This section describes new Metadata features.
The Decompose Date feature, available when WebFOCUS synonyms are first created or refreshed, or when using the Synonym Editor, has been enhanced to create DEFINE fields that provide additional standard date formats for a date or a date-time field.
When creating a synonym, select the following option:
Decompose Date fields into separate Year, Month, Day fields
Eight additional DEFINE fields will be created. For example, if the date-time field is called DT, the synonym will contain the following defines:
DT_YEAR/I4 MISSING ON ALL=DTPART(DT, YEAR); DT_QUARTER/I1 MISSING ON ALL=DTPART(DT, QUARTER); DT_MONTH/I2 MISSING ON ALL=DTPART(DT, MONTH); DT_DAY/I2 MISSING ON ALL=DTPART(DT, DAY); DT_YEAR_Y/YYMDy MISSING ON ALL=DTRUNC(DT, YEAR); DT_QUARTER_Q/YYMDq MISSING ON ALL=DTRUNC(DT, QUARTER); DT_MONTH_M/YYMDm MISSING ON ALL=DTRUNC(DT, MONTH); DT_DAY_D/YYMD MISSING ON ALL=DTRUNC(DT, DAY);
If the field DT contained the value 2014-03-10 00:00:00.000, the new fields would contain the following values:
2014 1 3 10 2014 2014 Q1 2014/03 2014/03/10
In the Synonym Editor, right click any date or date-time field to expose the Decompose Date option.
The following enhancements have been made to the Metadata Wizard dialogue.
For application directories or synonyms in the Web Console and Data Management Console, the context menu option for New/Synonym is now labeled Synonym (Create or Update).
When this option is selected for a relational data source, a check list of available properties is displayed. The user can select the properties that should be updated for the database metadata.
This provides more control than the Refresh Synonym option for individual synonyms where the properties that are updated are fixed. The Refresh Synonym option is still available.
The CHECK FILE command shows the number of segments, number of fields, and the length of all fields in a synonym.
To perform the CHECK FILE command on a synonym:
For information, see the DataMigrator User's Guide.
Using DataMigrator to create and load a new table normally creates the synonym FIELD and ALIAS values and the target table with column names in uppercase. For data sources such as Greenplum, where lowercase column names are preferred, the case can be specified from the Data Management Console or Web Console. Right-click the folder for the adapter and select Change Settings from the context menu. For the ALIAS_CASE drop-down menu in the Metadata section, select Enforce lower case or Enforce upper case.
The case also be specified with the profile setting
ENGINE engine SET ALIAS_CASE {LOWER|UPPER}
where:
Is the suffix for the data source, for example, SQLGPDB,
Indicates that column names should be lowercase.
Indicates that column names should be uppercase.
Enhancements have been made to the way in which WebFOCUS processes a request against a generic cube synonym.
A generic cube is a cluster synonym in which a cube structure has been superimposed on top of one or more of the joined synonyms in the cluster. This is typically the case for a synonym that represents a Star Schema in a Relational Database, which consists of fact and dimension tables.
A cube structure has one or more measures (data that can be aggregated). Measures are usually grouped into placeholders called Measure Groups. Measures are usually identified with a Fact table which forms the root of the Cluster Synonym. The Cluster can have one or more root segments (multi-fact).
Dimensions are ways to analyze measures. A cube can have one or more dimensions, which are children of the fact table. If the synonym has more than one root segment, first level child segments can be shared across parents (shared dimension) or joined (linked) to only one parent (non-shared dimension). A dimension can have dependents (snowflakes), in which case they form hierarchies. Dimensions can also have attributes, which can be part of the levels of a hierarchy.
A generic cube structure is created using Dimension View feature of Synonym Editor.
A SET command has been introduced that will limit the scope of a request against a generic cube to only those fields referenced in the OLAP structure.
The new behavior is activated with the following SET command
SET REQSCOPE = {DEFAULT|CUBE}
where:
Does not limit a request to field in an OLAP-defined structure. This is the default value.
Limits a request scope to fields specified in the OLAP structure. DBA restrictions will be applied if any field in the request has a relationship to the Dimension table that has DBA, even if the Dimension table itself is not referenced. Fields used in DBA rules also have to be within the scope of the OLAP structure.
If a request attempts to retrieve fields that are out of the scope, the following messages are generated, and the request halts:
(FOC32655) FIELD USED IN REQUEST IS NOT PART OF THE CUBE: name(FOC009) INCOMPLETE REQUEST STATEMENT
BYPASSING TO END OF COMMAND
In the Web Console, this setting is under the Table Miscellaneous Service category of the Core Engine Settings page.
The Dimension View option of the Synonym Editor on the Web Console now allows Dialogue Manager variables, filters, defines, computes, and sort objects to be included as candidates for selection. This feature is used to create a generic cube structure superimposed on the physical structure of the synonym.
The following SET command propagates certain OLAP metadata from the referenced synonyms in a request to the output HOLD file that the request generates.
SET HOLDATTR=CUBE
The following describes what is transferred:
If any error is detected in processing the cube metadata, the HOLD file is processed as if the HOLDATTR=ON setting is active.
Master File formats DATE and the date portion of date-time fields can be customized during SQL or Excel synonym creation. This allows you to control date format appearance in reports.
Available formats are:
The Synonym Editor now supports the BYTEORDER attribute for file data sources. Previously, the BYTEORDER attribute could only be set from a text editor.
This option specifies the order in which bytes are stored for integer values. This is generally only required when the data source is a file transferred from a system with a different byte order.
The values are:
BE (Big Endian) the most significant byte is first. Hardware using this order includes IBM zSeries, POWER, and Sparc.
LE (Little Endian) the least significant byte is first. Hardware using this order includes Intel x85, x86-64, and Itanium.
The option to fully qualify field names in the join criteria of a cluster synonym has been added to the Synonym Editor.
This new setting can be found under Options > General > Column Management. It has two values, Always and For duplicate fields. The default is For duplicate fields in order to maintain compatibility with earlier releases.
The Synonym Editor now provides a simplified process for specifying USAGE attributes for DATE and TIME columns. You select the date components (such as, Year, Month, Day, and Hour) and their order first. Then you select the display options for the individual components.
WebFOCUS |