Pseudo columns are columns in the index table that define the parameters of the search and provide access to the results data. (For more information about index tables, see “The Index Table”.) These columns are valid only in the context of a query; that is, the information in the columns is valid only for the duration of the query. If the query that follows contains a different set of parameters, the pseudo columns contain a different set of values.
Each pseudo column in an index table describes a different search attribute. For example, if you indicate the score column, the query displays only the result set that falls within the parameters you define. For example, the following query displays only the results that have a score value greater than 90:
index_table_name.score > 90
Other pseudo columns (like highlight) are used to retrieve data generated by Verity for a particular document. Table 5-1 describes the pseudo columns that are maintained by the Full-Text Search engine.
Pseudo Column Name |
Description |
Datatype |
Length (in Bytes) |
|---|---|---|---|
cluster_number |
Enhanced Full-Text Search engine only. Contains the cluster that the row is part of. Clusters are numbered starting with 1. You can use the cluster_number column only in the select clause of a query. |
int |
4 |
cluster_keywords |
Enhanced Full-Text Search engine only. Contains the keywords that Verity uses to build the cluster. You can use cluster_keywords only in the select clause of a query. |
varchar |
255 |
highlight |
Offsets within the document all words from the query. You can use highlight only in the select clause of a query. |
text |
16 |
id |
Uniquely identifies a document within a collection. Used to join with the IDENTITY column of the source table. You can use id in the select clause or where clause of a query. |
numeric |
6 |
index_any |
Provides a Verity language query to the Full-Text Search engine. You can use index_any only in a where clause. |
varchar |
255 |
max_docs |
Limits results to the first n documents, based on the default sort order. In a clustered result set, limits results to the first n documents in each cluster. You can use max_docs only in a where clause. |
int |
4 |
score |
The normalized measure of correlation between search strings and indexed columns. The score associated with a specific document has meaning only in reference to the query used to retrieve the document. You can use score in a select clause or a where clause. |
int |
4 |
sort_by |
Specifies the sort order in which to return the result set. The Enhanced Full-Text Search engine allows up to 16 sort specifications in the sort_by column. You can use sort_by only in a where clause. |
varchar |
35 |
summary |
Selects summarization data. You can use the summary column only in the select clause of a query. |
varchar |
255 |
The following sections describe the functionality of the pseudo columns.
