I think that anyone who has ever done any GUI programming for a day-to-day application that empowers people to do their job, regardless of the language or environment, will fully agree with me when I say: “end-users are never satisfied.” End-users do not want to click a button and have their GUI freeze up because your application is crunching data on your main program thread. Enter multi-threading. I do not claim to be anything close to an authority on threading (check out http://www.albahari.com/threading/ for everything you’ll ever need to know about C# threading), but I wanted to at least introduce asynchronous web requests because they definitely enhance the end-user experience. Yeah… that’s the story I’m going with anyway, Elliot has been harassing me to actually post here too.

Asynchronous web requests are terribly easy within the .NET framework, someone at Microsoft has already done all of the footwork for us. They wrapped it all up in a WebClient (http://msdn.microsoft.com/en-us/library/system.net.webclient(VS.80).aspx) object, as well.

Lets get started, only minimal code is required to actually implement an asynchronous call. My intention is just to show how to pull down the HTML of a web page, it can be adapted for a lot of other things. Check out the MSDN link (or just mess around with the AutoComplete within Visual Studio)  to see what else the WebClient object has to offer.

First, you have to instantiate our WebClient object and setup our event handler for when the DownloadDataAsync call is completed. This is different then the Generic DownloadData call which blocks the main thread until the download is completed. Our call will be better because it will allow the user, for example, to click a button and have their data download on a different thread and be able to do things on main gui thread. Waiting sucks. Code for this is:

            using (WebClient asyncWebRequest = new WebClient())
            {
                asyncWebRequest.DownloadDataCompleted += asyncWebRequest_DownloadDataCompleted;
                Uri urlToRequest = new Uri("http://www.google.com/");
                asyncWebRequest.DownloadDataAsync(urlToRequest);
            }

The only thing left is actually the code in the method that the data/error will return to, and handle any exceptional cases.

void asyncWebRequest_DownloadDataCompleted(object sender, DownloadDataCompletedEventArgs e)
 {
 if (e.Error != null)
 {
 MessageBox.Show(e.Error.Message);
 return;
 }

 if (e.Result != null && e.Result.Length > 0)
 {
 string downloadedData = Encoding.Default.GetString(e.Result);
 // do things with data here
 }
 else
 {
 MessageBox.Show("No data was downloaded.");
 }

 }

In the end, this functionality enhances the end user experience by not locking up their application while doing background work, and also provides you (the programmer) an easy way to download data in the background without dealing with your own thread. Other considerations may be actually promoting the WebClient to a private variable so you can give the user the option to cancel the request or cancel when the form is closing. This method is meant to be a demonstration of the syntax of a asynchronous call.

PHP is a very popular language for accessing databases for Websites. It is only natural, therefore, that developers may wish to use Microsoft Silverlight to send data to a PHP script, perhaps for database addition, deletion or retrieval. You can use the C# WebClient class to connect to a PHP script for this purpose as follows:

// Test Variables for username and password.

string username=”test”;

string password = “mypassword”;

// Set the Parameters for the PHP Script.

string parameters = String.Format(“?username={0}&password={1}”,username,password);

// Create a Web Client for contacting the PHP Script.

WebClient client = new WebClient();

// Call the PHP Script.

// Add the appropriate header.

client.Headers["Content-Type"] = “application/x-www-form-urlencoded”;

// Add a callback handle for when the Save operation is completed.

client.UploadStringCompleted += new UploadStringCompletedEventHandler(LogonCompleted);

// Invoke the script and its parameters.

client.UploadStringAsync(new Uri(“http://localhost/logonscript.php”+parameters,UriKind.Absolute),”POST”);

In the above code we are calling a sample logon script called ‘logonscript.php’. This dummy script takes a username and password as a parameter. In the sample code above, we first create variables for storing a username and password (‘test’ and ‘mypassword’) respectively. We then format these variables as parameters that the PHP script can read so they will appear to the PHP script as ‘?username=test&password=mypassword’.

We then create a Web Client, set a URL Encoded Header for this client and add an UploadStringCompletedEventHandler (as we are sending parameters to the script). We then invoke the UploadStringAsync method to call the script with its parameters. In our example we are using an absolute path (UriKind.Absolute) and we are posting data to the script as denoted by the “POST” parameter.

In the example the full path is ‘http://localhost/logonscript.php?username=test&password=mypassword’ where the dummy script name is ‘logonscript.php’. This script would reside on a Web Server such as Apache or Microsoft IIS (Internet Information Services).

We referred to a callback handler above. Once the script has run this function is invoked to notify the Silverlight application that it is completed. An example implementation of this callback function is shown below:

private void LogonCompleted(Object sender, UploadStringCompletedEventArgs e)

{

// We assume here that the script prints out “Success” or “Failure”.

if(e.Result.Equals(“Success”))

{

MessageBox.Show(“Logon Succeeded”);

}

else if(e.Result.Equals(“Failure”))

{

MessageBox.Show(“Logon Failed”);

}

}

In this sample function the PHP script prints “Success” if the logon succeeded or “Failure” if the logon failed. This print out is stored in the Result member of event arguments parameter ‘e’. We then check for these strings and display an appropriate message box.

The above example is for sending data to a PHP script. However, we can also retrieve data from a PHP script by print out the results and reading from the Result member of the events argument parameter ‘e’ in our callback function. Alternatively, if we are not sending parameters to the PHP script, we can use WebClient’s DownloadStringAsync() method, set up a DownloadStringCompleted event and define a callback function with a signature similar to the following:

private void DownloadCompleted(Object sender, DownloadStringCompletedEventArgs e)

Atualizando a data de vencimento de um Service Ticket a partir da data do item.

Após alguns meses ausente devido ao trabalho, volto ao meu blog tratando de um assunto não muito comum no mercado SAP: IC WebClient.

Confesso que não tinha conhecimento sobre este framework até, claro, aparecer uma demanda para manutenção neste sistema. A principio, uma percepção da complexidade, posteriormente, a certeza!

Bom, para iniciar, algumas coisas têm que estar claras para um melhor entendimento. SAP CRM 5.0, ICWC e BSP são os assuntos que trataremos aqui.

SAP CRM
CRM – Customer Relationship Management – como o próprio nome sugere, CRM é um sistema de gestão de relacionamento com o cliente. As áreas de atuação deste sistema é voltado para maximizar a experiência e produtividade do cliente, atendendo as áreas de service center, call center, vendas, marketing, canais web, canais de parceiros, suppliers, etc.

ICWC
Interaction Center WebClient – é uma aplicação client-side executada via web browser. Os dados transmitidos para o CRM WebClient são emitidos pelo SAP NetWeaver AS. O SAP NW Application Server possui dois ambientes: J2EE e ABAP; possui conexões para diversos protocolos, como: SOAP/XML, HTTP/HTTPS, SMTP, RFC, etc.

A camada de UI (user interface) é baseada em BSP e para o AS realizar a comunicação com o browser o mesmo possui um camada chamada ICM (Internet Communication Manager) que faz a interface (HTTP) entre o WebClient e o BSP runtime.
A aplicação WebClient é divida em três camadas: Presentation Layer, Business Layer e Business Engine.

Na camada de apresentação se concentram os códigos HTMLB e JavaScript.
A camada de negócio consiste de duas sub-camadas: BOL (Business Objet Layer) e GenIL (Generic Interaction Layer). Na sub-camada BOL estão concentradas os objetos de negócio, como: parceiro de negócio, ordens de serviço, ordens de venda, etc. GenIL é usada para a transferência de informações (ou dar sustentação) entre a BOL e as APIs da camada de business engine.
Na camada de Business Engine está concentrada toda a base de dados, isto é, o SAP CRM .

Tela da inbox do IC WebClient

BSP
Business Server Pages – a grosso modo é uma mistura de código ABAP Objects, HTMLB (HTML-Business, se preferirem) e JavaScript. Atualmente, BSP implementa o design pattern MVC (Model View Controller).

Assim como o JSP (Java ServerPages), PHP e ASP (Active ServerPages); o BSP é uma aplicação server-side utilizada para o desenvolvimento de aplicações para Web permitindo acesso ao conteúdo do servidor de aplicação.

O PROBLEMA
Para determinados perfis de usuário (como diretores, gerentes e presidente), sempre que a data de SLA do item de uma determinada ordem de serviço for alterada, a mesma deverá automaticamente ser refletida na data de SLA do cabeçalho e, consequentemente, o flag de sinalização da proximidade da data de vencimento deverá ser atualizado. Pois, atualmente, as datas de SLA do item estão sendo alteradas, mas a mesma não é refletida no cabeçalho (campo de vencimento).

A solução encontrada para este problema foi, primeiramente, criar uma view "Z" (qual ABAPer que nunca fez uma cópia do standard? Rs! Neste caso, isto é SAP recommendation), no meu caso, ZSrvTHead.

View ZSrvTHead

View de SLA do item.

Depois de criados, é necessário criar os atributos necessários para carregar dentro do mesmo contexto as datas de item e cabeçalho.
Lembrando que os campos no CRM são do tipo timestamp.

Com os atributos e setters e getters criados, é necessário instanciarmos os objetos do cotrollers e models, como: CuCoBT, Collection Wrapper e BOL Entity.
Isto será necessário para a implementação dos métodos que serão responsáveis por buscar os dados do item dentro do contexto (runtime).

Em runtime temos acesso ao conteúdo do cabeçalho do service ticket, então, para obtermos os dados do item será necessário a navegação por entre os elementos BOL. Isto é, os dados do item estão dentro do componente BTDate e, como temos acesso ao cabeçalho (BTAdminH), criar-se a navegação da seguinte forma:

Usaremos o controler para executar o CuCoBT para termos acesso ao item (BTAdminI).

CuCoBT = ClassController->get_custom_controller( controller_id = 'CuCoBT' ).

Posteriormente, chegaremos até o BTDate:

BTAdminI -> (rel) BTItemDatesSet -> BTDatesSet -> (rel) BTDatesAll -> BTDate

Com o objeto BTDate localizado, basta buscar o campo timestamp do item.

Após concluir os métodos para buscar a data do item, será necessário programarmos os métodos de atualização da data/hora do cabeçalho.

Para isso, usando a mesma linha de raciocínio para navegarmos até o BTDate do item, devemos encontrar o objeto BTDate do cabeçalho.

Dica:

BTAdminH -> (rel) BTHeaderDatesSet -> BTDatesSet -> (rel) BTDatesAll -> BTDate

Mas, uma aviso importante, para alterarmos a data de vencimento é necessário localizar o BTDate cujo campo APPT_TYPE do registro seja igual a SRV_CUST_END, pois a tabela também possui registros com valores iguais a SRV_CUST_BEG, ORDERACTUAL, etc. Meu caso!

Após localizar o BTDate correto, converter a nova data/hora em formato adequado, comparar com a data antiga e, caso sejam diferentes, atualizá-la (óbvio).

object = entity_col->set_property( ‘TIMESTAMP_FIELD’ ).

Esta implementaçã será feita para os campos FROM e TO do objeto BTDate.

Com os getters/setters implementados, temos que escolher o ponto onde o processo será iniciado.
Então, crie uma redefinição do evento do botão da GRAVAR da view "Z", EH_ONSAVE.

Instancie um objeto da classe do nó de contexto responsável pelos atributos e métodos criados anteriormente. E, neste objeto foi carregado os dados do cabeçalho (BTAdminH).

Depois de carregados os dados do cabeçalho no objeto, chamar o método GET do item do service ticket.
Posteriormente, chamar o método SET para a atualização da data do cabeçalho.

Depois de atualizados os dados do cabeçalho, chamar o método SAVE do controlador CuCoBT.

Fim do problema.

Espero ter ajudado seja na resolução do problema ou na agregação de conhecimento.

Até a próxima…

En estos últimos días he tenido un problema bastante curioso. Era una aplicación multi-hilo (50 hilos simultáneos de ejecución) que realizaba peticiones HTTP Get a distintos servidores. El problema es que se presentaban bastante timeouts en las peticiones HTTP y sin embargo el servidor estaba en un correcto funcionamiento.

El código era algo como esto:

public static string Get2(string uri)
{
    string responseData = string.Empty;
    WebRequest request = WebRequest.Create(uri) as HttpWebRequest;
    request.Method = "GET";
    request.Timeout = 35000;
    using(HttpWebResponse response = (HttpWebResponse)request.GetResponse() as HttpWebResponse)
    {
        using(Stream dataStream = response.GetResponseStream ())
        {
            using(StreamReader reader = new StreamReader (dataStream))
            {
                responseData = reader.ReadToEnd();
            }
        }
    }
    return responseData;
}

Tras buscar, encontré bastantes links y este bastante interesante.

El problema era que .NET sólo permite 2 conexiones HTTP a cada sitio, tal y como establece el el protocolo HTTP1.1. Para aumentar este número, lo único que hace falta es incluir esta configuración en el app.config o web.config.

<connectionManagement>
<addaddress="*"maxconnection="40"/>
</connectionManagement>

Neste artigo será mostrado como carregar uma dll de usercontrol dinamicamente em Silverlight. Pois bem, o motivo que me inspirou a fazer este exemplo foi um amigo meu do MSN que insistiu para ajudá-lo a carregar animações dinâmicas vindo externamente. Além disso, percebi um big fator, às vezes podemos ter uma aplicação grande, e como sabemos, o Silverlight cria o arquivo XAP, que é um Zip disfarçado, e fazemos o download deste santo arquivo. Neste ponto, você pode imaginar que, se seu arquivo Xap for grande o usuário irá ficar extremamente zangado e desistir de ver sua aplicação, não é mesmo?. Então o que devemos fazer ?

                Simplesmente devemos é quebrar nossa aplicação em partes, ou seja, criar class Library e usar algumas classes importantes para isso.  Vamos ver então:

Passo 1.

                Crie uma class Library, indo em VS: File -> New->Project. Em Project types escolha silverlight e em Templates Silverlight Class Library. Forneça um nome, tal como: Bola e pressione ok.

                Na solution Explorer, botão direito e adicione um novo item, selecione Silverlight User Control, forneça o nome de Bola.Xaml  e pressione Ok.   

                Para este demo eu criei uma animação, e chamei no meu construtor, como segue:

<UserControl x:Class=”Bola.Bola”

    xmlns=”http://schemas.microsoft.com/winfx/2006/xaml/presentation”

    xmlns:x=”http://schemas.microsoft.com/winfx/2006/xaml”

    Width=”400″ Height=”300″>

                <UserControl.Resources>

                               <Storyboard x:Name=”SbBola”>

                                               <DoubleAnimationUsingKeyFrames BeginTime=”00:00:00“ Storyboard.TargetName=”ellipse” Storyboard.TargetProperty=”(UIElement.RenderTransform).(TransformGroup.Children)[3].(TranslateTransform.X)”>

                                                               <SplineDoubleKeyFrame KeyTime=”00:00:01.2000000″ Value=”64″/>

                                               </DoubleAnimationUsingKeyFrames>

                                               <DoubleAnimationUsingKeyFrames BeginTime=”00:00:00“ Storyboard.TargetName=”ellipse” Storyboard.TargetProperty=”(UIElement.RenderTransform).(TransformGroup.Children)[3].(TranslateTransform.Y)”>

                                                               <SplineDoubleKeyFrame KeyTime=”00:00:01.2000000″ Value=”160″/>

                                               </DoubleAnimationUsingKeyFrames>

                               </Storyboard>

                </UserControl.Resources>

    <Grid x:Name=”LayoutRoot” Background=”White”>

                <Ellipse Height=”80″ HorizontalAlignment=”Left” Margin=”24,48,0,0″ VerticalAlignment=”Top” Width=”80″ Fill=”#FF9B2B2B” Stroke=”#FF000000″ RenderTransformOrigin=”0.5,0.5″ x:Name=”ellipse”>

                               <Ellipse.RenderTransform>

                                               <TransformGroup>

                                                               <ScaleTransform/>

                                                               <SkewTransform/>

                                                               <RotateTransform/>

                                                               <TranslateTransform/>

                                               </TransformGroup>

                               </Ellipse.RenderTransform>

                </Ellipse>

    </Grid>

</UserControl>

Na parte do CodeBehind :

namespace Bola

{

    public partial class Bola : UserControl

    {

        public Bola()

        {

            InitializeComponent();

            SbBola.Begin();

        }

    }

}

Bem até ai nada demais. 

Agora compile para gerar a dll.

Passo 2

Agora, o que precisamos fazer é criar um projeto Silverlight e adicionar a dll, tal como “Bola.dll” dentro da Aplicação Asp.Net, conforme Figura 1. Eu forneci o nome para o meu projeto de LoadAnimation. 

Passo 3

O código seguinte mostra como carregar e mostrar o controle silverlight, ou seja, a dll.

Algumas explicações do código:

• Adicione o seguinte namespace, using System.Reflection, que irá permitir pegar o Assembly do objeto.
• Usar WebClient  para abrir uma leitura assíncrona da dll.
• Pegar o caminho absoluto da dll.
• Quando o download é terminado, o assembly é carregado e uma instancia do controle é criada.
• Finalmente, é adicionado dentro do Layoutroot.

[Code] 

 namespace LoadAnimation

{

    public partial class Page : UserControl

    {

        //assembly para ler a dll

        private Assembly _assembly;

        public Page()

        {

            InitializeComponent();

            Load();

        }

        private void Load()

        {

            //puxar o arquivo da Aplicacao Asp.Net, precisa ser assincrono

            WebClient down = new WebClient();

            down.OpenReadCompleted += new OpenReadCompletedEventHandler(download_OpenReadCompleted);

            string classe = “Bola.dll”;

            string absoluteUri = System.Windows.Application.Current.Host.Source.AbsoluteUri;

            string caminho = absoluteUri.Substring(0, absoluteUri.IndexOf(“ClientBin”)) + classe;

            //assincrono

            down.OpenReadAsync(new Uri(caminho, UriKind.Absolute));

        }

        //Ao terminar de ler é carregada na aplicacao

        void download_OpenReadCompleted(object sender, OpenReadCompletedEventArgs e)

        {

            AssemblyPart assemblyPart = new AssemblyPart();

            _assembly = assemblyPart.Load(e.Result);

            //Bola.Bola, esta é minha classe

            UserControl control = (UserControl)_assembly.CreateInstance(“Bola.Bola”);

            LayoutRoot.Children.Add(control);

        }

    }

}

Muito obrigada,

Flávia Moreira

Unable to save .pdf files from Adobe Acrobat Reader to the Sharepoint Document Library.  As per http://support.microsoft.com/kb/265867/en-us this is a known issue and the workaround is mentioned in the Microsoft KB article.  During troubleshooting it is discovered there is another way this can be done.

For this we first need to ensure that WebDav is enabled in Web Service Extensions in the IIS.  Also make sure that the WebClient service is running in the Services.msc in Windows Server 2003.  Once we have this done, we add a network place using the UNC path.  For instance, if your document library url is http://sharepoint/documents then type in the path as \\sharepoint\documents and this would ask for your valid credentials.  Once you enter your credentials, a network share folder would be created in your My Network Places.  Now you can open your favorite .pdf files in the Adobe Acrobar Reader and save a copy of it directly into the Sharepoint Document Library using the netwrok share in My Network Places.

Note:  The above procedure can be used only to create network shares for folder in the Sharepoint site running on port 80 only.  This is because Web Folders are created for other port numbers and as per the above quoted Microsoft KB article, it is a known issue that .pdf files cannot be saved.  This applies to all non Office applications like .jpg, .rtf etc…

“Screen scraping” means grab the contents of a web page. ASP.NET makes scraping easy. You can grab the contents of any site and then use that contents according to your needs.

For example, you can grab the contents of stock exchange site and then extract useful information from that one. We will remain at very basic level in this article. In future we will move to some useful practical knowledge.

Scraping through ASP.NET:

You have to include following name spaces

1. System.Net (For WebClient)      2. System.Text (For UTF8Encoding)

// make object of WebClient.

WebClient MyClient = new WebClient();

// the url of site whose contents you want to grab.
string Url = “http://www.google.com.pk”

// to get string, as downloaded data returns byte[]
UTF8Encoding MyEncoding = new UTF8Encoding();

// Calling DownloadData function. It will return all contents inform of byte[] array. You can get the                    string by calling GetString method of UTF8Encoding object
string str = MyEncoding.GetString(MyClient.DownloadData(Url));

In this string you have all the contents. you can extract any thing from it.

Enjoy.

Ok, SkyDrive is still a baby, and personally I’ve used other services providing file space in the cloud and enjoyed them a little better. But this post isn’t about if SkyDrive is good or bad, it’s just about a missing feature that is very painful. Someone wanted to share some fotos, uploaded them to SkyDrive and all I wanted was to download them all to my PC. Tough look, you can click on each and every image to get to the preview page, where you click on the preview picture to then finally get at the actual picture. Multiply that by about 100. I have better things to do than waste my time on that.

So a Dev does what he does best, fires up Visual Studio 2008 and hacks away (did I just say I had something better to do – well I lied partially, but before I go off to do that, there is always time for some good ol’ C#).

I’ve posted it here not as a finished utility (there are no binaries) but as a small sample. Using WebClients, RegEx and some other stuff it downloads the list page of the SkyDrive folder, fetches the preview page and then downloads the actual image to a folder on the hard disk. Not really rocket science and of course there are a few quirks (no real error handling for example), but it’s just a sample. Feel free to extend as you wish, don’t blame me if it starts downloading Gigabytes of files overnight, because you accidentally crawled a HoneyPot. (And yes, it only downloads jpgs at the moment. I didn’t need any other types.)

May those SkyDrive bytes be with you…

/**********************************************************************************
*
* Example Application for crawling web pages and downloading images.
*
* This code works if you pass in a SkyDrive Folder Url (http://…. /browse.aspx/…)
* and will download any jpg images it finds in there.
*
* Permission to use, copy, modify, distribute and sell this software and its
* documentation for any purpose is hereby granted without fee.
* I make no representations about the suitability of this software for any purpose.
* It is provided “as is” without express or implied warranty.
*
* Alex Duggleby – 24.05.08 – V0.9 – http://alexduggleby.com
*
**********************************************************************************/
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Net;
using System.Text.RegularExpressions;
using System.IO;
using System.Web;
using System.ComponentModel;

namespace Tools.SkyDrive.DownloadAll
{
class Program
{
// Used for tracking how many items we have left
private static int _wcInnerCount = 0;
private static int _wcInnerCompleted = 0;

// We have to start somewhere
private static Uri _uriStart;

// Work we have already done
private readonly static List _urisCrawled = new List();
private readonly static List _imagesDownloaded = new List();

// Download images to?
private readonly static DirectoryInfo _diDownloadTo = new DirectoryInfo(Path.Combine(Path.Combine(System.Environment.GetFolderPath(Environment.SpecialFolder.Personal), “Downloads”),”Images”));

// This finds urls in the page
private readonly static Regex _regexUrl = new Regex(“href\\s*=\\s*(?:(?:\\\”(?[^\\\"]*)\\\”)|(?[^\\s]* ))”);

// This finds the open url in the image page
private readonly static Regex _regexUrlOpen = new Regex(“href\\s*=\\s*(?:(?:\\\”(?[^\\\"]*)\\\”)|(?[^\\s]*)) title=\\\”Open\\\”");

///

try
{
// First parameter is url
_uriStart = new Uri(args[0]);
}
catch (Exception _ex)
{
Console.WriteLine(“Invalid Url. ” + _ex.Message);
return;
}

// Make sure download directory exists
if (!_diDownloadTo.Exists) _diDownloadTo.Create();

using (WebClient _wc = new WebClient())
{
// This is the index with all the images
string _pageContents = _wc.DownloadString(_uriStart);

// Each image has a preview page, so we get the url to that, before we get the url to the actual image
foreach (Match _matchUrlToImagePage
in _regexUrl.Matches(_pageContents))
{
Uri _uriToImagePage =
new Uri(_uriStart, HttpUtility.HtmlDecode(_matchUrlToImagePage.Groups["url"].Value));

CrawlPreviewPage(_uriToImagePage);
}
}

// Wait for the async web clients to complete…
while (_wcInnerCompleted < _wcInnerCount)
{
Console.WriteLine(“Wait for images to complete…”);
Console.ReadLine();
}

Console.WriteLine(“Should be finished!”);
Console.ReadLine();
}

///

if (uriToImagePage.ToString().ToLower().EndsWith(“.jpg”))
{
string _pageContents = _wc.DownloadString(uriToImagePage);

// Find the image we want to download… There should be
// only one link with title=”Open” in it.
foreach (Match _matchImage in _regexUrlOpen.Matches(_pageContents))
{
Uri _uriToImage = new Uri(_matchImage.Groups["url"].Value);

DownloadImage(_uriToImage);
}
}
}
}
}

///

if (!_imagesDownloaded.Contains(uriToImage.ToString()))
{
_imagesDownloaded.Add(uriToImage.ToString());
string _lowerUrl = uriToImage.ToString().ToLower();

// Simple checking
if (_lowerUrl.EndsWith(“.jpg”) &&
(!_lowerUrl.Contains(“browse”)) &&
(!_lowerUrl.Contains(“self”)))
{
// HtmlDecode here because some urls have encoded characters
string _localFilename = HttpUtility.HtmlDecode(
uriToImage.Segments[uriToImage.Segments.Length - 1]);

// Create a valid local filename
Path.GetInvalidPathChars().ToList().ForEach(
c => _localFilename = _localFilename.Replace(c, ‘_’));

Console.Write(“Downloading {0}…{1}”, _localFilename, Environment.NewLine);

// Create a seperate web client for each image (uses async, and you can’t
// issue two downloads at the same time for the same client). Of course
// here we should be using some kind of pooling but this is the quickest
// way to do it.
using (WebClient _wcInner = new WebClient())
{
_wcInnerCount++;
_wcInner.DownloadFileAsync(uriToImage, Path.Combine(_diDownloadTo.ToString(), _localFilename));
_wcInner.DownloadFileCompleted += new AsyncCompletedEventHandler(_wcInner_DownloadFileCompleted);
}
}
}
}

// Is fired when a download complete. We output status and check if we are finished!
private static void _wcInner_DownloadFileCompleted(object sender, AsyncCompletedEventArgs e)
{
// Increase the completed counter
_wcInnerCompleted++;

// Ok, we could do some more extensive checking, this could trigger
// even if there are still items to download… but hey, it’s just a
// quick utility!
if (_wcInnerCompleted == _wcInnerCount)
{
Console.WriteLine(“{0}{1}{2}”, Environment.NewLine, “Finished all files!”, Environment.NewLine);
Console.ReadLine();
}
else
{
Console.WriteLine(“File {0} of {1} completed!”, _wcInnerCompleted, _wcInnerCount);
}
}
}
}