Use to create and configure attributes of your datastores.
Datastores describe data accessed by Oracle Data Integrator through two-dimensional tabular structures. Oracle Data Integrator contains the metadata related to the attributes. The data contained in the applications is stored in rows.
Attribute editors can be invoked by double-clicking the individual attributes or on the attributes tab from the corresponding data store.
The Attribute Editor has the following tabs:
| Properties | Description |
|---|---|
| Name | The name of the attribute, as recognized by the data server. |
| Datatype | The datatype, corresponding to a type recognized by the data server. |
| Storage Type | This field is only displayed for HBase attributes.
The storage type, using which you can define how a data type is physically stored in HBase. Possible values are |
| Short Description | The descriptive label of the attribute. This label makes the user interface easier to read if the attribute names are not explicit. |
The description of the physical storage format of the data.
| Properties | Description |
|---|---|
| Order | The order number or physical rank of the attribute in the datastore. |
| Start | The physical position of the attribute. This information is to be given for fixed format files. The first position is position 1. |
| Length | The physical length in number of bytes. An attribute with start position '1' and length '4' occupies bytes 1,2,3,4 in each row.
Note: For delimited files, the maximum length taken by the attribute should be shown. If the maximum length of the delimited file is not known, try to ensure that the length is greater than the maximum length. This length is used to reserve sufficient memory space for reading the information. |
| Record codes | The list of record codes that allows files with several record patterns to be processed.
For example, for a file INVOICE containing different header and invoices rows marked by a record code CODE_REC with the value 'HEAD' or 'LINE', you must create a datastore for each of these conceptual entities. Specify HEAD in the list of record codes for the attribute CODE_REC of the datastore INVOICE, and LINE in the corresponding attribute for the datastore INVOICE_LINE. You can specify several values separated by the semicolon (;) character. |
| Exclude the Record Codes | If this box is selected, the rows respecting the record codes are excluded (instead of included) when loading the file. |
The description of the data representation.
| Properties | Description |
|---|---|
| Logical Length | The logical length of the attribute, as it is functionally seen by the user. For a number, this is the total number of numerals (Precision).
Specify this length even if the physical length is given. |
| Format | The format used for dates. This format is a Java date format that matches your computer's local parameters. |
| Scale | The number of digits that occur after the decimal. |
| Decimal separator | The decimal separator (by default, a period (.)). |
This section is only displayed for Hive attributes.
| Properties | Description |
|---|---|
| Used for Partitioning | Select to use Hive columns for partitioning.
Note: This value is filled in by reverse-engineering a Hive table and should not be changed manually. |
| Used for Clustering | Select to use Hive columns for clustering.
Note: This value is filled in by reverse-engineering a Hive table and should not be changed manually. |
| Used for Sorting | Select to use Hive columns for sorting.
Note: This value is filled in by reverse-engineering a Hive table and should not be changed manually. |
Use to configure the generation and the deployment of data services for this attribute. More
| Properties | Description |
|---|---|
| Allowed operations | Select the box that corresponds to the type of actions (INSERT, UPDATE, and SELECT) you want to allow data services to perform on this attribute. One important use of this tab is to lock an attribute against being written to by data services. |
| Properties | Description |
|---|---|
| Default Value | The value inserted in this attribute if no value is specified. |
| Read only | Select this box if this attribute cannot be used in an INSERT or UPDATE command. This is the case, for example, for attributes whose values are automatically set and cannot be modified, such as IDENTITY attributes. |
| Slowly Changing Dimensions Behavior | The attribute behavior when loading a slowly changing dimension table for OLAP. The following attribute behaviors are supported:
Some of the Knowledge Modules support slowly changing dimensions (SCDs). See the Oracle Fusion Middleware Connectivity and Knowledge Modules Guide for Oracle Data Integrator for more information. |
| Description | The detailed description of the attribute. |
Use to display and configure the quality control properties.
The quality control properties are taken into account during a flow control, during a static control, or while loading a file.
| Properties | Description |
|---|---|
| Mandatory | If selected, the attribute must be completed. |
| Control | Type of quality control for which the mandatory status is checked. |
| Flow | During a flow control for data integrated into this datastore, the mandatory status of the attribute is verified if this box is selected. This information is a default value that can be modified when the mapping is designed. It is recommended that you activate this type of check to ensure data quality, even if the target technology also verifies the information. |
| Static | During a static control, that is, when data already present in this datastore is checked, the mandatory status of this attribute is also verified if the box is selected. |
| On Error | Data in the file may be inconsistent with the datastore definition. When reading the file, if one value from the row is inconsistent with the attribute description, the On Error option defines the action Oracle Data Integrator will perform. |
| Reject Error | The row containing the error is moved into a file with the BAD extension, and an explanation of the error is placed in a file with the ERROR extension.
The BAD and ERROR files are located in the same directory as the file being read. |
| Null if error (inactive trace) | The row is kept in the flow and the erroneous value is replaced by null. |
| Null if error (active trace) | The row is kept in the flow, the erroneous value is replaced by null, and an explanation of the error is placed in a file with the ERROR extension. |
Oracle Fusion Middleware Developing Integration Projects with Oracle Data Integrator
Oracle Fusion Middleware Integrating Big Data with Oracle Data Integrator