QuandlCS is a simple wrapper around the Quandl API to provide easy access. The library currently provides download functionality supporting simple, multiset, favourite, metadata and search request downloads.
Started as a project in an attempt at my colleagues Project365
Visit the online reference for more information about the Quandl API.
The CS API is simple to use and provides type safety garuantees and basic validation on data before requesting. Interaction is based around the concept of a request with all download requests implementing a single interface. Once a request is created this can be used with the QuandlConnection object to download and return the data or to generate the request string URL (www.quandl.com/api/v1/datasets/PRAGUESE/PX.json) to handle manually.
The API has a few fundamental types that mimick Quandl variables and options.
The datacode class represents the Quandl Code that uniquely identifies each dataset. It is made up of a Source and a Code.
Datacode code = new Datacode("PRAGUESE", "PX"); // PRAGUESE is the source, PX is the datacodeWhen used this class will validate that the source and code provided do not contain any invalid characters but will not validate that the code does exist.
The FileFormats enum represents each of the available output formats from the Quandl API. Not all of the formats will be valid for each of the different requests. (e.g. Metadata requests are only available in XML and JSON formats... this is validated by the QuandlMetadataRequest class when used)
The Frequencies enum is used to specify which frequency collapse is required for the dataset being downloaded.
The SortOrders enum, as you can probably imagine, specified if the data should be sorted ascending or descending.
The Transformations enum specifies which transformation should be applied to the before downloading.
The simplest and most useful request to be made is a download.
using System;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
QuandlDownloadRequest request = new QuandlDownloadRequest();
request.APIKey = "1234-FAKE-KEY-4321";
request.Datacode = new Datacode("PRAGUESE", "PX"); // PRAGUESE is the source, PX is the datacode
request.Format = FileFormats.JSON;
request.Frequency = Frequencies.Monthly;
request.Truncation = 150;
request.Sort = SortOrders.Ascending;
request.Transformation = Transformations.Difference;
Console.WriteLine("The request string is : {0}", request.ToRequestString());
}
}
}The multiset download provides an almost identical interface to the simple download documented above, but instead of taking just one datacode object it provides two methods to add columns for the multiset. The columns are downloaded in the same order as they are added to this object.
using System;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
QuandlMultisetRequest request = new QuandlMultisetRequest();
request.APIKey = "1234-FAKE-KEY-4321";
Datacode first = new Datacode("GOOG", "NASDAQ_GOOG");
request.AddColumns(first);
Datacode second = new Datacode("GOOG", "NASDAQ_AAPL");
request.AddColumn(second, 4);
request.Format = FileFormats.HTML;
request.Frequency = Frequencies.Daily;
request.Transformation = Transformations.None;
request.StartDate = new DateTime(2000, 01, 01); // Terrible date to start as Google IPO was 19/08/2004...
request.EndDate = new DateTime(2014, 01, 19);
Console.WriteLine("The request string is : {0}", request.ToRequestString());
}
}
}The metadata download is useful for gathering the key details about a dataset. It provides a very simple interface needing only a dataset, format and an API key.
using System;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
QuandlMetadataRequest request = new QuandlMetadataRequest();
request.APIKey = "1234-FAKE-KEY-4321";
request.Datacode = new Datacode("NSE", "OIL");
request.Format = FileFormats.XML;
Console.WriteLine("The request string is : {0}", request.ToRequestString());
}
}
}The favourites download is used to gather information about the current user (as specified through the required API key) and which datasets are their favourites. I imagine this would be useful if you were to write a desktop application to allow users to manage their Quandl data... that's what I am planning to use it for.
using System;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
QuandlFavouritesRequest request = new QuandlFavouritesRequest();
request.APIKey = "1234-FAKE-KEY-4321";
request.Format = FileFormats.XML;
Console.WriteLine("The request string is : {0}", request.ToRequestString());
}
}
}The search is pretty self explanatory in terms of what it is for...
using System;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
QuandlSearchRequest request = new QuandlSearchRequest();
request.APIKey = "1234-FAKE-KEY-4321";
request.Format = FileFormats.XML;
request.SearchQuery = "OIL";
Console.WriteLine("The request string is : {0}", request.ToRequestString());
}
}
}In order to make using the request objects easier there is also a QuandlConnection class that will take in any of the requests and then download the data returning it as a string to the caller...
using System;
using System.IO;
using QuandlCS.Connection;
using QuandlCS.Interfaces;
using QuandlCS.Requests;
using QuandlCS.Types;
namespace QuandlCSTest
{
class Program
{
static void Main(string[] args)
{
IQuandlRequest request = CreateRequest(); // Implementation of CreateRequest() left as a challenge to the reader
IQuandlConnection connection = new QuandlConnection();
string data = connection.Request(request);
using (StreamWriter writer = new StreamWriter(@"C:\TestOutput.xml"))
{
writer.Write(data);
}
}
}
}:koala: