Things to note about mysql full text search

This article gives some important instructions about full-text search.

1. When indexing full text data, short words are ignored and excluded from the index. Short words are defined as those words with 3 or fewer characters (this number can be changed if necessary).

2.MySQL comes with a built-in list of stopwords, which are always ignored when indexing full-text data. This list can be overridden if desired (see the MySQL documentation to learn how to accomplish this).

1. Many words appear so frequently that searching for them is useless (too many results are returned). Therefore, MySQL stipulates a 50% rule. If a word appears in more than 50% of the rows, it is ignored as a non-word. The 50% rule does not apply to IN BOOLEANMODE.

2. If the number of rows in the table is less than 3 rows, the full text search returns no results (because each word either does not appear, or appears in at least 50% of the rows).

3. Ignore single quotes in words. For example, don't is indexed as dont .

4. Languages ​​that do not have word separators (including Japanese and Chinese) do not return full-text search results appropriately.

5. As mentioned before, full-text search is only supported in the MyISAM database engine.

No Proximity Operator Proximity search is a feature supported by many full-text searches that searches for adjacent words (in the same sentence, in the same paragraph, or in a section of a specific number of words, etc). MySQL full-text search does not currently support the proximity operator, but there are plans to support this operator in future versions.

The above is the detailed content of Things to note about mysql full text search. For more information, please follow other related articles on the PHP Chinese website!