All Products
Search
Document Center

Tablestore:Full-text search

Last Updated:Jun 23, 2026

Tablestore supports full-text search through match query and match phrase query. Full-text search matches data based on tokens and highlights matched keywords in query results.

Data query

The following table describes the data query features available for full-text search.

Feature

Description

Scenario

Match query

Match query searches for data that contains a specific keyword based on approximate matches. Tablestore tokenizes Text field values and query keywords based on the specified tokenization method, then matches data against the resulting tokens.

Suitable for big data analytics, content search, knowledge management, social media analysis, log analysis, intelligent Q&A systems, and compliance review.

Match phrase query

Match phrase query searches for data that contains a specific keyword based on approximate matches, with the additional requirement that the tokens appear in the same order and positions as the tokenized keyword.

Suitable for big data analytics, content search, and intelligent recommendation.

Data processing

The following table describes the data processing features used in full-text search.

Feature

Description

Tokenization

When you specify a tokenization method for a Text field, Tablestore splits the field values into tokens accordingly. Supported methods: single-word tokenization, delimiter tokenization, minimum semantic unit-based tokenization, maximum semantic unit-based tokenization, and fuzzy tokenization.

Highlight

The highlight feature marks matching or related text in query results, helping you quickly locate relevant content.