InnoDB Full-Text Search improvements¶
By default all Full-Text Search indexes check the stopwords list, to see if any indexed elements contain one of the words on that list.
Using this list for n-gram indexes isn’t always suitable, as an example, any
item that contains
i will be ignored. Another word that can’t be
east, this one will find no matches because
a is on the
FTS stopword list.
Although this variable is introduced to resolve n-gram issues, it affects all Full-Text Search indexes as well.
Being a stopword doesn’t just mean to be a one of the predefined
words from the list. Tokens shorter than innodb_ft_min_token_size
or longer than innodb_ft_max_token_size
are also considered stopwords. Therefore, when
innodb_ft_ignore_stopwords is set to
ON even for non-ngram
innodb_ft_max_token_size will be
ignored meaning that in this case very short and very long words will
also be indexed.
By default, full text search is unable to find words with various punctuation
characters in boolean search mode, although those characters are
indexed with ngram parser. A new variable
was introduced in Percona Server
5.7.21-20 to solve this issue.
When it’s enabled, all the non-whitespace symbols are considered to be word symbols by FTS query parser, except for the boolean search syntax symbols (which are specified by ft_boolean_syntax variable). The latter ones are also considered to be word symbols inside double quotes. This only applies for the query tokenizer, and the indexing tokenizer is not changed in any way. Because of this, the double quote symbol itself is never considered a word symbol, as no existing indexing tokenizer does so, thus searching for it would never return documents.
Command Line: Yes Config File: Yes Scope: Session, Global Dynamic: Yes Variable Type: Boolean Default Value:
When enabled, this variable will make all non-whitespace symbols (including punctuation marks) to be treated as word symbols in full-text search queries.