How to cache query result in ember data

2020-02-26 13:53发布

I want to cache the result of a query in ember-data. (findQuery)

To make it clear: I don't want to cache the entire models; just what models are the result of the query. Where is the right place for this?

I was thinking about implementing this in the adapter and cache the result of the AJAX call, but I don't think this is a good solution since I don't wanna override the loaded and maybe newer and/or modified model data.

I don't thinks its possible to just return a list of ID's, and to manipulate the adapter and the serializer for this simple use-case seems to be messy!

Actually I don't want that findQuery is called for specific types of querys. Like the behavior of findAll. Nice would be something like a queryShouldBeCached hook.

Is there a good Solution for this?

2条回答
男人必须洒脱
2楼-- · 2020-02-26 14:27

I'm not an Ember expert but I think you can address your problem with a pure JS solution.

Given Ember Data queries return Promises, e.g. return this.store.findAll('blog-post'); // => Promise, we can cache promises in a simple object with higher order functions (functions that return functions). The object cache could be replaced with localStorage, sessionStorage, Map or even WeakMap but I'm using the object cache to make things simple to understand.

What you want to essentially do is to replace following call:

return this.store.findAll('blog-post');

with something more or less like:

return cachedStore.findAll('blog-post');

actually, with the solution below it might look more like:

return cachedStore.call(this, 'findAll', 'blog-post');

As a result, you will request data once and always return from cache in subsequent calls.

Let me show you how the implementation might look like:

var cachedStore = (function () {
  // Your cache - in this case simple object literal
  var cache = {};

  // Actual method that will check cache for results before trying to query services
  return function (method) {
    var args = Array.prototype.slice.call(arguments, 1);
    var serializedArgs = JSON.stringify(args);

    if (!(serializedArgs in cache)) {
      cache[serializedArgs] = this.store[method].apply(this, args);
    }
    return cache[serializedArgs];
  };
}());

And here's a sample usage:

// Fires a request
cachedStore.call(this, 'findAll', 'blog-post');
// Returns from cache
cachedStore.call(this, 'findAll', 'blog-post');
// Returns from cache
cachedStore.call(this, 'findAll', 'blog-post');

// Fires a request
cachedStore.call(this, 'findRecord', 'blog-post', 123);
// Returns from cache
cachedStore.call(this, 'findRecord', 'blog-post', 123);
// Returns from cache
cachedStore.call(this, 'findRecord', 'blog-post', 123);

Does that help in any way?

查看更多
乱世女痞
3楼-- · 2020-02-26 14:50

When you fire a request based on params Ember Data doesn't know how those params necessarily translate into the models (aka it doesn't know that you have all of the records that have some sort of relationship param1). You can cache it yourself, but then you'd still need some sort of way of knowing those records from other records in your store.

查看更多
登录 后发表回答