grunt-sencha-dependencies
Version:
A Grunt.js plugin which will figure out the order of Ext classes your Ext.application uses so the list can be passed on to further commands like concat, jshint, etc
60 lines (50 loc) • 1.93 kB
JavaScript
/**
* @class Ext.data.TwitterProxy
* @extends Ext.data.proxy.JsonP
*
* This simple proxy allows us to use Twitter's JSON-P API to search for tweets. All we're really doing in this
* class is setting a few defaults (such as the number of tweets per page and a simple JSON Reader), and using
* any Filters attached to the read Operation to modify the request url (see buildRequest).
*
*/
Ext.define('Twitter.proxy.Twitter', {
extend: 'Ext.data.proxy.JsonP',
alias: 'proxy.twitter',
config: {
// This is the url we always query when searching for tweets
url: 'http://search.twitter.com/search.json',
extraParams: {
result_type: 'mixed', // Returns popular tweets at the top of the result
lang: 'en' // Only show English results
},
reader: {
type: 'json',
rootProperty: 'results'
}
},
filterParam: undefined,
/**
* We need to add a slight customization to buildRequest - we're just checking for a filter on the
* Operation and adding it to the request params/url, and setting the start/limit if paging
*/
buildRequest: function(operation) {
var request = this.callParent(arguments),
filter = operation.getFilters()[0],
params = request.getParams();
Ext.apply(params, {
rpp: operation.getLimit(),
page: operation.getPage()
});
if (filter) {
delete params.filter;
Ext.apply(params, {
q: filter.getValue() // pass in the query string to the search api
});
request.setParams(params);
request.setUrl(this._url);
// As we're modifiying the request params, we need to regenerate the url now
request.setUrl(this.buildUrl(request));
}
return request;
}
});