Network Programming in .NET

Uploading a file to AWS S3 via C# has been documented a million times, but I thought I’d put my own version here, a simple upload from a HTML page, a bit of ajax, and an ASPX file to upload the file to S3, and return a unique file name.

Firstly, you need to include the Nuget package with Install-Package AWSSDK.S3

Here’s the javascript:

$(document).ready(function(e) {
$(“#uploadimage”).on(‘submit’,
(function(e) {
e.preventDefault();
$.ajax({
url: “/api/imageUpload.aspx”, // Url to which the request is sent
type: “POST”, // Type of request to be send, called as method
data: new
FormData(
this), // Data sent to server, a set of key/value pairs (i.e. form fields and values)
contentType: false, // The content type used when sending data to the server.
cache: false, // To unable request pages to be cached
processData: false, // To send DOMDocument or non processed data file it is set to false
success: function(data) // A function to be called if request succeeds
{
$(“#previewing”).attr(“src”,data);
}
});}));});

Then, here is the api/imageUpload.aspx file codebehind (with keys removed)

var MyFiles = Request.Files;
for (int l = 0; l < MyFiles.Count; l++)
{
if (MyFiles.GetKey(l) != “file”) continue;
var file = MyFiles.Get(“file”);
var strExt = Path.GetExtension(file.FileName);
var filename = Guid.NewGuid().ToString(“D”) + strExt;
var cfg = new AmazonS3Config {RegionEndpoint = Amazon.RegionEndpoint.EUWest1};
const string bucketName = “xxxxxx”;
var s3Client = new AmazonS3Client(“xxxxx”, “xxxxx”, cfg);
var fileTransferUtility = new TransferUtility(s3Client);
var fileTransferUtilityRequest = new TransferUtilityUploadRequest
{
BucketName = bucketName,
InputStream = file.InputStream,
StorageClass = S3StorageClass.ReducedRedundancy,
Key = filename,
CannedACL = S3CannedACL.PublicRead
};
fileTransferUtility.Upload(fileTransferUtilityRequest);
Response.Write(“https://s3-eu-west-1.amazonaws.com/xxxxxx/&#8221; + filename);
}

The Bing Image search based on the DataMarket service by Microsoft is closing down, and it is moving to cognitive services. This means that if you have code that uses an endpoint like this;

https://api.datamarket.azure.com/Bing/Search/v1/Image …

It needs to be changed to code like this:

https://api.cognitive.microsoft.com/bing/v5.0/images/search …

And your quote is reduced to 1,000 per month… 😦

Here’s some code I’ve written in C# to do a basic image search

public static List<String> GetImages(string searchText)
{
var strUrl = “https://api.cognitive.microsoft.com/bing/v5.0/images/search?q=&#8221; + searchText + “&count=1000”;
var wc = new WebClient();
wc.Headers[“Ocp-Apim-Subscription-Key”] = “xxxxxxxxxxx”;
var strJson = wc.DownloadString(strUrl);
var bing = JavascriptDeserialize<BingSearchObject>(strJson);
var lResults = new List<string>();
foreach (var result in bing.value)
{
var qs = HttpUtility.ParseQueryString(result.contentUrl);
lResults.Add(qs[“r”]);

}
return lResults;
}

Where “BingSearchObject” is defined as:

#region BingSearchResults
public class Instrumentation
{
public string pageLoadPingUrl { get; set; }
}

public class Thumbnail
{
public int width { get; set; }
public int height { get; set; }
}

public class InsightsSourcesSummary
{
public int shoppingSourcesCount { get; set; }
public int recipeSourcesCount { get; set; }
}

public class Value
{
public string name { get; set; }
public string webSearchUrl { get; set; }
public string thumbnailUrl { get; set; }
public string datePublished { get; set; }
public string contentUrl { get; set; }
public string hostPageUrl { get; set; }
public string contentSize { get; set; }
public string encodingFormat { get; set; }
public string hostPageDisplayUrl { get; set; }
public int width { get; set; }
public int height { get; set; }
public Thumbnail thumbnail { get; set; }
public string imageInsightsToken { get; set; }
public InsightsSourcesSummary insightsSourcesSummary { get; set; }
public string imageId { get; set; }
public string accentColor { get; set; }
}

public class Thumbnail2
{
public string thumbnailUrl { get; set; }
}

public class QueryExpansion
{
public string text { get; set; }
public string displayText { get; set; }
public string webSearchUrl { get; set; }
public string searchLink { get; set; }
public Thumbnail2 thumbnail { get; set; }
}

public class Thumbnail3
{
public string thumbnailUrl { get; set; }
}

public class Suggestion
{
public string text { get; set; }
public string displayText { get; set; }
public string webSearchUrl { get; set; }
public string searchLink { get; set; }
public Thumbnail3 thumbnail { get; set; }
}

public class PivotSuggestion
{
public string pivot { get; set; }
public List<Suggestion> suggestions { get; set; }
}

public class Thumbnail4
{
public string url { get; set; }
}

public class SimilarTerm
{
public string text { get; set; }
public string displayText { get; set; }
public string webSearchUrl { get; set; }
public Thumbnail4 thumbnail { get; set; }
}

public class BingSearchObject
{
public string _type { get; set; }
public Instrumentation instrumentation { get; set; }
public string readLink { get; set; }
public string webSearchUrl { get; set; }
public int totalEstimatedMatches { get; set; }
public List<Value> value { get; set; }
public List<QueryExpansion> queryExpansions { get; set; }
public int nextOffsetAddCount { get; set; }
public List<PivotSuggestion> pivotSuggestions { get; set; }
public bool displayShoppingSourcesBadges { get; set; }
public bool displayRecipeSourcesBadges { get; set; }
public List<SimilarTerm> similarTerms { get; set; }
}
#endregion

Image processing in C# is surprisingly fast, and not too difficult. Here is a simple example of how to remove greys from an image, an extract only colour.

Firstly, a pure Grey is one defined as a colour where the Red, Green, and blue component are exactly equal. This spans the spectrum from White (all 255), to Black (all 0). In natural photographs, then the Grey may not be exactly pure, with small variations in the colour components.

So, lets see the core processing code:

public static byte[] PreprocessStep(byte[] input)
{
Stream sInput = new MemoryStream(input);
var imgEasy = Bitmap.FromStream(sInput) as Bitmap;
var img2dEasy = new Color[imgEasy.Width, imgEasy.Height];
for (int i = 0; i < imgEasy.Width; i++)
{
for (int j = 0; j < imgEasy.Height; j++)
{
Color pixel = imgEasy.GetPixel(i, j);

// Preprocessing
if (pixel.R == pixel.G && pixel.B == pixel.G)
{
// Convert all greys to white.
pixel = Color.White;
}

img2dEasy[i, j] = pixel;
}
}

var imgOut = new Bitmap(imgEasy.Width, imgEasy.Height);
for (int i = 0; i < imgEasy.Width; i++)
{
for (int j = 0; j < imgEasy.Height; j++)
{
imgOut.SetPixel(i, j, img2dEasy[i, j]);
}
}
var stream = new MemoryStream();
imgOut.Save(stream, ImageFormat.Png);
var bytes = stream.ToArray();
return bytes;
}

This accepts an image as a byte array, and outputs a byte array in PNG format. by accepting and returning a byte array, it makes it more “pluggable” into different inputs and outputs, like Files, HTTP streams etc.

Here’s how to use it with an input and output file.

static void Main(string[] args)
{
var strExePath = AppDomain.CurrentDomain.BaseDirectory;
var strEasyImage = strExePath + “il_570xN.1036233188_neqb.jpg”;
var sIn = new FileStream(strEasyImage,FileMode.Open);
var br = new BinaryReader(sIn);
var bIn = br.ReadBytes((int)sIn.Length);
sIn.Close();
var bOut = PreprocessStep(bIn);
var sOut = new FileStream(strExePath + “flower.png”,FileMode.Create);
sOut.Write(bOut,0,bOut.Length);
sOut.Close();
}

If you need to be 100% sure that your customer works or lives at the address they have provided you? Our service ensures this by mailing a secure pin code to your customer’s address, which they use to verify that they are physically at the address they have provided you.

It works internationally, and we’ve got an API, so you can integrate into the system programmatically, as part of your customer-onboarding process.

The website is at www.mailaddressverification.com – and you get $1 free to test the service once you open a test account.

The Finnish API for Car registrations; which you can access at http://www.rekisterinumeroapi.com/ now supports both all types of vehicles, not only cars – this includes;

• Private Cars [M1]
• Commercial vehicles (Vans / Trucks) [N1]
• Motorcycles [L3e]
• Mopeds [L1e]
• Snowmobiles
• Tractors
• Trailers