utf8mb4unicodeci

utf8mb4_unicode_ci is a collation for the utf8mb4 character set used by MySQL (and compatible with MariaDB). The collation defines how strings are compared, sorted, and indexed, and is widely used for multilingual databases. The commonly referenced name is utf8mb4_unicode_ci, with the informal form “utf8mb4unicodeci” sometimes seen without the underscores.

This collation implements the Unicode Collation Algorithm to provide language-aware ordering and comparison. It is case-insensitive, as indicated by the “ci” suffix, meaning that uppercase and lowercase forms are treated as equal for equality and most sorts. It aims to yield linguistically reasonable results across a broad range of languages and scripts, including those that require handling of characters beyond the basic Latin set. Since utf8mb4 supports all Unicode code points, including supplementary characters, utf8mb4_unicode_ci can store and compare emoji and rare characters.

Compared with the older utf8mb4_general_ci, utf8mb4_unicode_ci often offers more accurate and predictable results for non-English text, at the cost of slower performance and larger index sizes in some workloads. It is not a locale-specific collation; it provides a general, language-blind set of rules that are still intended to be reasonably intuitive for many languages.

Usage and considerations: set the collation at the database, table, or column level, and ensure the connection (client) charset/collation match to avoid implicit conversions. Changing collations can affect ORDER BY results and index usage, so testing is advised. Other modern options in MySQL include utf8mb4_0900_ai_ci and related collations, which offer different strengths, such as accent-insensitivity.