Spec-Zone .ru
спецификации, руководства, описания, API
|
The INNODB_FT_INDEX_CACHE
table displays token information about newly inserted
rows in a FULLTEXT
index for an InnoDB
table. To avoid
expensive index reorganization during DML operations for an InnoDB
FULLTEXT
index, the information about newly indexed words is stored separately,
and combined with the main search index only when you issue the OPTIMIZE TABLE
statement for the InnoDB
table.
This table is only accessible to users with the SUPER
privilege. Before you query
this table, set the configuration variable innodb_ft_aux_table
to the name (including the database name) of the table
that contains the FULLTEXT
index, for example test/articles
.
Table 20.22. INNODB_FT_INDEX_CACHE
Columns
Column name | Description |
---|---|
WORD |
A word extracted from the text of a newly inserted row. |
FIRST_DOC_ID |
The first document ID that this word appears in in the FULLTEXT
index.
|
LAST_DOC_ID |
The last document ID that this word appears in in the FULLTEXT
index.
|
DOC_COUNT |
The number of rows this word appears in in the FULLTEXT index. The
same word can occur several times within the cache table, once for each combination of DOC_ID and POSITION values.
|
DOC_ID |
The document ID of the newly inserted row. This value might reflect the value of an ID column
that you defined for the underlying table, or it can be a sequence value generated by InnoDB when the table does not contain a suitable column.
|
POSITION |
The position of this particular instance of the word within the relevant document identified by
the DOC_ID value. The value does not represent an absolute
position; it is an offset added to the POSITION of the previous
instance of that word.
|
Notes:
This table initially appears empty, until you set the value of the configuration
variable innodb_ft_aux_table
.
You must have the PROCESS
privilege to query this
table.
For more information, see Section 12.9, "Full-Text Search Functions".