Aside from quickly generating code via 'Quick Generation', you will eventually have to create a Project in Project Explorer.

Projects are designed to help organize the different configurations you may need to successfully generate structured code.
Each project may contain one or more configurations.

 

Most small to medium sized projects may only need one configuration.
Larger projects however, may need different configurations for different environments or situations. For example, in the development environment, if you use Intercepts to implement logging, you may want to enable Intercepts in all Layers and sub layers. However, as your project/code matures towards the production environment, you may decide to enable Intercepts only for the Business Layer objects.

Structured Intercepts are methods defined in a partial class that, the developer may change to influence the outcome of the generated methods. These are optional methods that are called right before the body of a generated method.

With Structured Intercepts, you can do the following:

• Throw exceptions, based on the value of passed in parameters.
• Stop the execution of the calling method and execute custom code or call a different method.
• Change the values of passed in parameters and continue with the execution of the generated method.

Views are one of the primary object types in most RDBMSs and starting from version 2.0, you may generate structured code that involves Views.

Views were excluded from the initial versions of OxyGen Code because generating code based on Views usually means the loss of relationship information.

For view centric environments, this leads to hand-coding a lot of the code that normally would have been generated. In version 2.0, we provide the option of propagating relationships, from Tables, unto the Views that are based on those Tables. This option will also propagate the primary keys as well. This is called Table Virtualization for Views. It may be configured via the "Virtualization Settings" View Processing Instructions and the "Persistent Storage" Processing Instruction (available for Views and Tables)

Another key enhancement for version 2.0 is the generation of structured code based on custom Stored Procedures.

The generation of structured code for Stored Procedures led to the creation of a new class of database related objects: Virtual Views.

Virtual Views
Virtual Views are used to represent the result-set of Stored Procedures. You may also choose to represent the result-set using Tables, Views, or a small selection of system types.

MultiResultset
To propagate the data from a Stored Procedure that returns multiple result-sets, a specialized class, the MultiResultset, is used.
The MultiResultset class is exceedingly simple and inherits from List (List of T). As always, the code will be part of your project.

Command Execution
OxyGen Code version 2.0 supports all the command execution types as well as a few derived ones.

• ExecuteReader
• ExecuteNonQuery
• ExecuteNonQuery (with rows affected)
• ExecuteNonQuery (with row count)
• ExecuteScalar
• ExecuteXmlReader
• ExecuteSingleRow

Stored Procedures that use the ExecuteXmlReader execution type have the option of three System types as return types: System.Xml.XmlDocument, System.Xml.XmlReader, System.String.

Since some databases may contain hundreds of stored procedures, two new wizards have been created to help with the configuration of Stored Procedures:

• The Virtual View Generator
• The StoredProcedure Generator

The idea behind the folder structure changes is to separate the three types of code that can be found in a solution that uses OxyGen Code:

• Generated code that has the potential to change from session to session
• Generated code that does not change from session to session
• Custom code written by the user

  To help with the separation, four special folders are used

• Custom
• Generated
• Intercepts
• Durable

Custom
When the code is first generated, an empty partial class for each entity is added to the 'Custom' folder. The code generator will not delete or replace any files in this folder. The generation of the empty partial class is optional and if this option is not selected, the custom folder will not be generated.

Generated
The Generated folder is the primary location for the generated files. Files stored directly by this folder may be replaced by the code generator on subsequent code generation sessions.

Intercepts
The Intercepts folder location contains partial classes with private methods that maybe used to modify the behavior of the generated methods. The generator will not delete or replace files in this folder. On subsequent generations, it may become necessary to modify some of these files to include additional methods. The generator will inject new methods into the matching files within this folder. The generator will not delete or replace methods in any file contained by this folder.

You may modify the body of the methods contained by files in the Intercepts folder, but if you wish to write custom code unrelated to the Intercepts, the partial classes defined in the Custom folders are the right way to do it.

Durable
Files in the Durable folder will never be deleted or changed in any way by the code generator. This location was conceived as a container of files that don’t change across generation sessions. Typically, it contains base classes and utility/infrastructure classes. You may modify the classes in this location without fear that they may be overwritten.

With the Code Publication feature of OxyGen Code version 2.0, you are able to precisely decide what you want to publish into your Work folder. This works hand in hand with the OxyGen Code Folders, and allows you to decide what should be moved into the Work folders.

The generated code can be published in the following ways:

• By configuring the code publishing options as part of the code generation settings
• Using Source Code Explorer

The programming model for writing provider-independent code is based on the use of the "factory" design pattern, which uses a single API to access databases across multiple providers.

The Data Provider class generated by OxyGen Code version 2.0 uses this design pattern and includes a new abstract base class. A "Common Types" enumeration is also included.

The WHERE Clause of the generated search stored Procedures have been enhanced and with version 2.0, you will have the option of either Union or Intersection search for the Search Query.

You may also select the column you want to search by.

Optionally, the name of the search method can be specified instead of using the auto generated search method name.

For Tables and Views, you are now able to decide if they are modifiable (insert, update, delete). This is also applicable to individual columns as well (insert, update).

Tables are modifiable by default and Views are not modifiable by default.