Перейти к основному содержимому

MySQL MATCH AGAINST Clauses

The `MATCH ... AGAINST` clause in MySQL is used for full-text search, allowing for complex search queries within text fields. It enables searching for words and phrases in text columns efficiently.

Usage

The `MATCH ... AGAINST` clause is used when you need to perform a full-text search on a text-based column in a table. It requires a full-text index on the columns being searched.

Prerequisites

  • Storage Engine Support: Full-text search is supported by specific storage engines such as InnoDB and MyISAM.
  • Column Types: Only certain column types like CHAR, VARCHAR, and TEXT can be indexed for full-text search.
sql
SELECT column1, column2, ...
FROM table_name
WHERE MATCH (column1, column2, ...) AGAINST ('search_term' [IN NATURAL LANGUAGE MODE | IN BOOLEAN MODE | WITH QUERY EXPANSION]);

In this syntax, `MATCH (column1, column2, ...) AGAINST ('search_term')` specifies the columns to search and the term to search for.

Examples

1. Basic Full-Text Search

sql
SELECT title
FROM articles
WHERE MATCH (title, content) AGAINST ('database');

This example retrieves articles where the `title` or `content` fields contain the word "database".

2. Full-Text Search with Boolean Mode

sql
SELECT title
FROM articles
WHERE MATCH (title, content) AGAINST ('+database -SQL' IN BOOLEAN MODE);

In this query, `+database -SQL` specifies that the word "database" must be present and "SQL" must be absent in the `title` or `content`.

3. Full-Text Search with Query Expansion

sql
SELECT title
FROM articles
WHERE MATCH (title, content) AGAINST ('search' WITH QUERY EXPANSION);

Here, `WITH QUERY EXPANSION` expands the search to include related terms, providing a broader range of results for the term "search".

Search Modes Explained

  • Natural Language Mode: Processes the search term as a natural language phrase, ranking results based on relevance.
  • Boolean Mode: Allows for more complex queries using operators like `+`, `-`, and `*` to refine search results.
  • Query Expansion: Uses the initial search results to further search for related terms, broadening the search scope.

Tips and Best Practices

  • Ensure Full-Text Indexes. Create full-text indexes on columns where you intend to use `MATCH ... AGAINST` for optimal performance.
  • Choose the right search mode. Use natural language mode for simple searches, boolean mode for more complex queries, and query expansion when broader results are desired.
  • Optimize column selection. Include only necessary columns in the `MATCH` clause to improve search speed and relevance.
  • Handle stopwords and minimum word length. Be aware of MySQL's default stopwords and word length settings, which might filter out common or short words from searches.