {
const keys = await AsyncStorage.getAllKeys();
const cacheKeys = keys.filter(key => key.startsWith(this.CACHE_KEY_PREFIX));
for (const key of cacheKeys) {
const cachedData = await AsyncStorage.getItem(key);
if (cachedData) {
const parsed: CachedContentResult = JSON.parse(cachedData);
// Remove results from this scraper
parsed.results = parsed.results.filter(r => r.scraperId !== scraperId);
if (parsed.results.length === 0) {
// Remove entire cache entry if no results left
await AsyncStorage.removeItem(key);
} else {
// Update cache with remaining results
await AsyncStorage.setItem(key, JSON.stringify(parsed));
}
}
}
logger.log(`[LocalScraperCache] Invalidated cache for scraper ${scraperId}`);
} catch (error) {
logger.error('[LocalScraperCache] Error invalidating scraper cache:', error);
}
}
/**
* Clear all cached results
*/
async clearAllCache(): Promise<void> {
try {
const keys = await AsyncStorage.getAllKeys();
const cacheKeys = keys.filter(key => key.startsWith(this.CACHE_KEY_PREFIX));
await AsyncStorage.multiRemove(cacheKeys);
logger.log(`[LocalScraperCache] Cleared ${cacheKeys.length} cache entries`);
} catch (error) {
logger.error('[LocalScraperCache] Error clearing cache:', error);
}
}
/**
* Clean up old cache entries to stay within size limit
*/
private async cleanupOldEntries(): Promise<void> {
try {
const keys = await AsyncStorage.getAllKeys();
const cacheKeys = keys.filter(key => key.startsWith(this.CACHE_KEY_PREFIX));
if (cacheKeys.length <= this.MAX_CACHE_SIZE) {
return; // No cleanup needed
}
// Get all cache entries with their timestamps
const entriesWithTimestamps = await Promise.all(
cacheKeys.map(async (key) => {
const data = await AsyncStorage.getItem(key);
if (data) {
const parsed: CachedContentResult = JSON.parse(data);
return { key, timestamp: parsed.timestamp };
}
return { key, timestamp: 0 };
})
);
// Sort by timestamp (oldest first)
entriesWithTimestamps.sort((a, b) => a.timestamp - b.timestamp);
// Remove oldest entries
const entriesToRemove = entriesWithTimestamps.slice(0, cacheKeys.length - this.MAX_CACHE_SIZE);
const keysToRemove = entriesToRemove.map(entry => entry.key);
if (keysToRemove.length > 0) {
await AsyncStorage.multiRemove(keysToRemove);
logger.log(`[LocalScraperCache] Cleaned up ${keysToRemove.length} old cache entries`);
}
} catch (error) {
logger.error('[LocalScraperCache] Error cleaning up cache:', error);
}
}
/**
* Get cache statistics
*/
async getCacheStats(): Promise<{
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
}> {
try {
const keys = await AsyncStorage.getAllKeys();
const cacheKeys = keys.filter(key => key.startsWith(this.CACHE_KEY_PREFIX));
let totalSize = 0;
let oldestTimestamp: number | null = null;
let newestTimestamp: number | null = null;
for (const key of cacheKeys) {
const data = await AsyncStorage.getItem(key);
if (data) {
totalSize += data.length;
const parsed: CachedContentResult = JSON.parse(data);
if (oldestTimestamp === null || parsed.timestamp < oldestTimestamp) {
oldestTimestamp = parsed.timestamp;
}
if (newestTimestamp === null || parsed.timestamp > newestTimestamp) {
newestTimestamp = parsed.timestamp;
}
}
}
return {
totalEntries: cacheKeys.length,
totalSize,
oldestEntry: oldestTimestamp,
newestEntry: newestTimestamp
};
} catch (error) {
logger.error('[LocalScraperCache] Error getting cache stats:', error);
return {
totalEntries: 0,
totalSize: 0,
oldestEntry: null,
newestEntry: null
};
}
}
}
export const localScraperCacheService = LocalScraperCacheService.getInstance();
export default localScraperCacheService;

View file

@ -4,6 +4,8 @@ import { Platform } from 'react-native';
import { logger } from '../utils/logger';
import { Stream } from '../types/streams';
import { cacheService } from './cacheService';
import { localScraperCacheService } from './localScraperCacheService';
import { hybridCacheService } from './hybridCacheService';
import CryptoJS from 'crypto-js';
// Types for local scrapers
@ -859,7 +861,7 @@ class LocalScraperService {
}
}
// Execute scrapers for streams
// Execute scrapers for streams with caching
async getStreams(type: string, tmdbId: string, season?: number, episode?: number, callback?: ScraperCallback): Promise<void> {
await this.ensureInitialized();
@ -876,20 +878,51 @@ class LocalScraperService {
logger.log('[LocalScraperService] No enabled scrapers found for type:', type);
return;
}
// Check cache for existing results (hybrid: global first, then local)
const { validResults, expiredScrapers, allExpired, source } = await hybridCacheService.getCachedResults(type, tmdbId, season, episode);
logger.log('[LocalScraperService] Executing', enabledScrapers.length, 'scrapers for', type, tmdbId);
// Immediately return cached results for valid scrapers
if (validResults.length > 0) {
logger.log(`[LocalScraperService] Returning ${validResults.length} cached results for ${type}:${tmdbId} (source: ${source})`);
for (const cachedResult of validResults) {
if (cachedResult.success && cachedResult.streams.length > 0) {
// Streams are already in the correct format, just pass them through
if (callback) {
callback(cachedResult.streams, cachedResult.scraperId, cachedResult.scraperName, null);
}
} else if (callback) {
// Return error for failed cached results
const error = cachedResult.error ? new Error(cachedResult.error) : new Error('Scraper failed');
callback(null, cachedResult.scraperId, cachedResult.scraperName, error);
}
}
}
// Determine which scrapers need to be re-run
const scrapersToRerun = enabledScrapers.filter(scraper =>
expiredScrapers.includes(scraper.id) || !validResults.some(r => r.scraperId === scraper.id)
);
if (scrapersToRerun.length === 0) {
logger.log('[LocalScraperService] All scrapers have valid cached results');
return;
}
logger.log(`[LocalScraperService] Re-running ${scrapersToRerun.length} scrapers (${expiredScrapers.length} expired, ${scrapersToRerun.length - expiredScrapers.length} not cached) for ${type}:${tmdbId}`);
// Generate a lightweight request id for tracing
const requestId = `rs_${Date.now().toString(36)}_${Math.random().toString(36).slice(2, 6)}`;
// Execute each scraper
for (const scraper of enabledScrapers) {
this.executeScraper(scraper, type, tmdbId, season, episode, callback, requestId);
// Execute only scrapers that need to be re-run
for (const scraper of scrapersToRerun) {
this.executeScraperWithCaching(scraper, type, tmdbId, season, episode, callback, requestId);
}
}
// Execute individual scraper
private async executeScraper(
// Execute individual scraper with caching
private async executeScraperWithCaching(
scraper: ScraperInfo,
type: string,
tmdbId: string,
@ -904,8 +937,6 @@ class LocalScraperService {
throw new Error(`No code found for scraper ${scraper.id}`);
}
// Skip verbose logging to reduce CPU load
// Load per-scraper settings
const scraperSettings = await this.getScraperSettings(scraper.id);
@ -939,20 +970,57 @@ class LocalScraperService {
// Convert results to Nuvio Stream format
const streams = this.convertToStreams(results, scraper);
// Cache the successful result (hybrid: both local and global)
await hybridCacheService.cacheScraperResult(
type,
tmdbId,
scraper.id,
scraper.name,
streams,
null,
season,
episode
);
if (callback) {
callback(streams, scraper.id, scraper.name, null);
}
// Skip verbose logging to reduce CPU load
} catch (error) {
logger.error('[LocalScraperService] Scraper', scraper.name, 'failed:', error);
// Cache the failed result (hybrid: both local and global)
await hybridCacheService.cacheScraperResult(
type,
tmdbId,
scraper.id,
scraper.name,
null,
error as Error,
season,
episode
);
if (callback) {
callback(null, scraper.id, scraper.name, error as Error);
}
}
}
// Execute individual scraper (legacy method - kept for compatibility)
private async executeScraper(
scraper: ScraperInfo,
type: string,
tmdbId: string,
season?: number,
episode?: number,
callback?: ScraperCallback,
requestId?: string
): Promise<void> {
// Delegate to the caching version
return this.executeScraperWithCaching(scraper, type, tmdbId, season, episode, callback, requestId);
}
// Execute scraper code in sandboxed environment
private async executeSandboxed(code: string, params: any): Promise<LocalScraperResult[]> {
// This is a simplified sandbox - in production, you'd want more security
@ -1285,6 +1353,44 @@ class LocalScraperService {
await this.ensureInitialized();
return Array.from(this.installedScrapers.values()).some(scraper => scraper.enabled);
}
// Cache management methods (hybrid: local + global)
async clearScraperCache(): Promise<void> {
await hybridCacheService.clearAllCache();
logger.log('[LocalScraperService] Cleared all scraper cache (local + global)');
}
async invalidateScraperCache(scraperId: string): Promise<void> {
await hybridCacheService.invalidateScraper(scraperId);
logger.log('[LocalScraperService] Invalidated cache for scraper:', scraperId);
}
async invalidateContentCache(type: string, tmdbId: string, season?: number, episode?: number): Promise<void> {
await hybridCacheService.invalidateContent(type, tmdbId, season, episode);
logger.log('[LocalScraperService] Invalidated cache for content:', `${type}:${tmdbId}`);
}
async getCacheStats(): Promise<{
local: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
};
global: {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
hitRate: number;
};
combined: {
totalEntries: number;
hitRate: number;
};
}> {
return await hybridCacheService.getCacheStats();
}
}
export const localScraperService = LocalScraperService.getInstance();

View file

@ -0,0 +1,453 @@
import { supabase } from './supabaseClient';
import { logger } from '../utils/logger';
import { Stream } from '../types/streams';
export interface GlobalCachedScraperResult {
streams: Stream[];
timestamp: number;
success: boolean;
error?: string;
scraperId: string;
scraperName: string;
contentKey: string; // e.g., "movie:123" or "tv:123:1:2"
}
export interface GlobalCacheStats {
totalEntries: number;
totalSize: number;
oldestEntry: number | null;
newestEntry: number | null;
hitRate: number;
}
class SupabaseGlobalCacheService {
private static instance: SupabaseGlobalCacheService;
private readonly TABLE_NAME = 'scraper_cache';
private readonly DEFAULT_TTL_MS = 30 * 60 * 1000; // 30 minutes default TTL
private readonly FAILED_RETRY_TTL_MS = 5 * 60 * 1000; // 5 minutes for failed scrapers
private readonly SUCCESS_TTL_MS = 60 * 60 * 1000; // 1 hour for successful scrapers
private readonly MAX_CACHE_AGE_MS = 7 * 24 * 60 * 60 * 1000; // 7 days max age
private readonly BATCH_SIZE = 50; // Batch size for operations
// Cache hit/miss tracking
private cacheHits = 0;
private cacheMisses = 0;
private constructor() {}
public static getInstance(): SupabaseGlobalCacheService {
if (!SupabaseGlobalCacheService.instance) {
SupabaseGlobalCacheService.instance = new SupabaseGlobalCacheService();
}
return SupabaseGlobalCacheService.instance;
}
/**
* Generate cache key for content
*/
private getContentKey(type: string, tmdbId: string, season?: number, episode?: number): string {
if (season !== undefined && episode !== undefined) {
return `${type}:${tmdbId}:${season}:${episode}`;
}
return `${type}:${tmdbId}`;
}
/**
* Generate unique key for scraper result
*/
private getScraperKey(contentKey: string, scraperId: string): string {
return `${contentKey}:${scraperId}`;
}
/**
* Check if cached result is still valid based on TTL
*/
private isCacheValid(timestamp: number, success: boolean): boolean {
const ttl = success ? this.SUCCESS_TTL_MS : this.FAILED_RETRY_TTL_MS;
return Date.now() - timestamp < ttl;
}
/**
* Get cached results for content from global cache
*/
async getCachedResults(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<{
validResults: GlobalCachedScraperResult[];
expiredScrapers: string[];
allExpired: boolean;
}> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const { data, error } = await supabase
.from(this.TABLE_NAME)
.select('*')
.eq('content_key', contentKey)
.gte('created_at', new Date(Date.now() - this.MAX_CACHE_AGE_MS).toISOString());
if (error) {
logger.error('[GlobalCache] Error fetching cached results:', error);
this.cacheMisses++;
return {
validResults: [],
expiredScrapers: [],
allExpired: true
};
}
if (!data || data.length === 0) {
this.cacheMisses++;
return {
validResults: [],
expiredScrapers: [],
allExpired: true
};
}
// Filter valid results and identify expired scrapers
const validResults: GlobalCachedScraperResult[] = [];
const expiredScrapers: string[] = [];
for (const row of data) {
const result: GlobalCachedScraperResult = {
streams: row.streams || [],
timestamp: new Date(row.created_at).getTime(),
success: row.success,
error: row.error,
scraperId: row.scraper_id,
scraperName: row.scraper_name,
contentKey: row.content_key
};
if (this.isCacheValid(result.timestamp, result.success)) {
validResults.push(result);
} else {
expiredScrapers.push(result.scraperId);
}
}
// Track cache hits
if (validResults.length > 0) {
this.cacheHits++;
} else {
this.cacheMisses++;
}
logger.log(`[GlobalCache] Retrieved ${validResults.length} valid results, ${expiredScrapers.length} expired scrapers for ${contentKey}`);
return {
validResults,
expiredScrapers,
allExpired: validResults.length === 0
};
} catch (error) {
logger.error('[GlobalCache] Error getting cached results:', error);
this.cacheMisses++;
return {
validResults: [],
expiredScrapers: [],
allExpired: true
};
}
}
/**
* Cache results for specific scrapers in global cache
*/
async cacheResults(
type: string,
tmdbId: string,
results: Array<{
scraperId: string;
scraperName: string;
streams: Stream[] | null;
error: Error | null;
}>,
season?: number,
episode?: number
): Promise<void> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const now = new Date().toISOString();
// Prepare batch insert data
const insertData = results.map(result => ({
scraper_key: this.getScraperKey(contentKey, result.scraperId),
content_key: contentKey,
scraper_id: result.scraperId,
scraper_name: result.scraperName,
streams: result.streams || [],
success: !result.error && result.streams !== null,
error: result.error?.message || null,
created_at: now,
updated_at: now
}));
// Use upsert to handle duplicates
const { error } = await supabase
.from(this.TABLE_NAME)
.upsert(insertData, {
onConflict: 'scraper_key',
ignoreDuplicates: false
});
if (error) {
logger.error('[GlobalCache] Error caching results:', error);
} else {
logger.log(`[GlobalCache] Cached ${results.length} results for ${contentKey}`);
}
} catch (error) {
logger.error('[GlobalCache] Error caching results:', error);
}
}
/**
* Cache a single scraper result
*/
async cacheScraperResult(
type: string,
tmdbId: string,
scraperId: string,
scraperName: string,
streams: Stream[] | null,
error: Error | null,
season?: number,
episode?: number
): Promise<void> {
await this.cacheResults(type, tmdbId, [{
scraperId,
scraperName,
streams,
error
}], season, episode);
}
/**
* Get list of scrapers that need to be re-run (expired or not cached globally)
*/
async getScrapersToRerun(
type: string,
tmdbId: string,
availableScrapers: Array<{ id: string; name: string }>,
season?: number,
episode?: number
): Promise<string[]> {
const { validResults, expiredScrapers } = await this.getCachedResults(type, tmdbId, season, episode);
const validScraperIds = new Set(validResults.map(r => r.scraperId));
const expiredScraperIds = new Set(expiredScrapers);
// Return scrapers that are either expired or not cached globally
const scrapersToRerun = availableScrapers
.filter(scraper =>
!validScraperIds.has(scraper.id) || expiredScraperIds.has(scraper.id)
)
.map(scraper => scraper.id);
logger.log(`[GlobalCache] Scrapers to re-run: ${scrapersToRerun.join(', ')}`);
return scrapersToRerun;
}
/**
* Get all valid cached streams for content from global cache
*/
async getCachedStreams(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<Stream[]> {
const { validResults } = await this.getCachedResults(type, tmdbId, season, episode);
// Flatten all valid streams
const allStreams: Stream[] = [];
for (const result of validResults) {
if (result.success && result.streams) {
allStreams.push(...result.streams);
}
}
return allStreams;
}
/**
* Invalidate cache for specific content globally
*/
async invalidateContent(
type: string,
tmdbId: string,
season?: number,
episode?: number
): Promise<void> {
try {
const contentKey = this.getContentKey(type, tmdbId, season, episode);
const { error } = await supabase
.from(this.TABLE_NAME)
.delete()
.eq('content_key', contentKey);
if (error) {
logger.error('[GlobalCache] Error invalidating cache:', error);
} else {
logger.log(`[GlobalCache] Invalidated global cache for ${contentKey}`);
}
} catch (error) {
logger.error('[GlobalCache] Error invalidating cache:', error);
}
}
/**
* Invalidate cache for specific scraper across all content globally
*/
async invalidateScraper(scraperId: string): Promise<void> {
try {
const { error } = await supabase
.from(this.TABLE_NAME)
.delete()
.eq('scraper_id', scraperId);
if (error) {
logger.error('[GlobalCache] Error invalidating scraper cache:', error);
} else {
logger.log(`[GlobalCache] Invalidated global cache for scraper ${scraperId}`);
}
} catch (error) {
logger.error('[GlobalCache] Error invalidating scraper cache:', error);
}
}
/**
* Clear all cached results globally (admin function)
*/
async clearAllCache(): Promise<void> {
try {
const { error } = await supabase
.from(this.TABLE_NAME)
.delete()
.neq('id', 0); // Delete all rows
if (error) {
logger.error('[GlobalCache] Error clearing cache:', error);
} else {
logger.log('[GlobalCache] Cleared all global cache');
}
} catch (error) {
logger.error('[GlobalCache] Error clearing cache:', error);
}
}
/**
* Clean up old cache entries (older than MAX_CACHE_AGE_MS)
*/
async cleanupOldEntries(): Promise<void> {
try {
const cutoffDate = new Date(Date.now() - this.MAX_CACHE_AGE_MS).toISOString();
const { error } = await supabase
.from(this.TABLE_NAME)
.delete()
.lt('created_at', cutoffDate);
if (error) {
logger.error('[GlobalCache] Error cleaning up old entries:', error);
} else {
logger.log('[GlobalCache] Cleaned up old cache entries');
}
} catch (error) {
logger.error('[GlobalCache] Error cleaning up old entries:', error);
}
}
/**
* Get global cache statistics
*/
async getCacheStats(): Promise<GlobalCacheStats> {
try {
// Get total count
const { count: totalEntries, error: countError } = await supabase
.from(this.TABLE_NAME)
.select('*', { count: 'exact', head: true });
if (countError) {
logger.error('[GlobalCache] Error getting cache stats:', countError);
return {
totalEntries: 0,
totalSize: 0,
oldestEntry: null,
newestEntry: null,
hitRate: 0
};
}
// Get oldest and newest entries
const { data: oldestData } = await supabase
.from(this.TABLE_NAME)
.select('created_at')
.order('created_at', { ascending: true })
.limit(1);
const { data: newestData } = await supabase
.from(this.TABLE_NAME)
.select('created_at')
.order('created_at', { ascending: false })
.limit(1);
const oldestEntry = oldestData?.[0] ? new Date(oldestData[0].created_at).getTime() : null;
const newestEntry = newestData?.[0] ? new Date(newestData[0].created_at).getTime() : null;
// Calculate hit rate
const totalRequests = this.cacheHits + this.cacheMisses;
const hitRate = totalRequests > 0 ? (this.cacheHits / totalRequests) * 100 : 0;
return {
totalEntries: totalEntries || 0,
totalSize: 0, // Size calculation would require additional queries
oldestEntry,
newestEntry,
hitRate
};
} catch (error) {
logger.error('[GlobalCache] Error getting cache stats:', error);
return {
totalEntries: 0,
totalSize: 0,
oldestEntry: null,
newestEntry: null,
hitRate: 0
};
}
}
/**
* Reset cache hit/miss statistics
*/
resetStats(): void {
this.cacheHits = 0;
this.cacheMisses = 0;
}
/**
* Get cache hit/miss statistics
*/
getHitMissStats(): { hits: number; misses: number; hitRate: number } {
const totalRequests = this.cacheHits + this.cacheMisses;
const hitRate = totalRequests > 0 ? (this.cacheHits / totalRequests) * 100 : 0;
return {
hits: this.cacheHits,
misses: this.cacheMisses,
hitRate
};
}
}
export const supabaseGlobalCacheService = SupabaseGlobalCacheService.getInstance();
export default supabaseGlobalCacheService;