Requirements for pattern matching searches

Table and time series definition requirements

The time series table and the time series definition must conform to the following requirements:

• The time series table must have only one TimeSeries column.
• The time series table must have a primary key. The primary key can be a composite of multiple columns.
• The columns in the TimeSeries subtype on which you run pattern matching searches must have one of the following data types: SMALLFLOAT, FLOAT, INTEGER, SMALLINT, BSON. For a BSON column, you can search and index individual fields within the BSON document that contain numeric data.
• The time series must be regular.

Server requirements

In addition to the TimeSeries extension, pattern matching searches require the TimeSeries Pattern Matching extension. If you create pattern match indexes, the TimeSeries Pattern Matching Index is required.

The TimeSeries Pattern Matching extension is automatically registered the first time that you run the TSPatternMatch function. The name of the extension is TSPMatch.bld and it is in the $INFORMIXDIR/extend/TSPMatch.version/ directory, where version is the internal version number of the extension.

The TimeSeries Pattern Matching Index extension is automatically registered the first time that you run the TSCreatePatternIndex function. The name of the extension is TSPIndex.bld and it is in the $INFORMIXDIR/extend/TSPIndex.version/ directory, where version is the internal version number of the extension.

Replication

You cannot replicate pattern matching indexes with Enterprise Replication.