• Overview and Background
• Why IndexedDB?
• Installation
• Usage
• Initializing in a Lowcoder Parent Application
• Using Shared Cache in Nested Modules
• Fallback to Local Cache in Nested Modules
• API Methods
  • createDB()
  • init()
  • get()
  • set()
  • getAll()
  • invalidate()
  • preload()
• TTL, Concurrency, and Preload
• Debugging and Logging
• Best Practices
• Troubleshooting

AppCacheManager is a UMD-based JavaScript library for client-side caching using IndexedDB. It provides a simple key-value API that wraps the low-level IndexedDB database, making it easy to store and retrieve complex data in the browser. In a Lowcoder application (or any web app), this library enables large-capacity, persistent caching of data (e.g., API responses) across page loads and offline sessions. Unlike simple localStorage, IndexedDB can store structured data and blobs in a transactional database without blocking the main thread.

AppCacheManager builds on IndexedDB to give developers an intuitive interface: you create or open a database, define object stores, and then use get/set methods with optional expiration (TTL) semantics. Many client-side libraries (e.g., LocalForage) wrap IndexedDB behind a friendly API. Similarly, AppCacheManager abstracts the boilerplate of IndexedDB so you can focus on your cache data, TTL policies, and concurrency needs.

It is particularly useful in Lowcoder’s modular apps: you can initialize a cache in the parent app and then share or fallback gracefully in nested modules.

IndexedDB is a transactional, asynchronous, NoSQL database built into modern browsers. It is designed to handle large amounts of structured data on the client (far beyond what localStorage can handle). Key concepts:

• Object stores: Similar to tables in relational DBs, but NoSQL. A database can have multiple object stores to organize data.
• Keys and values: Every stored value is associated with a key. Keys can be simple or complex, but most libraries (including AppCacheManager) use a string key for easy lookups.
• Asynchronous API: All operations (open, read, write) are asynchronous. This avoids blocking the UI thread, but requires callbacks or promises.

Because IndexedDB’s native API is verbose, AppCacheManager wraps it in a promise-based API. When you call createDB({dbName, stores}), it uses indexedDB.open and creates any missing object stores. It provides a neat object with methods (get, set, etc.) so you don’t have to write the boilerplate IndexedDB transaction code yourself.

Note: Unlike localStorage, IndexedDB has no built-in expiration or “cache eviction” policy. AppCacheManager handles expiration (TTL) by storing a timestamp in each record’s meta. On reads, it checks if data has expired and flags it.

Browser script include:

<script src="path/to/AppCacheManager.umd.js"></script>  
<script>  
  console.log(typeof AppCacheManager.createDB === 'function');
</script>

Node / Bundler:

const AppCacheManager = require('appcache-manager');  
import AppCacheManager from 'appcache-manager';

Typical usage pattern:

1. Create a cache instance with createDB({dbName, stores, version, debug}).
2. Initialize the cache by calling .init().
3. Use get/set to read/write data to a given store with a given key.
4. (Optional) Use invalidate or preload for batch operations.

window.AppCache = AppCacheManager.createDB({  
  dbName: 'MyLowcoderAppCache',  
  stores: ['users', 'settings', 'reports'],  
  version: 1,  
  debug: true  
});  

window.AppCache.init().then(cache => {  
  console.log('AppCache initialized:', cache.name);  
});

const cache = window.AppCache || AppCacheManager.createDB({  
  dbName: 'MyLowcoderAppCache',  
  stores: ['users', 'settings', 'reports']  
});  

await cache.init();  
cache.get('settings', 'theme').then(theme => console.log('Shared theme setting:', theme));

let cache;  
if (window.AppCache) {  
  cache = window.AppCache;  
} else {  
  cache = AppCacheManager.createDB({  
    dbName: 'MyModuleLocalCache',  
    stores: ['localData'],  
    debug: false  
  });  
}  

await cache.init();  
const data = await cache.get('localData', 'key1');  
console.log('Local cached data:', data);

AppCacheManager.createDB({ dbName, version = 1, stores = [], debug = false })

await cache.init()

const value = await cache.get(storeName, key)

await cache.set(storeName, key, value, { ttlMs })

const arrayOfValues = await cache.getAll(storeName)

await cache.invalidate(storeName, key)

await cache.preload(defs, { concurrency, batchDelayMs, retries })

• TTL: Per-item expiration via expiresAt.
• Concurrency: Limit parallel fetches with concurrency.
• Retry logic: Automatic retry in preload().
• Batch delays: Throttle between tasks with batchDelayMs.

Enable via debug: true. Logs include:

[AppCacheManager] DB ready: MyLowcoderAppCache
[AppCacheManager] Cache hit for users:allUsers, skipping fetch.
[AppCacheManager] Preload complete.

• Organize stores logically.
• Use consistent key naming.
• Version schema changes carefully.
• Apply TTL only to data that should expire.
• Monitor cache size and purge expired entries.
• Use DevTools to inspect IndexedDB.
• Always handle errors with try/catch.

• Database not opening / errors: Check console for version or store issues.
• Missing store error: Include all needed stores at creation.
• Unexpected null returns: Ensure await cache.init() is called.
• __expired flags: Indicates data TTL has passed.
• Debug logs missing: Verify environment and UMD build version.
• Module conflicts: Ensure modules share schema or only parent initializes the cache.