Full-text searches are not supported for partitioned tables. See Section 20.6, “Restrictions and Limitations on Partitioning”.
Full-text searches can be used with most multibyte character sets. The exception is that for Unicode, the
utf8character set can be used, but not the
ucs2character set. Although
ucs2columns cannot be used, you can perform
IN BOOLEAN MODEsearches on a
ucs2column that has no such index.
The remarks for
utf8also apply to
utf8mb4, and the remarks for
ucs2also apply to
Ideographic languages such as Chinese and Japanese do not have word delimiters. Therefore, the built-in full-text parser cannot determine where words begin and end in these and other such languages.
In MySQL 5.7.6, a character-based ngram full-text parser that supports Chinese, Japanese, and Korean (CJK), and a word-based MeCab parser plugin that supports Japanese are provided for use with
Although the use of multiple character sets within a single table is supported, all columns in a
FULLTEXTindex must use the same character set and collation.
MATCH()column list must match exactly the column list in some
FULLTEXTindex definition for the table, unless this
IN BOOLEAN MODEon a
MyISAMtables, boolean-mode searches can be done on nonindexed columns, although they are likely to be slow.
The argument to
AGAINST()must be a string value that is constant during query evaluation. This rules out, for example, a table column because that can differ for each row.
Index hints are more limited for
FULLTEXTsearches than for non-
FULLTEXTsearches. See Section 9.9.4, “Index Hints”.
InnoDB, all DML operations (
DELETE) involving columns with full-text indexes are processed at transaction commit time. For example, for an
INSERToperation, an inserted string is tokenized and decomposed into individual words. The individual words are then added to full-text index tables when the transaction is committed. As a result, full-text searches only return committed data.
The '%' character is not a supported wildcard character for full-text searches.