MATCH Processing

How to:

Reference:

The way MATCH merges data depends on the order in which you name data sources in the request, the BY fields, display commands, and the merge phrases you use. In general, however, processing is as follows:

  1. MATCH retrieves requested records from the first data source you name, and writes them to a temporary work area.
  2. MATCH retrieves requested records from the second data source you name, and writes them to a temporary work area.
  3. It compares the retrieved records common high-order sort fields as specified in the merge phrase (for example, OLD-OR-NEW). For more information, see Merge Phrases.
  4. It writes the merged results of the comparison to a temporary data source (if there are more MATCH operations). It cycles through all data sources named until END is encountered.
  5. It writes final records to the HOLD file.

Top of page

x
Reference: Merge Phrases

MATCH logic depends on the concept of old and new data sources. Old refers to the first data source named in the request, and new refers to the second data source. The result of each merge creates a HOLD file until the END command is encountered.

The following diagram illustrates the general merge process:


Top of page

x
Syntax: How to Specify Merge Phrases
AFTER MATCH HOLD [AS 'name'] mergetype					

where:

AS 'name'

Specifies the name of the extract data source created by the MATCH command. The default is HOLD.

mergetype

Specifies how the retrieved records from the files are to be compared.

The results of each phrase are graphically represented using Venn diagrams. In the diagrams, the left circle represents the old data source, the right circle represents the new data source, and the shaded areas represent the data that is written to the HOLD file.

OLD-OR-NEW specifies that all records from both the old data source and the new data source appear in the HOLD file. This is the default if the AFTER MATCH line is omitted.

OLD-AND-NEW specifies that records that appear in both the old and new data sources appear in the HOLD file. (The intersection of the sets.)

OLD-NOT-NEW specifies that records that appear only in the old data source appear in the HOLD file.

NEW-NOT-OLD specifies that records that appear only in the new data source appear in the HOLD file.

OLD-NOR-NEW specifies that only records that are in the old data source but not in the new data source, or in the new data source but not in the old, appear in the HOLD file (the complete set of non-matching records from both data sources).

OLD specifies that all records from the old data source, and any matching records from the new data source, are merged into the HOLD file.

NEW specifies that all records from the new data source, and any matching records from the old data source, are merged into the HOLD file.


WebFOCUS