#Javascript Data Conductor Provides tools to make efficient access to remote data sources trivial.

Main components

At the time of this writing, the following main components are available (with more to come. See roadmap.md)

• AdapterService - Provides a global location to store adapters.
• IndexedPagerService - A service which allows automatic caching and dropping of data according to the currently set index.

##What it does

###IndexedPagerService The simplest way to describe the behaviour of the pager is in the following image

Above you can see that as the selection changes, the pager loads the desired pages, including a page to the left and to the right of the selected page. Additionally the pager ensures the that there are a maximum of 3 pages stored, and drops the page furthest from the current index in order to accomadate newly loaded pages. The pager configuration for the behaviour above is as follows:

pager.setAdapter('myAdapter');  //Your adapter or adapter name here, see below
pager.setPageSize(10);          //How many records you want with each page
pager.setMaxPages(3);           //Ensure no more than 3 pages cached
pager.setPadLow(1);             //Load one page below the current index
pager.setPagHigh(1);            //Load one page above the current index
pager.setIndex(0);              //Kick off the loads by setting the initial index

###AdapterService This provides named base storage for data adapters, as well as a global singleton (AdapterService.global) which can be used to store adapters, and later set them by name in components such as the pager.

##Getting Started In order to get started with this library, 3 steps are required.

1 Download

git clone https://github.com/major-mann/data.conductor.git

Optionally, you may wish to develop, or view the demo, in which case you should do

npm install

Once done, launch the demo server and demo in your browser by

npm start

2 Create an adapter to hook up to your data source. Here is some sample code to get you going.

var adapter = {
    //In both cases the argument structure is the same
    //filter will be fully defined by the you (for example by calling setFilter on the pager)
    //args contains the following:
    //  args.skip - The data index to start reading at
    //  args.limit - The number of records to retrieve
    //  args.cancel - A promise that will be resolved if the consumer wishes to cancel the call
    find: function (filter, args) {
        //We will be returning a promise to be resolved once we have the requested data
        var deferred = $q.defer(),
            filt = {
                where: filter,
                skip: args.skip,
                limit: args.limit
            };

        //This would usually be some kind of remote service call (eg. using http)
        mydb.find(filt, function(err, data) {
            if (err) {
                deferred.reject(err);
            } else {
                //Note: We expect that data will be an array.
                //  If there is no data at the specified index, null should be returned.
                //  If there are not enough records in the dataset to fulfil the request,
                //      as many records as possible should be returned.
                deferred.resolve(data);
            }
        });

        return deferred.promise;
    },
    //Note: this is optional, but is often quite easily implemented, and quite useful to consumers.
    count: function (filter, args) {
        //We will be returning a promise to be resolved once we have the count
        var deferred = $q.defer(),
            filt = {
                where: filter,
                skip: args.skip,
                limit: args.limit
            };

        //This would usually be some kind of remote service call (eg. using http)
        mydb.count(filt, function(err, count) {
            if (err) {
                deferred.reject(err);
            } else {
                //We expect a number for count
                deferred.resolve(count);
            }
        });

        return deferred.promise;
    }
}

Note: You may optionally add this adapter to the AdapterService singleton, and in future refer to it by name. For example:

var adapter = { };
AdapterService.global.add('myAdapterName', adapter);

//The pager may be then referenced by name
pager.setAdapter('myAdapterName');
//Or, you may choose to use the object directly, without involving the global AdapterService.
pager.setAdapter(adapter);

3 Once you have an adapter, you may create a pager, and set it's adapter to the previously created adapter.

var pager = new IndexedPager();
pager.setAdapter(myAdapter);
//Set any other options here (See the docs for more info)
pager.setIndex().then(function () {
    var pages = pager.pages(),
        i, j;

    for (i = 0; i < pages.length; i++) {
        console.log('Index: ' + pages[i].index);
        console.log('Loading: ' + pages[i].loading);
        for (j = 0; j < pages[i].data.length; j++) {
            console.log('Page ' + i + ', Item ' + j);
        }
        //Note: Each page has a cancel function. If called while loading, the load will be cancelled.
    }
});

To view the full capabilities, launch the server by doing npm install && npm start

For more information, see the documentation.