Serializer

• responsible for (Tamper) serialization protocol .serialize(obj) : [Tamper] packs the data .deserialize(obj) : [Tamper] unpacks the data

AbstractSerializer Class that extends --> defines public API/variables that will be shared for all serializers

• two methods: throw Error('not yet implemented') --> serialize and deserialize + utility functions
• shows the API for what a serializer should be
• look at AbstractServices Class in LVP framework

npm install cache-man

CacheManager GarbageCollector AbstractSerializer SetSerializer SingleSerializer

-- JSON serialization protocol -- Tamper serialization protocol

CacheManager --> config to tell which storage it's pointed at

• responsible for managing window.localStorage object window.localStorage stingifies everything

Public API .getItem(key) --> returns an item that is stored in the local cache .getItems(keys) .setItem(key, obj={}) --> adds an item to the localCache .setItems(key, type, obj={})

individual object --> not use Tamper collection of objects (list of articles) --> integrates Tamper

.removeItem(obj) --> removes an item from localCache .removeItems(objs) .hasKey(objKey) --> returns whether an obj with the objKey exists in local cache .length() --> returns number of elements in the object

** .capacity() --> returns the capacity at which localStorage is at .clear() --> clears the entire cache

Garbage Collector

• (1) providing enough space for the cacheManager to store things and (2) maximizing the hit ratio for the cache
• responsible for storage needs --> removing items as they are no longer needed Ex., trying to store in cache, returns error, then garbageCollector.cleanup()

Private API functionality Helper methods to support public API methods

Storage

• capacity : when capacity is reached for localStorage (FIFO)
• popularity :
• time-based : when data is expired/needs to be updated

JSON.stringify(localStorage).length() --> intensive; don't need to expose to public api for capacity unless for debugging

window.localStorage = {

}

In case of failure, do we want to do anything? --> very unlikely that it will happen that often provided it checks capacity first

Examples

var cache = new CacheManager(opts, storageSpace)

opts would contain a name and a debug flag that will console.log every

var article = {_id: 123, id: 123, name: 'Do it yourself', category: 'Tips'}

cache.setItem(article) //only returns an error in case of failure

cache.getItem(article)

• use Asana's subtasks to keep track of all individualized use cases -- easier to get a test script and to get it running quickly
• mocha test -- tests defines use cases

Example of Use Cases reading an expired item reached capacity think about use cases --> will define what helper methods are needed

• Object.defineProperty() ; can set custom getting and setters on an object; for capacity exception

Steps

1. Get something that runs --> create all the classes with public API methods w/ comments --> call methods that won't do anything

2. Start with cacheManager, pick a use case, keep following all the steps down into the classes; tests as you go --> write other classes -- start w/ one use case; setting, just focus on setting, then move to capacity

3. Create mock --test in node test.js creates an instance of cacheManager and pass in a fake window.localStorage object

4. Node testing

5. Browser testing