Hi
Untyped DataSets
Untyped
datasets are hard to maintain since the developer has to know the in
depth details of the database. This scenario is not practical since in
actual programming the developer does not have any idea of the database
layer and its hidden from him by the database administrators.
Another
disadvantage of using Untyped DataSets is that the code looks hideous.
Consider the code below which simply retrieves the first row of the
first column from the database.
| // Weak Type DataSet
DataSet weakDataSet = new DataSet();
ad.Fill(weakDataSet,"Articles");
string articleId = weakDataSet.Tables["Articles"].Rows[0][0].ToString();
|
I am simply
make a dataset object and filling the dataset using SqlDataAdapter. Once
the dataset is loaded with data I am retrieving the first row of the
first column in the dataset. As you can see that by just looking at the
code to retrieve the data one cannot simply understand that what is
going to be returned.
Now if you are using a dataset and need to print all the rows than you might have to code something like this:
| foreach(DataTable dt in weakDataSet.Tables)
{
foreach(DataRow dr in dt.Rows)
{
foreach(DataColumn dc in dt.Columns)
{
myList.Items.Add(dr[dc].ToString());
}
}
}
|
As you
can see the code above is hard to understand since we have not mentioned
any column in the database. Although it will work fine but its very
hard to understand that what is going on.
Most of these problems are solved by using Strongly Typed DataSets. Let's see how we can use them.
Strongly Typed DataSets:
You
can start using Strongly Typed DataSets by simply adding a new dataset
item from right clicking the project and selecting add new item. Later
you can drag and drop a table from the server explorer to create a Strongly Typed DataSets.
Once
you have created Typed DataSet you can simply make an object of that
dataset and than you are ready to use it.
Below I made an instance of
the typed dataset "MyDataSet".
| MyDataSet ds = new MyDataSet(); |
Once the DataSet is created we need to fill it using SqlDataAdapter.
Now let's see that how easy it is to refer to the column and rows that are present in the dataset.
| ds.Articles[0].ArticleID;
ds.Articles[0].Author;
ds.Articles[0].DateCreated;
|
As you can see
above that now we know that which columns we are referring to. The
Articles is the name of the table and [0] represents the first row of
the column ArticleID, Author and DateCreated.
Now let's see how we can iterate through the dataset.
| foreach(MyDataSet.ArticlesRow articleRow in ds.Articles)
{
myList.Items.Add(articleRow.Title);
}
|
As you see
that now we don't have to type much code in order to iterate through the
dataset. We just used the dataset row to iterate through the dataset
collection and finally we added the items in the Listbox.
Typed DataSets are more faster than the UnTyped DataSets. Strongly typed datasets are early bound and their columns represents the properties.
Also that if you are getting errors in Typed DataSets they will most
probably be compile time and not runtime so you can always fix them.
when you make Typed DataSets you also get some extra features and methods that are very useful.
| ds.Articles[0].IsAuthorNull(); |
As
you can see above each column that is included in the Typed DataSet
comes with a method which let's the user know that if the current row
that he is accessing is null or not. This is very convenient for columns
that can have null values. If you don't check for null values "Object
reference not set to an instance" error will be fired.
Adding new columns in the Typed DataSets are also very easy. Just switch back to the xsd schema graphical user interface and add the column.
Now after adding a new column just build the application and you are ready to use the new column.
The coolest feature of using Typed DataSets is that you can convert the whole dataset schema into a class which can be used by other users. You can easily do this by using a tool called XSD.exe.
Go to the Visual Studio.NET Command prompt. And type this command.
xsd MyDataSet.xsd /c
The /c in the end means that we are creating a class file from the xsd file. There are many things you can do depending on the type of the escape sequence you apply.
Remember
that this command must be run from the same directory where the xsd
file is present. Once you the command it will create the class file with the same name as that of the xsd file.
Now you schema has become portable and so others can easily use this.
Many
people are using Typed DataSets as business objects. There are some
problems in this approach which I will address in my future article
"Typed DataSets Vs Entity Classes".
I hope you enjoyed the article, happy programming !
refer
http://www.codersource.net/microsoft-net/c-advanced/untyped-datasets-and-strongly-type-datasets.aspx