First, download the actual non-aplha release of LINQ to XSD and extract it.  You will need to have .NET 3.5 installed to run LINQ to XSD.  The project itself is a Visual Studio 2010, .NET 4.0 project.  You should be able to change it to .NET 3.5 with no problems.  I’m not sure about using .NET 2.0.   I still assume these steps should work with Mapguide Open Source 2.2 given its pretty much the same thing as MGE 2011.  Please let me know if you try it and that is not the case.

Building the classes

Last time around we had to mess around with Visual Studio projects, building temporary code, finding and extracting that code from temporary files.  This time around, we’re given a nice little executable that can be run in a batch file.   I’ve posted a copy of my batch file below but it was simply made using a dir /b > CreateMapguideNetObjects.cmd in the Mapguide server schema folder (which by default is to c:\Program Files\Autodesk\MapGuideEnterprise2011\Server\Schema) .  I then edited that file with a text editor that support macros and removed the following schema files:

• FdoProviderCapabilities-1.0.0.xsd
• LoadProcedure-1.0.0.xsd
• LoadProcedure-1.1.0.xsd
• LayerDefinition-1.0.0.xsd
• LayerDefinition-1.1.0.xsd
• LayerDefinition-1.2.0.xsd
• SiteInformation-1.0.0.xsd
• SiteVersion-1.0.0.xsd
• SymbolDefinition-1.0.0.xsd
• WebLayout-1.0.0.xsd

These files are deprecated object definitions from previous Mapguide releases.  In the end each line in the batch file looks a little like this:

LinqToXsd ApplicationDefinition-1.0.0.xsd /filename:ApplicationDefinition.cs

Executing the batch file will create a number of C# files containing appropriately named classes.

Setting up your project

At this point you should be able to fire up Visual Studio and create a new C# project.   If you have an existing project from a previous version of Mapguide its good to start fresh.  Also, don’t try this in an existing solution that is dependant on the project that contains the Mapguide objects.  This will just result in a bunch of extra screwing around to avoid compiler errors from missing code during the process.    This time around its a lot easier.  Create a new DLL project.  Add a reference to the Xml.Schema.Linq.dll file that was included with LinqToXsd.exe.   Add a reference to the following Mapguide dlls:

• OSGeo.Mapguide.Foundation
• OSGeo.Mapguide.Geometry
• OSGeo.Mapguide.MapguideCommon
• OSGeo.Mapguide.PlatformBase
• OSGeo.Mapguide.Web

You’ll also need to ensure that the appropriate unmanaged assemblies are available in the calling applications bin folder when you go to run this stuff. 

Next place all of the generated C# files in the new project.  Once you build you’re going to see a ton of errors.  We’ll clean those up.

To resolve a lot of these errors I did a global search and replace on “global::” and replaced it with nothing.  Also, I wrapped each class in a unique wrapper classes to prevent duplicate type errors.  Finally, the case issue with DataType was still an issue – and I resolved it by changing the case on DataType to be Datatype:

   1: public string DataType {

   2: public string Datatype {

For a complete view of the changes I made do a file compare between the code in the attached zip file and your newly generated code.

<insert 3 week gap here, wherein I had a hell of a time getting MGE 2011 installed and properly licensed so that I could actually test this newly created code, oh and I also went on vacation for a few days too =)/>

And back.  The best part about this is that it seems ALL my old RADE code just worked with upgraded basic layouts after changing it to look for the object definitions within the new wrapper classes.   Now I just need to add support for flexible layouts and we’ll be laughing.

The Code

Please note – I didn’t rebase the code.  It’s all in the RADE.MGE namespace.  If you would like to use it, feel free to re-base it – or just leave it as is.  Also, if you create any unit tests or enhancements and would like to share them – please feel free!  

As I update the project myself, or receive updates I will update this file.

. 

Finally, the example code I posted long ago will still work with this new project.  Perhaps one day, I’ll post some fancy new code in C# =)

Hope this comes in handy, I welcome your comments.

Came across Automapper.   This is a perfect match for my needs:

AutoMapper uses a fluent configuration API to define an object-object mapping strategy.   AutoMapper uses a convention-based matching algorithm to match up source to destination values. Currently, AutoMapper is geared towards model projection scenarios to flatten complex object models to DTOs and other simple objects, whose design is better suited for serialization, communication, messaging, or simply an anti-corruption layer between the domain and application layer.

It was quick to setup, and painless to use.  First define your object to object mappings:

   1: Mapper.CreateMap<RADE.BO.Domain.Application, RADE.BO.Entity.Application>();

   1: /// <summary>

   2: /// Convert a DTOApplication to abnd 

   3: /// </summary>

   4: /// <param name="app"></param>

   5: /// <returns></returns>

   6: public static RADE.BO.Domain.Application Convert(Application app)

   7: {

   8:   if (app == null) return null;

   9:   LoadDtoMappings();

  10:   return AutoMapper.Mapper.Map<Application, RADE.BO.Domain.Application>(app);

  11: }

  12:  

  13: /// <summary>

  14: /// Convert a list of domain Application objects to DTO Application objects

  15: /// </summary>

  16: /// <param name="apps">List of domain Application objects</param>

  17: /// <returns>List of DTO Application objects</returns>

  18: public static List<Application> Convert(List<RADE.BO.Domain.Application> apps)

  19: {

  20:     LoadDtoMappings();

  21:     return apps.Select(AutoMapper.Mapper.Map<RADE.BO.Domain.Application, Application>).ToList();

  22: }

It’s important to note that if your source object contains nested objects – a mapping for each nested object must also be defined.  This goes on recursively through nested objects.  My domain Application object contains some numerous sub objects – Map, Layer etc.  In this case rather than establish mappings for these child objects – I removed these objects from my DTO as the client consuming these objects would never need that data.  This also keeps the size of any data possibly being serialized down.   Automapper dealt with this automatically.

I’ve only scratched the surface of what Automapper can do.  Next step is to define unit test coverage using the configuration validation, but it’s sleep time now… =)

I’ve upgraded the solution to 2010.  I’ve done all the refactoring and finished my work.  Everything tests out great in both unit tests and “developers computer” tests.  Just need to get the installer updated and get the pesky “other peoples computers” thing out of the way. 

I’ve updated the pre-requisites on my installer to include .NET 4.0 – rebuilt the installer and boom. 

Error 1001.  Exception occurred while initializing the installation.  System.BadImageFormatException: Could not load file or assembly ‘file:///C:\Program Files\Landor\ … \RADE.Connect.Service.exe’ or one of its dependencies.  This assembly is built by a runtime newer than the currently loaded runtime and cannot be loaded.

After much mucking around I found the hidden property that I needed to change.  I needed to change the .NET version of the .NET Launch Condition.  To find this gem right click on the deployment project in the solution explorer and choose View\Launch Conditions.  Next right click on the “.NET Framework” under “Launch Conditions” and choose “properties”.  In the properties dialog change the Version to be “.NET Framework 4”.

Update 4/20/2010

Don’t forget to edit the .NET version for both Release and Debug configurations.

http://msdn.microsoft.com

wooo!

In the meantime, The Register has a good write up on some of the new goodies coming in an hour or so.  That will learn me for getting up early.  Oh well, time for another cap!

System.InvalidCastException: Unable to cast object of type ‘Oracle.DataAccess.Client.OracleConnection’ to type ‘System.Data.Common.DbConnection’

This was one of those awesome “doesn’t happen on any of my machine” errors.  After some mucking around we determined that the client machine was using the Oracle 10g  R1 client.  The machines and VMs here we used for testing were all running either 11g, or 10g R2.  Doh!

The simple resolution to this was to modify the NHhibernate config and add the following property:

   1: <property name="hbm2ddl.keywords">none</property>

Depending on your underlying databases and mappings this could cause problems with your code project.  Hopefully you’re not using reserved words, or bad Oracle syntax =).

Under the hood, the problem seems to be that the Oracle10gDialect does not provide an implementation of IDataBaseSchema for 10gR1.   Fabio Maulo has provided some sample code and the steps on how to create the appropriate metadata for your database and submit it to the NHibernate project for inclusion.

The NUnit test engine is great.   I didn’t realize this was a ReSharper tool at first.  Now, I can’t live without it.

The localization plugin is going to save me days of effort on an upcoming action item. 

The class/code generating tools are indispensible – especially if you create a lot of classes (e.g NHibernate POCO’s).

The refactorings have been helping me slowly increase the legibility of my code – and also helping me build better coding habits.  Check out the comparison between version 5.0 and 4.5.

If you’ve never tried ReSharper – now is the time.  only I could make the time to properly learn all the different tools available…

NH used to assume Castle – but now it needs to be configured.  In addition to the Castly proxy, there is a new LinFu proxy available too.   Don’t bother with NHibernate the release notes! , Davy Brion has it all for you here =). 

I opted to go with the Castly Dynamic Proxy – I’m not in a position to do any performance tests between– Castle has been around a long time and has a pretty big user base.  Good post on stackoverflow on this subject.  However, LinFu would have won if I was basing my choice on name alone…

After a bit of fiddling, I’m back up and running with NH 2.1 and shiny new validation attributes.   

mmmm validation…

1. 1. performance and memory usage improvements
2. 2. Solution-wide code inspections (tons of cool tools to find unused params, privates, etc)
3. 3. VB9 support (too bad I’ve been busy moving to C# =])
4. 4. Native MSTest Support (I much prefer NUnit though)

Check it out.  ReSharper is one of the tools I use every day and couldn’t live without…One more day and I’ll be done in Visual Lisp and can actually try out the new release a little more =)

   1: #header {width:100%; height:50px;    background: #7D7D7D url(images/header.gif) left top no-repeat;}

When the master page is used from a page deeper than the site root, the paths in the CSS become invalid.  To work around this I made my CSS dynamic!  I added a new page to my project in the CSS folder.  I called this page allcss.aspx.css.   Open up this new aspx file and remove all the html.  Leave only the page header definition:

   1: <%@ Page Language="C#" AutoEventWireup="true" CodeBehind="all.css.aspx.cs" Inherits="landorIS.com.css.allcss" ContentType="text/css" %>

Note that you may need to update your classnames and stuff depending on how you name the file.  I wanted my file name to reflect the fact that it was a CSS file – but it’s still needed to end with .aspx to ensure the file would be properly parsed by the ASP.NET engine without having to mess with the extension mapping on the web server.  This resulted in Visual Studio making some illegal class names.

Also be sure that you add the ContentType attribute to the @Page definition, and set it to “text/css”, or you will get an error in some browsers stating that the Mime Type “text/html” is not “text/css”.  Nice of Firefox to do a little validation on the content.

At this point you can add your CSS to the file and use <%%> tags or codebehind to manipulate the values.  Now the CSS that was causing problems with the images looks like this:

   1: background:url('<%=ResolveUrl("~/img/header-bg.png")%>') no-repeat 0 0;

Now when the CSS is requested by the browser and rendered out, the URL to the image file will be properly displayed as /RootFoolder/img/header-bg.png.  This is a very simple modification, but one could go a little more crazy and do something like  detect browser make/model and send out browser specific css in pre-defined literals, or possibly add authentication checks. 

This complexity resulted in infrequent builds and as a result the code was not tested as well as it should have been.  Not to mention, I’d get really frustrated and even pissed off on build days.   It took me a while to find a system that really made me happy.  On a whim one night I downloaded Automated Build Studio (ABS) and gave it a shot.  After installing ABS on a new virtual machine that would become our dedicated build system I fired it up.  I was pretty impressed with the wide range of actions and  tools it supported.  In our environment we had several must have items:

• 1. SourceGear Vault
• 2. Visual Studio 2008
• 3. Oracle/SQL Server
• 4. Installshield 2009

ABS worked with all of them.  In addition to the key requirements I had – it supported a plethora of other operations.

First thing I did was sketch out a rough outline of what was involved in creating a build.  Get latest source code, compile, delete temporary files, etc etc etc.  I started selecting various options from the ABS and fleshing out my macro.  Within a couple of hours, I had a sweet macro in place.  My macro started to look like this:

• Getting the latest code from Vault
• Compile all assemblies
• Delete the temporary files
• Build the needed release folder structure
• Move the new files into the needed structure
• Execute Remotesoft Protector on all assemblies
• Increment the build version, and set the version on all assemblies
• Create a label of the build in Vault
• Build a zip of the release
• Build the Installshield installer
• Connect to the build FTP and upload the new build

The best part – all of this was occurring in a 100% reliable fashion, oh and it took minutes for a full build to occur.  Over time, this project was tweaked, and copied.  It evolved as I learned more about ABS – or thought up new requirements.  Take FullCircle for example, it has evolved into three editions, sharing a similar yet different set of source files.  ABS lets me rebuild all three editions with the click of a button – in minutes.   Long long ago, in the before times something like this would have taken days, and years off my life.

I’ve only scratched the surface of what I’ve been able to do with ABS – some other cool features worth mentioning is that builds can be scheduled to run, there is a web application for viewing build history as well as kicking off builds.  It has a scripting language, and the support I’ve received is fantastic.  The price of the product makes the decision to buy ABS a non issue.

Man, now I’m starting to sound like a sales guy. =)  Seriously though, check out Automated Build Studio if you develop boxed, or even custom software.

In long – for a number of reasons there was not a strait line between upgrading my machine to SP1 and noticing the problem.  This made diagnosing the problem significantly more complicated because I had no idea when it actually started.  At the time, I was dealing with x64 and Oracle clients.  This had prompted me to start doing all my dev in an XP virtual machine.  Eventually, the slowless of developing in a VM drove me insane and I just had to get my machine working again.  So I got Oracle dealt with – but that left this annoying error.

It was a complete fluke that I noticed this.  The stars aligned, and I manually compiled some assemblies, instead of pushing the big green button on my build system.  This resulted in my assemblies not being run through the Remotesoft Protector.  All of a sudden the problem went away.  At the time I had been using Protector 2.x which would not support .NET 3.5 SP1 fully. So, either stop using Protector, or upgrade to the latest version which resolves this problem =)

Press enter and you are prompted for the class name.  So Visual Studio has this of course, but this is a small example of some of the time savings you can achieve.  Enter the class name and you start off with an empty class.  Next, lets define a couple of private members.   Let’s run with the following example.

   1: namespace RADE.BO.Domain

   2: {

   3:     public class SampleClass

   4:     {

   5:         private Int32 _ID;

   6:         private String _Description;

   7:         private Int64 _BiggerInt;

   8:     }

   9: }

Simple class.  Three member variables.  Here is where the big time savers come in.  Next, click Alt+Insert again and you will be prompted with the following options:

Choose Properties.  You will now be given the option to select one or many fields, set the access rights, read only and virtual properties as well a bunch of others. 

Execute that and all of your get/sets are defined.  My one complaint is that ReSharper is not maintaining the type on the properties.  My Int32’s become int, and Int64’s become long’s.  Apparently this is slated to be fixed.  

So yes, this sample class is tiny, but the time savings on larger classes, or projects full of classes are significant.  The last feature I’ve been using extensively in this project is the generate constructor tool.  Again, a form is displayed with the defined properties, select the ones you want and boom.  You can create half a dozen different constructors in seconds.  I’ve created a full object model on, roughly 20 different mildly complex objects in less than an hour, around midnight on a Sunday =)

Some of the other things ReSharper does is suggest code cleanup ideas by removing unused directives, easy execution of NUnit tests, improved code completion.  I encourage you to check it out, download a ReSharper trial.

Obfuscation was one of my first answers to this problem.  An obfuscator ships with Visual Studio Pro, free and there are many available on the market.  Obfuscation just didn’t do it for me.  I once helped a customer troubleshoot problems with one of their software solutions from an unnamed vendor using Reflector and walking through the obfuscated code.  This was really a painful experience, it does make it harder to figure out what is going on – but a friend of mine suggested a product that takes code protection one step further.

Hello Remotesoft Protector.  This product is pretty cool.  If you purchase the protector product you will receive three components.  Salamander .NET Decompiler, .NET Obfuscator, and .NET Protector.  Initially I was processing my assemblies with both the obfuscator and the protector.   Now a days, I pretty much only run my assemblies through the protector.

Once you’ve processed an assembly with the protector and you open it up in reflector things are going to look a little different.  Here is a little before and after action for you:

Now lets take a look at the same code, but after being protected:

That’s it.  Protector has made all your code go bye bye =)  What’s happened here?  As I understand it, Protector compiles all your managed .NET code into native code.  So, yes, is it possible to disassemble native binaries.  The difference here is the height of the bar – with plain .NET assemblies even my grand mother could get my code.  Reverse engineering a native assembly is a different story.  If someone with the skill to do that wants your code – well you must be writing some damn fine code.  It would probably be easier for that kind of person to write it from scratch =)

I’ve been working on increasing my Joel Test score lately.  One of my biggies is the one step build for RADE.  That sentence really doesn’t do the task justice.  The first step I’m tackling in the one step build is automating the process of protecting my .NET assemblies.  I could not find any resources on doing with with MSBuild.  Once I get it working, I’ll post some code.

All that said, I highly recommend you check out Protector if code protection is your thing.  The price is a little bit steep at 1899$ for 1-5 developers – but how much money have you invested in that one little DLL or EXE file?