Documentation Home
MySQL 9.1 Reference Manual
Related Documentation Download this Manual
PDF (US Ltr) - 40.4Mb
PDF (A4) - 40.5Mb
Man Pages (TGZ) - 259.5Kb
Man Pages (Zip) - 366.7Kb
Info (Gzip) - 4.1Mb
Info (Zip) - 4.1Mb


12.9 Unicode Support

The Unicode Standard includes characters from the Basic Multilingual Plane (BMP) and supplementary characters that lie outside the BMP. This section describes support for Unicode in MySQL. For information about the Unicode Standard itself, visit theUnicode Consortium website.

BMP characters have these characteristics:

  • Their code point values are between 0 and 65535 (orU+0000 andU+FFFF).

  • They can be encoded in a variable-length encoding using 8, 16, or 24 bits (1 to 3 bytes).

  • They can be encoded in a fixed-length encoding using 16 bits (2 bytes).

  • They are sufficient for almost all characters in major languages.

Supplementary characters lie outside the BMP:

  • Their code point values are betweenU+10000 andU+10FFFF).

  • Unicode support for supplementary characters requires character sets that have a range outside BMP characters and therefore take more space than BMP characters (up to 4 bytes per character).

The UTF-8 (Unicode Transformation Format with 8-bit units) method for encoding Unicode data is implemented according to RFC 3629, which describes encoding sequences that take from one to four bytes. The idea of UTF-8 is that various Unicode characters are encoded using byte sequences of different lengths:

  • Basic Latin letters, digits, and punctuation signs use one byte.

  • Most European and Middle East script letters fit into a 2-byte sequence: extended Latin letters (with tilde, macron, acute, grave and other accents), Cyrillic, Greek, Armenian, Hebrew, Arabic, Syriac, and others.

  • Korean, Chinese, and Japanese ideographs use 3-byte or 4-byte sequences.

MySQL supports these Unicode character sets:

  • utf8mb4: A UTF-8 encoding of the Unicode character set using one to four bytes per character.

  • utf8mb3: A UTF-8 encoding of the Unicode character set using one to three bytes per character. This character set is deprecated andsubject to removal in a future release; useutf8mb4 instead.

  • utf8: A deprecated alias forutf8mb3; useutf8mb4 instead.

    Note

    utf8 is expected in a future version of MySQL to become an alias forutf8mb4.

  • ucs2: The UCS-2 encoding of the Unicode character set using two bytes per character. Deprecated; expect support for this character set to be removed in a future release.

  • utf16: The UTF-16 encoding for the Unicode character set using two or four bytes per character. Likeucs2 but with an extension for supplementary characters.

  • utf16le: The UTF-16LE encoding for the Unicode character set. Likeutf16 but little-endian rather than big-endian.

  • utf32: The UTF-32 encoding for the Unicode character set using four bytes per character.

Note

Theutf8mb3 character set is deprecated and you should expect it to be removed in a future MySQL release. Please useutf8mb4 instead.utf8 is currently an alias forutf8mb3, but it is now deprecated as such, andutf8 is expected subsequently to become a reference toutf8mb4. MySQL 9.1 also displaysutf8mb3 in place ofutf8 in the columns of Information Schema tables, and in the output of SQLSHOW statements.

In addition, you should be aware that collations using theutf8_ prefix in older releases of MySQL have since been renamed using the prefixutf8mb3_, instead.

To avoid ambiguity about the meaning ofutf8, consider specifyingutf8mb4 explicitly for character set references.

Table 12.2, “Unicode Character Set General Characteristics”, summarizes the general characteristics of Unicode character sets supported by MySQL.

Table 12.2 Unicode Character Set General Characteristics

Character SetSupported CharactersRequired Storage Per Character
utf8mb3,utf8 (deprecated)BMP only1, 2, or 3 bytes
ucs2BMP only2 bytes
utf8mb4BMP and supplementary1, 2, 3, or 4 bytes
utf16BMP and supplementary2 or 4 bytes
utf16leBMP and supplementary2 or 4 bytes
utf32BMP and supplementary4 bytes

Characters outside the BMP compare asREPLACEMENT CHARACTER and convert to'?' when converted to a Unicode character set that supports only BMP characters (utf8mb3 orucs2).

If you use character sets that support supplementary characters and thus are“wider” than the BMP-onlyutf8mb3 anducs2 character sets, there are potential incompatibility issues for your applications; seeSection 12.9.8, “Converting Between 3-Byte and 4-Byte Unicode Character Sets”. That section also describes how to convert tables from the (3-byte)utf8mb3 to the (4-byte)utf8mb4, and what constraints may apply in doing so.

A similar set of collations is available for most Unicode character sets. For example, each has a Danish collation, the names of which areutf8mb4_danish_ci,utf8mb3_danish_ci (deprecated),utf8_danish_ci (deprecated),ucs2_danish_ci,utf16_danish_ci, andutf32_danish_ci. The exception isutf16le, which has only two collations. For information about Unicode collations and their differentiating properties, including collation properties for supplementary characters, seeSection 12.10.1, “Unicode Character Sets”.

The MySQL implementation of UCS-2, UTF-16, and UTF-32 stores characters in big-endian byte order and does not use a byte order mark (BOM) at the beginning of values. Other database systems might use little-endian byte order or a BOM. In such cases, conversion of values needs to be performed when transferring data between those systems and MySQL. The implementation of UTF-16LE is little-endian.

MySQL uses no BOM for UTF-8 values.

Client applications that communicate with the server using Unicode should set the client character set accordingly (for example, by issuing aSET NAMES 'utf8mb4' statement). Some character sets cannot be used as the client character set. Attempting to use them withSET NAMES orSET CHARACTER SET produces an error. SeeImpermissible Client Character Sets.

The following sections provide additional detail on the Unicode character sets in MySQL.