TerribleDev/ente
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Simplify text embeddings cache to use only database cache

Browse Source 
Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
laurenspriem
2025-09-10 15:35:58 +05:30
parent cc769fdd5b
commit cda3a5b149
1 changed files with 4 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
mobile/apps/photos/lib/services/text_embeddings_cache_service.dart
View File

@@ -1,7 +1,6 @@
import 'dart:async';
import 'package:logging/logging.dart';
import 'package:photos/core/cache/lru_map.dart';
import 'package:photos/db/ml/db.dart';
import 'package:photos/services/machine_learning/ml_computer.dart';
import 'package:synchronized/synchronized.dart';
@@ -9,7 +8,6 @@ import 'package:synchronized/synchronized.dart';
class TextEmbeddingsCacheService {
static final _logger = Logger('TextEmbeddingsCacheService');
LRUMap<String, List<double>> _memoryCache = LRUMap<String, List<double>>(50);
final _cacheLock = Lock();
TextEmbeddingsCacheService._privateConstructor();
@@ -17,36 +15,22 @@ class TextEmbeddingsCacheService {
Future<List<double>> getEmbedding(String query) async {
return _cacheLock.synchronized(() async {
// 1. Check memory cache
final cachedResult = _memoryCache.get(query);
if (cachedResult != null) {
_logger.info('Text embedding cache hit (memory) for query');
return cachedResult;
}
// 2. Check database
// 1. Check database cache
final dbResult =
await MLDataDB.instance.getRepeatedTextEmbeddingCache(query);
if (dbResult != null) {
_logger.info('Text embedding cache hit (database) for query');
_memoryCache.put(query, dbResult);
_logger.info('Text embedding cache hit for query');
return dbResult;
}
// 3. Compute new embedding
// 2. Compute new embedding
_logger.info('Computing new text embedding for query');
final embedding = await MLComputer.instance.runClipText(query);
// 4. Store in both caches
_memoryCache.put(query, embedding);
// 3. Store in database cache
await MLDataDB.instance.putRepeatedTextEmbeddingCache(query, embedding);
return embedding;
});
}
void clearMemoryCache() {
_memoryCache = LRUMap<String, List<double>>(50);
_logger.info('Cleared text embeddings memory cache');
}
}