SQL/Data Types

This section describes the data types that are used when defining Input and Output Mappings in a data service:

Array data type

Array parameter type enables data services to dynamically provide a set of values without knowing the size of the data set beforehand. Generally, multiple data types are used for building queries for more complex inputs. Array data type is available as a parameter type when creating input mapping queries that are defined while creating a data service.

Let's see how array types are handled internally in the SOAP message-level. When the parameters are serialized (in scalar types), there is only one element per parameter in the SOAP message. In arrays, the element representing the parameter is repeated. For example, refer to the sample SOAP message below, which contains the array type employeeNumbers.

...
<s:Body>
   <p:setSalaryForEmployees xmlns:p="http://ws.wso2.org/dataservice/samples/rdbms_sample">
      <salary xmlns="http://ws.wso2.org/dataservice/samples/rdbms_sample">15000</salary>
      <employeeNumbers xmlns="http://ws.wso2.org/dataservice/samples/rdbms_sample">1002</employeeNumbers>
      <employeeNumbers xmlns="http://ws.wso2.org/dataservice/samples/rdbms_sample">2014</employeeNumbers>
      <employeeNumbers xmlns="http://ws.wso2.org/dataservice/samples/rdbms_sample">4411</employeeNumbers>
   </p:setSalaryForEmployees>
</s:Body>
</s:Envelope>

Array types are properly mentioned in the WSDL generation. As a result, it is suitably presented in a service that is code generated.

Note

Note that the ARRAY parameter type cannot be used with the QUERY_STRING data type.

Binary data type

When using data services, you might come across the need to transfer binary data from/to the server. This is handled using Base64 encoding. When sending, binary data must be encoded in Base64 format. Similarly, when receiving, the receiving character data must be Base64-decoded in order to retrieve the original binary data.

Binary data type is available as an SQL Type when creating input mappings for your queries.

When defining the result of a data service, in order to declare that a binary value is expected, add a suitable entry in Output Mappings. The Schema Type of the Output Mapping should be selected as " xs:base64Binary."

When using code-generated clients, the encoding/decoding of Base64 data need not be explicitly done by the user since the existence of binary data is mentioned in the WSDL. For example, in Axis2 code-generated stubs, the binary types are handled using the "DataHandler" class.

User-defined data types

Apart from the standard data types (such as Varchar, Integer, etc.) it is also possible to query custom objects, which are usually called 'User Defined Types' (UDT). Users can query UDTs with ordinary SQL queries as well as OUT parameters of stored procedures.

Query UDTs with ordinary SQL queries

In this option, you are only required to define an output mapping corresponding to the UDT that is to be queried. An important point is that the attributes of a UDT is queried depending on the order they are specified in the UDT. For example, if the UDT carries the structure below, the attribute index "0" maps to the attribute "Id".

"SampleUDT{Id Integer, Name Varchar(100)}"

Similarly, the "Name" attribute can be retrieved via the index "1".

Query UDTs via OUT parameters of a stored procedure

It is possible to retrieve the values of UDTs via the OUT parameters defined in stored procedures. The user should define an input mapping as an OUT parameter as well as an output mapping in order to actually retrieve the values of the UDT attributes.

TIMESTAMP data type

If you are writing a query for a column in the database that requires a value in the form of a date and time, the TIMESTAMP data type should be used for the mapping in the data service. The format of the TIMESTAMP value should be as follows: ' YYYY-MM-DD T hh🇲🇲ss[.mmm]'. Note that ' T ' should be included to indicate the start of the time value in the time stamp.

Note

The TIMESTAMP data type is synonymous to the DATETIME data type. Therefore, if you are writing a query for a column of DATETIME type , you can use TIMESTAMP in your input/output mapping.

Top