Using Microsoft SQL Server Management Studio with Subversion

Features include:

Easy to install, easy to configure, easy to use Add solutions, projects and files to Subversion, all from within the IDE Check out, check in, get latest version all without leaving the IDE Revert file changes from within the IDE Provides the option of automatic file locking on check out and unlocking on check in The power of Subversion with the feel of Visual Source Safe View file differences, file status and file history from within the IDE Can use an external difference tool to view file differences

Described below are the steps required to get Agent SVN to work seamlessly with Microsoft SQL Server Management Studio and Subversion.

As an introductory tutorial, it is recommended you follow the steps as described.

Step: 1 Agent SVN Provider Setup

From within the Microsoft SQL Server Management Studio IDE use the Tools, Options menu to bring up the Options dialog and in the Source Control section select Agent SVN as the active provider as shown below:

Step: 2 Configure Agent SVN

Bring up the Agent SVN configuration as shown below:

For this initial testing, define a local Subversion repository that uses the file protocol as shown below:

NOTE: It is highly recommended you start with this minimal configuration since this is by far the simplest Subversion configuration model.

This will give you an opportunity to become familiar with how the plug-in works and at a later point it will be easy to reconfigure Agent SVN to another repository location using a different protocol.

Step: 3 Create a Test SQL Project

To see how Agent SVN works we need something to play with so the next step is to create a simple test project.

From within Microsoft SQL Server Management Studio IDE create a new My Scripts project as shown below:

Create a few SQL Query files in the project folder and the end result should be a project folder that looks something like this:

From within the Microsoft SQL Server Management Studio IDE, add these SQL files to the project using the Solution Explorer and the end result should be a project that looks like this:

Step: 4 Add Solution to Source Control

At this point, we have a new project and we will use Agent SVN to import the project into the Subversion repository (as was defined in step 2).

NOTE: Agent SVN can also handle the case where the project already exists, but in that case, the project will need to be bound to the Subversion repository.

Use the menu shown below to import the project:

Using the repository browser to confirm the project files are now inside the source control repository as shown below:

Step: 5 Patch the Project File

At this stage, the project is in the repository and the project has been bound to the repository, but unfortunately, unlike Visual Studio, the Microsoft SQL Server Management Studio IDE does not seem to remember these binding details.

Because of this bug, the next time you open the project you might see this error message:

What has happened is the IDE has saved the binding details into the solution file but not into the project file. So when it opens the project file it seems surprised to find no binding information and hence the message.

Therefore, if you select the option that is circled, the IDE will again have the correct binding information that it needs and so it will continue one happily.

However, it will not have saved this information, so if you close and reopen the solution the same message will appear.

Now you as the user have the choice of putting up with this annoyance or alternatively you can do what the IDE should have done and save the binding information inside the project file.

To do this just open up the project file shown below inside of Notepad:

With the IDE not running, carefully add the SccProjectName, SccAuxPath, SccLocalPath and SccProvider details to the top of this file as shown below:

<?xml version="1.0"?>
<SqlWorkbenchSqlProject xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema" Name="SQL Server Scripts1">
<SccProjectName>My Scripts</SccProjectName>
<SccAuxPath />
<SccLocalPath>.</SccLocalPath>
<SccProvider>MSSCCI:Agent SVN</SccProvider>
<Items>
<LogicalFolder Name="Connections" Type="2">
....
...
....


Save the file and commit the changes back into the repository. Restart the IDE and open the project and this time the error message will not get displayed.

With the project now fully configured, you can now use the check out, check in, revert, file compare commands (or any of the other source control commands) to manage your SQL file.