We use the repository pattern to enable:

• Testability. Using the pattern we can create stubs that can replace the real data access objects. This can help us to test our business logic without concerning what the data access is doing.
• Dependency Injection. Using the pattern we can use DI containers to inject the relevant object that we want to use in the code.
  http://blog.fossmo.net/post/Murder-mysteries-and-Dependency-Injection.aspx
• Abstraction. Using the pattern we create an abstraction above our data access functionality. This abstraction can help us when we want to change the implementation of the data access without affecting our business logic code. For example, I had to change implementation of data access with a call to a web service. Using the pattern I only needed to change the object that I used and that is  it.
  http://blog.fossmo.net/post/Inversion-of-Control-Containers.aspx

Repository Pattern Resources:

• http://channel9.msdn.com/Series/The-Full-Stack/The-Full-Stack-Part-3-Building-a-Repository-using-TDD
• http://visualstudiomagazine.com/Articles/2009/09/01/Eliminate-Database-Dependencies-in-Test-Driven-Development.aspx
• http://codingsolutions.blogspot.com/2010/07/tdd-kata-for-ddd-building-simple-domain.html
• http://www.pearsonhighered.com/assets/hip/us/hip_us_pearsonhighered/samplechapter/0321268202.pdf

http://www.arrangeactassert.com/how-to-unit-test-asp-net-mvc-controllers/

http://www.asp.net/mvc/tutorials/older-versions/unit-testing/creating-unit-tests-for-asp-net-mvc-applications-cs

http://msdn.microsoft.com/en-us/magazine/dd942838.aspx

http://fpnotebook.wordpress.com/2011/08/14/asp-net-mvc-entity-framework-unit-testing-mocks-and-microsoft/

http://www.infoq.com/news/2012/03/aspnet-unit-test

http://dotnetslackers.com/articles/aspnet/ASPNETMVCFrameworkPart2.aspx

http://msdn.microsoft.com/en-us/library/ff847525.aspx

Time to setup all the usings. Basically, we include the usuals plus System.IO for file handling, and System.Data.SqlClient for access to the SQL Server.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Data.SqlClient;
using System.Text.RegularExpressions;
using System.IO;

Next We’re going to create a method for saving the bytes to a file. So our method takes in a string for the file name, and a byte array for the actual file data. Then to be save we put everything in a try catch block. See the comments inside the code for the individual steps, but there is no real magic here.

public static bool ByteArrayToFile(string _FileName, byte[] _ByteArray)
        {
            try
            {
                // Open file for reading
                System.IO.FileStream _FileStream = new System.IO.FileStream(_FileName, System.IO.FileMode.Create, System.IO.FileAccess.Write);

                // Writes a block of bytes to this stream using data from a byte array.
                _FileStream.Write(_ByteArray, 0, _ByteArray.Length);

                // close file stream
                _FileStream.Close();

                return true;
            }
            catch (Exception _Exception)
            {
                // Error
                Console.WriteLine("Exception caught in process: {0}", _Exception.ToString());
            }

            // error occured, return false
            return false;
        }

Okay, let’s move into the main part of the program. This is broken into two parts the first handles the nonemail attachments, and the second part handles the email attachments. We start by setting up the SQL connection string, and the query.

        static void Main(string[] args)
        {
            string connString = "Server=servername;Database=Database_MSCRM;User Id=username;password=password";
            string queryNonEmailAttachments = "SELECT DocumentBody, ObjectId, FileName, CONVERT(varchar(255), ObjectID) + '_' + CONVERT(varchar(255), Filename) AS SaveAsFileName FROM FilteredAnnotationUS WHERE IsDocument = 1 AND (objecttypecode = 1 or objecttypecode = 2 or objecttypecode = 3 or objecttypecode = 4 or objecttypecode = 4212 or objecttypecode = 4201 or objecttypecode = 4210)";

Next we setup a using block and define a SqlCommand.

            using (SqlConnection connection = new SqlConnection(connString))
            {
                SqlCommand command = new SqlCommand(queryNonEmailAttachments, connection);

Now another try catch block is used to work through the database. We open the connection to the SQL Server, set a longer time out to make sure we get all the data and create a reader result object that we can iterate though. Finally we setup the CSV file that is used by SalesForce to tie the attachments by ObjectID to the filename and proper CRM record.

                try
                {
                    connection.Open();
                    command.CommandTimeout = 300000;
                    SqlDataReader reader = command.ExecuteReader();
                    StreamWriter file = new StreamWriter(@"C:\\CRM\\NONEMAIL\\NonEmailMaster.csv", true);

Once we have the data it is time to read through it. A while loop is used to roll through the data. We setup a filename that contains the ObjectID plus the original filename. Next is the magic part of this. The files are stored in latin_1 collation in the SQL database, and we need to get it into UTF8 and decode it from the Base64 encoding it is currently in.

                    int i = 0;
                    while (reader.Read())
                    {
                        string filePathAndName = "C:\\CRM\\NONEMAIL\\" + reader.GetString(3);
                        /*
                         * Converts Latin-1 from the CRM database to UTF8 so we can use Base64 decoding to recreate the original file.
                         */
                        string fileDataString = Encoding.UTF8.GetString(Encoding.GetEncoding(1252).GetBytes(reader.GetString(0)));
                        byte[] fileBytes = Convert.FromBase64String(fileDataString);

Next we write the decrypted byte array to the file using the Method we setup earlier, and append the information required by Salesforce to correlate the CRM entries to the files. Wash, rinse, repeat until we looked through all the records.

                        /*
                         * Writes the original file to disk
                         */
                        bool fileWritten = ByteArrayToFile(filePathAndName, fileBytes);

                        /*
                         * Adds the record to the CSV reference file.
                         */
                        file.WriteLine("\"{0}\",\"{1}\",\"{2}\"", reader.GetValue(1).ToString(), reader.GetString(2), reader.GetString(3));
                        i++;
                    }

Finally we drop a line about how many records we processed, clean up the reader and close out the CSV file. The catch is there to get any errors, and we’re done with the nonemail attachments.

                    Console.WriteLine("Processed {0} non email records.", i);
                    reader.Close();
                    file.Close();
                }
                catch (Exception ex)
                {
                    Console.WriteLine(ex.Message);
                }

We repeat that exactly changing on the query string for the email attachments. Lastely, we create the C:\CRM\Email and C:\CRM\NonEmail directories to hold all the data. Now run it, and you should see those directories filling up with data.

You can get the whole solution from Github here.

So enough sappy story, and on to the meat of this article.  My overall plan is very simple.  I want to be a software creator for a living in 13 months.  Well that was easy to say, now how do I get started.  The first thing I did was start trying to decide what technology I needed to work on to feed my family.  I knew web development was what I wanted to do so that automatically required CSS, Javascript, HTML, etc; however, there are tons of technology stacks that can be used to surround those.  I’ve done tons of work in PHP over the years so that went to the top of the list.  I had written a tracking application in Rails about a year ago, and .NET is just crazy hot in Nashville.  So those were my three main focuses: PHP, Rails, and .NET.

I thought about how I would pick between the frameworks, and settled on developing the same application in all three languages.  That would let me find the one I liked the most.  Thus was born NetVis the IP tracking application.  I’ve written it in Rails and ASP.NET MVC C#at this point, and it helped me determine what I liked about each of those languages.  I’m starting to write in PHP now.  You might notice my first instinct in this was to start coding, and truthfully that should be the exact starting point for everyone.  Don’t start researching a ton of stuff, pick any language and start coding.  Regardless of what you pick in the end, getting the brain flowing around the concepts is this most important part.

The next thing I did was get involved locally.  I made the decision to sacrifice up to 4 nights a month, and a weekend to plug into the local development community.  In my case, I started attending Agile, .NET, PHP, and Ruby User groups.  We very active ones of all three in Nashville thankfully.  At this groups, I listened intently; however, you and I both know I didn’t understand it all.  The more I listened, the more I researched, and the more I learned.  I also mentally selected a few people that I wanted to meet for coffee or lunch, and I just asked.  I’ve not gotten a single no yet.  Those few moments of listening to a craftsman talk about his trade are critical to how you shape yours.  Everyday I read twitter and blogs, not skim!  I read them when I don’t understand I wikipedia enough to get a basic idea.  Following the leaders in each technology is easy, but follow local people, and engage into their social sphere.

Everyday you can should a study day.  We learn something new, we work through a book, or we watch video; however, we always code!  Some small thing should be built everyday.  For me it is a feature on my IP address management system, or something to get us around a little problem at work.  Just because everyday should be a study day doesn’t mean it is.  You can not neglect your family and friends as you make this time of a career decision as they will be the ones to be there for you on the rough days when it seems like you are making the wrong choice.  So spend time with them, and study around that time.  Pick longer breaks every few weeks (6 in my world), to step away for a weekend and just relax.

Last thought of this article is simply start with the end in mind.  If you want to be a craftsman, start today on doing things right.  For me right now means the following: Test Driven Development, Object Oriented Development, Source Control, Agile Methodologies, Removing Platform Bigotry, Do not Repeat Yourself, and Sharing.  During this learning time is the best time to set the patterns and habits you want to have in your career.  Plus if you screw up with Git and blow up your tiny solution you built to track your Lego collection that is no big deal.  This gives you code that you can take to user groups, and show to people for feedback on more than just the working of your simple app.  Unit testing has a ton of nifty tricks that are best learned by showing others and listening to their feedback.

I wanna thank a few people who have been helpful on my path so far: Cal Evans, David Neal, Pluralsight, TekPub, DHH, Jacques Woodcock, Adam Serbert and Code Year.

So, I’m working on getting my development chops back up to speed after a 6 month hiatus, althought I’ve done some Ruby on Rails and Perl during this time for small things.  Also, learning ASP.NET this time since Nashville is a Microsoft market for development in my mind.  So I signed up for Pluralsight’s training classes, and I’ve worked throught a couple of them and wanted to share my thoughts.

First, I took A Beginner’s Guide to the Microsoft Web Platform.  I thought this class was good at showing off all the different types of projects that you can do in ASP.NET such as web applications, web sites, and MVC.  It took the time to do the same 3 examples in each of the development types, which I thought was a great thing.  The purpose of this class wasn’t to learn syntax or particular tricks.  It was solely to focus on the .NET architecture, and to show off it’s flexibility.

Next up for me was Introduction to ASP.NET 4 WebForms.  This was the first class that was showing me how to really build things, and I’m really enjoying it.  The instructors in the videos are articulate, and see to have a good understanding of what they are talking about.  This class is doing a great job of breaking down all the parts of a modern web application: WCF, AJAX, server and client side controls, and data connectivity.  However, it is a bit slow, and occasionally they just skip ahead in the code example instead of walking through line by line.

Wrapping this up, I have to say so far I feel I’m getting a great product for my money.

Time to setup all the usings. Basically, we include the usuals plus System.IO for file handling, and all the .NET Email Components for all the IMAP Access and Control.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Reflection;
using System.Net;
using System.Net.Sockets;
using System.IO;
using System.Text.RegularExpressions;
using Email.Net.Common.Configurations;
using Email.Net.Common.MessageParts;
using Email.Net.Imap.Responses;
using Email.Net.Imap.Collections;
using Email.Net.Imap;
using Email.Net.Imap.Sequences;

Next we setup the IMAP Client and login by setting:

• the host to which we will be connecting
• username
• password
• IMAP port 993 or Secure IMAP in this case
• the directory that will be used to store the attachments
• Security type

Then we call the log in method.

namespace iSynergyEmailImport
{
    class Program<
    {
        static void Main(string[] args)
        {
            ImapClient client = new Email.Net.Imap.ImapClient();
            client.Host = "host";
            client.Username = "User";
            client.Password = "password";
            client.Port = (ushort)993;
            client.AttachmentDirectory = "PathForAttachments";
            client.SSLInteractionType = EInteractionType.SSLPort;
            CompletionResponse response = client.Login();

Next we setup variables to represent the mailbox and the messages, and check to make sure that the response we got back from the client login was good.

            Mailbox inbox;
            ImapMessage msg;
            CompletionResponse response = client.Login();
            if (response.CompletionResult == ECompletionResponseType.OK)
            {

Next we get a list of Folders that are in the tree and access the folder named inbox. After that for debugging purposes we print out the message count, and load all the message headers into a message collection so we can go through them.

                Console.WriteLine("We're In");
                Mailbox folders = client.GetMailboxTree();
                inbox = Mailbox.Find(folders, "INBOX");
                Console.WriteLine("{0} Messages", client.GetMessageCount(inbox));
                MessageCollection tmp = client.GetAllMessageHeaders(inbox);

Finally we interate through each message, open it and check to see if their are any attachments. If there are any attachments present, we loop through each attachment and save each one to the Directory we specified earlier. However, there is one wrinkle, We have to move the file from it’s temporary file name which is FullFilename to the name that IMAP had it stored as which is the TransferFilename.

                foreach(ImapMessage x in tmp)
                {
                    msg = client.GetMessageText(x.UID, inbox);
                    if (msg.Attachments.Count > 0)
                    {
                        for (int i = 0; i <msg.Attachments.Count; i++)
                        {
                            Attachment attach = msg.Attachments[i];
                            if (attach != null)
                            {
                                Attachment tmpAttach = new Attachment();

                                tmpAttach = client.GetAttachment(inbox, msg, msg.Attachments[i]);
                                string tmpName = tmpAttach.FullFilename.ToString();
                                string realName = client.AttachmentDirectory + msg.Attachments[i].TransferFilename.ToString();
                                File.Move(tmpName, realName);
                                Console.WriteLine("Attachment: {0}", realName);
                            }
                        }
                        ISequence squence = new SequenceNumber(x.UID);
                        CompletionResponse delResponse = client.MarkAsDeleted(squence, inbox);
                        delResponse = client.DeleteMarkedMessages(inbox);
                    }
                }

 public class Subnet
 {
     public int SubnetId { get; set; }
     public int SiteId { get; set; }
     public string Network { get; set; }
     public string Description { get; set; }
     public string Netmask { get; set; }
     public string Gateway { get; set; }
     public int Preference { get; set; }
     public Site Site { get; set; }
     public List Ips { get; set; }
 }

public class Ip
{
    public int IpId { get; set; }
    public int SubnetId { get; set; }
    public string Address { get; set; }
    public string Name { get; set; }
    public string Purpose { get; set; }
    public string Url { get; set; }
    public Subnet Subnet { get; set; }
}

As you can see each subnet contains a list of IP addresses. We can easily access these by including them in our view by putting them in the controller:

public ViewResult Details(int id)
 {
    Subnet subnet = db.Subnets.Include("Ips").Include("Site").Single(s => s.SubnetId == id);
    return View(subnet);
 }

However, if we want to sort those in our view we would typically do the follow:

@foreach (NetVis.Net.Models.Ip ip in Model.Ips.OrderBy(i => i.Address))
 {
    <tr>
        <td>@ip.Address</td>
        <td>@ip.Name</td>
    </tr>
}

That code would produce an address list like:
10.10.4.1
10.10.4.10
10.10.4.2
10.192.6.3
10.2.1.1

Ugh, that is not how we sort IP addresses. They should look like:
10.2.1.1
10.10.4.1
10.10.4.2
10.10.4.10
10.192.6.3

To solve this problem we can use IComparer. We can extend the IComparer class for our needs with IP addresses. So the class below takes two strings, splits them where ever there is a period, cast that as an integer, then compares each part as integers, and returns which one is the lower of the two.

public class IPComparer : IComparer
{
    public int Compare(String a, String b)
    {
        return Enumerable.Zip(a.Split('.'), b.Split('.'), (x, y) => int.Parse(x).CompareTo(int.Parse(y))).FirstOrDefault(i => i != 0);
    }
}

We can use this code by including it below:

@foreach (NetVis.Net.Models.Ip ip in Model.Ips.OrderBy(i => i.Address, new NetVis.Net.Models.IPComparer()))
{
    <tr>
        <td>@ip.Address</td>
        <td>@ip.Name</td>
    </tr>
}

Now we will get a nice properly sorted IP address listing.

Extra special thanks to Jason on StackOverflow

This is lifted from http://stackoverflow.com/questions/4785218/linq-lambda-orderby-delegate-for-liststring-of-ip-addresses

http://penny-arcade.com/patv/episode/so-you-want-to-be-a-developer-part-1

http://blog.pluralsight.com/2012/02/02/build-your-own-screaming-quiet-pc-just-like-pluralsights-dev-boxes/