The content of TestFactTable.xml is as below:

<?xml version="1.0" encoding="utf-8"?>

<Vulcan xmlns="https://tempuri.org/vulcan2.xsd">

    <Fact Name="TestFactTableA" ConnectionName="DataWarehouse">

        <Columns>

            <Column Name="ID" Type="INT64"/>

                  <Column Name="Name" Type="WSTR" />

                  <Column Name="Price" Type="INT32" />

                  <Column Name="Amount" Type="INT32" />

                  <Dimension DimensionName="TestDimA" OutputName="TestDimA_ID" />

        </Columns>

        <Keys>

                  <Identity Name="IK_TestFactTableA_ID" Clustered="true">

                        <Column ColumnName="ID"/>

                  </Identity>

            </Keys>

    </Fact>

</Vulcan>

It’s a straightforward file, defines some columns, an identity key and a foreign key.

Well, if you just use simple “Vulcan” command to process it, you’ll see errors or warnings when opening the generated solution. The reason is that this file contains only its own definition for clarity, so that the external definitions, such as the Connection, and the Dimension, also need to be referenced. Remember the dimension file created in last lesson? It’s time to use it in the command line as below:

vulcan TestFactTable.xml -i TestDim.xml

The “-i” switch tells Vulcan to look for external definitions in other xml files. Actually you can add more xml files after that, with “-i” only shown once.

After executing this command line, you’ll see the generated solution and its files, and of course you can run it and find a new table TestFactTableA is created in DataWarehouse. The columns and keys should be created as described in this xml file.

In the following lessons, we’ll learn how to write ETL process to load data to tables, especially fact tables. You’ll find out this time-consuming task becoming easier to control in Vulcan.

In this lesson, we’ll see how 2 demo dimensions are created, and how a foreign key is created.

It isn’t uncommon that dimensions and fact tables are referencing foreign keys from other dimensions and tables. Using Vulcan, this step will be convenient and easy to maintain. The experience you learn in this lesson can surely be applied to much more complicated scenarios for your own dimensions and fact tables.

Let’s take a look at the code first.

<?xml version="1.0" encoding="utf-8" ?>

<Vulcan xmlns="https://tempuri.org/vulcan2.xsd">

      <Connection Name="DataWarehouse" Type="OLEDB" ConnectionString="Data Source=localhost;Initial Catalog=DataWarehouse;Provider=SQLNCLI10.1;Integrated Security=SSPI;" />

      <Dimensions>

            <Dimension ConnectionName="DataWarehouse" Name="TestDimA">

                  <Columns>

                        <Column Name="ID" Type="INT64"/>

                        <Column Name="Text" Type="WSTR" Length="255"/>

                  </Columns>

                  <Keys>

                        <Identity Name="IK_TestDimA_ID" Clustered="true">

                              <Column ColumnName="ID"/>

                        </Identity>

                  </Keys>

                  <Sources>

                        <StaticSource>

                              <Row>-1,'Unknown'</Row>

                        </StaticSource>

                  </Sources>

            </Dimension>

            <Dimension ConnectionName="DataWarehouse" Name="TestDimB">

                  <Columns>

                        <Column Name="ID" Type="INT64"/>

                        <Column Name="Text" Type="WSTR" Length="255"/>

                        <Dimension DimensionName="TestDimA" OutputName="TestDimAID"/>

                  </Columns>

                  <Keys>

                        <Identity Name="IK_TestDimB_ID" Clustered="true">

                              <Column ColumnName="ID"/>

                        </Identity>

                        <UniqueKey Name="UK_TestDimB_TestDimAID">

                              <Column ColumnName="TestDimAID"/>

                        </UniqueKey>

                  </Keys>

                  <Sources>

                        <StaticSource>

                              <Row>-1,'Unknown',-1</Row>

                        </StaticSource>

                  </Sources>

            </Dimension>

      </Dimensions>

</Vulcan>

It’s as straightforward as the code shows.

By running Vulcan to process this DimensionDemo.xml, 2 packages are created in the binary folder you appointed, which can create TestDimA and TestDimB in DataWarehouse when you execute these 2 packages.

Both dimensions have 2 columns, and an identity key. Now the important line is as below:

                        <Dimension DimensionName="TestDimA" OutputName="TestDimAID"/>

The above line in the definition of TestDimB will create a foreign key named TestDimAID and will be linked to the identity or primary key of TestDimAID.

This is very easy, isn’t it?

The UK_TestDimB_TestDimAID is another unique key created upon the TestDimAID.

Many cases can use similar code as this sample, but if the requirement is to create foreign key on non-primary key, or multiple-column keys, the code will look different. Details will be shown in next lesson.

<Vulcan xmlns="https://tempuri.org/vulcan2.xsd">

</Vulcan>

All the table definitions, ETLs should only exist in <Vulcan> element.

<Connections>

</Connections>

The list of all the connections used in the tables, ETLs. Each connection is represented by <Connection> element as below.   

<Connection Name="TestDataWarehouse" Type="OLEDB" ConnectionString="Data Source=localhost\sql2008;Initial Catalog=TestDataWarehouse;Provider=SQLNCLI10.1;Integrated Security=SSPI;" />

The connection defined by Name, Type and ConnectionString. The Name attribute will be used in table definitions and ETLs.

<Table Name="HelloWorld" ConnectionName="TestDataWarehouse">

</Table>

This is a table definition framework. You need to give it a name to be referenced in other places. The ConnectionName attribute specifies the connection to be used by this table.

<Columns>

</Columns>

This is the list of columns in a table. Each column is defined by <Column> element as below.

<Column Name="HelloID" Type="INT64" />

Column is usually defined by Name, Type and other attributes depending on the column’s type. Such as Length, Precision, Scale, IsNullable, Computed.

<Keys>

</Keys>

This is the list of keys defined in a table. Each key is defined by a key element as below. You can choose from Identity, PrimaryKey, and UniqueKey.

<PrimaryKey Name="PK_HelloPrimary" Clustered="true">

      <Column ColumnName="HelloID" />

</PrimaryKey>

These 3 lines define a primary key by its Name, Clustered attributes and its underlying column. There are other attributes you can apply to a key and the underlying column. For example, SortOrder can be used to specify if the sorting order is ascending or descending.

<Sources>

      <StaticSource>

            <Row>0, 'Hello, World!'</Row>

            <Row>1000, 'Hello, World 1000 times!'</Row>

      </StaticSource>

</Sources>

The above lines just put initial values into the table and should be straightforward.

To save time in editing, it’s strongly suggested to set the correct schema file for these XML files.

For example, if you’re using Visual Studio, open the “Properties” Window of the current XML file, and click on the button in “Schemas” field. Then you can add vulcan2.xsd and other Vulcan schema files in the namespace of https://tempuri.org/vulcan2.xsd and https://tempuri.org/Values.xsd

These schema files are located in the xsd folder of your copy of Vulcan. You can easily find them to finish your schema setting, but to avoid issues, you should never change these schemas.

Once this is done, you’ll find the editor now is more convenient than before. It’ll prompt you possible element name or attribute values, and remind you for typos, such as wrong names or values.

It is also a good way to get familiar with Vulcan features.

If you start writing an XML file from scratch, the following lines are the minimum you need to create a simple outline:

<?xml version="1.0" encoding="utf-8" ?>

<Vulcan xmlns="https://tempuri.org/vulcan2.xsd">

</Vulcan>

When you try to insert element into <Vulcan> element, the editor will give you quite a few choices, such as Dimensions/Dimension, Packages/Package, Connections/Connection. These keywords will be explained in details in the following lessons.

Starting from now, we’ll work on our own test database, with a name “TestDataWarehouse”. So you need to create a database with the same name in your SQL Server.

The first Vulcan XML file looks as simple as below: “HelloWorld.xml”

<?xml version="1.0" encoding="utf-8" ?>

<Vulcan xmlns="https://tempuri.org/vulcan2.xsd">

      <Connections>

            <Connection Name="TestDataWarehouse" Type="OLEDB" ConnectionString="Data Source=localhost\sql2008;Initial Catalog=TestDataWarehouse;Provider=SQLNCLI10.1;Integrated Security=SSPI;" />

      </Connections>

      <Table Name="HelloWorld" ConnectionName="TestDataWarehouse">

            <Columns>

                  <Column Name="HelloID" Type="INT64" />

                  <Column Name="HelloName" Type="WSTR" Length="255" IsNullable="true" />

            </Columns>

            <Keys>

                  <PrimaryKey Name="PK_HelloPrimary" Clustered="true">

                        <Column ColumnName="HelloID" />

                  </PrimaryKey>

            </Keys>

            <Sources>

                  <StaticSource>

                        <Row>0, 'Hello, World!'</Row>

                        <Row>1000, 'Hello, World 1000 times!'</Row>

                  </StaticSource>

            </Sources>

      </Table>

</Vulcan>

Copy and save it to HelloWorld.xml.

The only possible change you may need to modify is the ConnectionString. Your SQL server instance may be a different one, or even the default noname instance. So just modify the localhost\sql2008 to your own instance name.

Let’s bring the command window work environment we set in Lesson 0, type:

vulcan HelloWorld.xml

You’ll notice the output information and a progress bar. Because we only have 1 simple test file here to be compiled, the progress bar will advance to 100% very quickly.

2 folders will be created along with the HelloWorld.xml file.

The “PackageConfigurations” folder needs to be copied to the root of C: drive, i.e., C:\.

Then you can step into the “Table\HelloWorld” folder, either execute it from command line or from the GUI tools.

This folder contains the following files:

·        “HelloWorld.dtproj”, the project file of HelloWorld, which you can open with Visual Studio to check its contents.

·        “HelloWorld.dtsx”, the package file of HelloWorld, which you can execute from command line or GUI, such as Visual Studio.

·        Create Table.sql, the automatically generated helper file.

Once you execute this package, you’ll find a table named “HelloWorld” generated in your database, with columns, keys, and 2 rows of values as shown in the XML code.

This is just the beginning of how to use Vulcan, only shows you the convenience you can benefit from it: fast development, clear code for anyone to read, and easy maintenance.

We’ll learn more in the following lessons.

·        Microsoft SQL Server 2008 with Integration Service

·        Microsoft .NET Framework 3.5

·        Vulcan binaries

·        Microsoft Visual Studio 2008 (Optional)

You can get .NET Framework 3.5 from Microsoft Download Center https://www.microsoft.com/downloads/en/default.aspx

, and get Vulcan from https://vulcan.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25666

Vulcan binaries can be unzipped to any folder you like, just make sure to add its path to your system’s PATH variable, so that you can type “Vulcan” from command line to run it conveniently.

The MSBuild tool of Microsoft Visual Studio is a powerful tool to build large scale applications. Since Vulcan does support MSBuild, it’s suggested to install it to enjoy the convenience. It’s not required to learn Vulcan.

OK, once you get all the above done, type “Vulcan” from command line, if you see the following information, then it’s all set.

__VULCAN MAIN:Notification: VulcanRuntime: Welcome to Vulcan Version 2.2.3356.28949

VULCAN [filename] [-t targetpath] [-r root=path] [-i includefile1] [includefile2] [...] [includefileN]

    filename          Specifies the XML file to be processed.
    -t                Specifies the target path for the generated packages and files.
    -r root=path      Specifies the root path. For example:
                      -r detego=c:\detego
                      sets the 'Detego' root to c:\detego
    -i includefiles   The XML files need to be included to process the source XML file.
                      It's suggested to use '-r' on command line and <Using/> in XML instead.