This section describes the collations available for Unicode character sets and their differentiating properties. For general information about Unicode, see Section 1.9, “Unicode Support”.
MySQL supports multiple 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.
utf8: An alias for
ucs2: The UCS-2 encoding of the Unicode character set using two bytes per character.
utf16: The UTF-16 encoding for the Unicode character set using two or four bytes per character. Like
ucs2but with an extension for supplementary characters.
utf32: The UTF-32 encoding for the Unicode character set using four bytes per character.
utf32 support Basic Multilingual Plane (BMP)
characters and supplementary characters that lie outside the
only BMP characters.
Most Unicode character sets have a general collation (indicated
_general in the name or by the absence of
a language specifier), a binary collation (indicated by
_bin in the name), and several
language-specific collations (indicated by language specifiers).
For example, for
utf8mb4_bin are its general and binary
utf8mb4_danish_ci is one of
its language-specific collations.
MySQL implements the
collations according to the Unicode Collation Algorithm (UCA)
collation uses the version-4.0.0 UCA weight keys:
collations have only partial support for the Unicode Collation
Algorithm. Some characters are not supported, and combining
marks are not fully supported. This affects primarily
Vietnamese, Yoruba, and some smaller languages such as Navajo.
A combined character is considered different from the same
character written with a single unicode character in string
comparisons, and the two characters are considered to have a
different length (for example, as returned by the
CHAR_LENGTH() function or in
result set metadata).
MySQL implements language-specific Unicode collations if the ordering based only on the Unicode Collation Algorithm (UCA) does not work well for a language. Language-specific collations are UCA-based, with additional language tailoring rules. Examples of such rules appear later in this section. For questions about particular language orderings, unicode.org provides Common Locale Data Repository (CLDR) collation charts at http://www.unicode.org/cldr/charts/30/collation/index.html.
A language name shown in the following table indicates a language-specific collation. Unicode character sets may include collations for one or more of these languages.
Table 1.3 Unicode Collation Language Specifiers
Danish collations may also be used for Norwegian.
For Classical Latin collations,
J compare as equal, and
V compare as
Spanish collations are available for modern and traditional
Spanish. For both,
ñ (n-tilde) is a
separate letter between
o. In addition, for traditional Spanish,
ch is a separate letter between
ll is a separate letter between
Traditional Spanish collations may also be used for Asturian and Galician.
Swedish collations include Swedish rules. For example, in Swedish, the following relationship holds, which is not something expected by a German or French speaker:
Ü = Y < Ö
For any Unicode character set, operations performed using the
collation are faster than those for the
collation. For example, comparisons for the
utf8_general_ci collation are faster, but
slightly less correct, than comparisons for
utf8_unicode_ci. The reason is that
utf8_unicode_ci supports mappings such as
expansions; that is, when one character compares as equal to
combinations of other characters. For example,
ß is equal to
German and some other languages.
utf8_unicode_ci also supports contractions
and ignorable characters.
is a legacy collation that does not support expansions,
contractions, or ignorable characters. It can make only
one-to-one comparisons between characters.
To further illustrate, the following equalities hold in both
utf8_unicode_ci (for the effect of this in
comparisons or searches, see
Section 1.8.6, “Examples of the Effect of Collation”):
Ä = A Ö = O Ü = U
A difference between the collations is that this is true for
ß = s
Whereas this is true for
which supports the German DIN-1 ordering (also known as
ß = ss
collations if the ordering with
utf8_unicode_ci does not work well for a
language. For example,
works fine for German dictionary order and French, so there is
no need to create special
utf8_general_ci also is satisfactory for
both German and French, except that
s, and not to
ss. If this is acceptable for your
application, you should use
because it is faster. Otherwise, use
utf8_unicode_ci because it is more
For all Unicode collations except the
(binary) collations, MySQL performs a table lookup to find a
character's collating weight. If a character is not in the
table (for example, because it is a “new”
character), collating weight determination becomes more
For BMP characters in general collations (
), the weight is the code point.
For BMP characters in UCA collations (for example,
and language-specific collations), the following algorithm applies:
if (code >= 0x3400 && code <= 0x4DB5) base= 0xFB80; /* CJK Ideograph Extension */ else if (code >= 0x4E00 && code <= 0x9FA5) base= 0xFB40; /* CJK Ideograph */ else base= 0xFBC0; /* All other characters */ aaaa= base + (code >> 15); bbbb= (code & 0x7FFF) | 0x8000;
The result is a sequence of two collating elements,
U+04cf CYRILLIC SMALL LETTER PALOCHKAcurrently is, with all UCA collations, greater than
U+04c0 CYRILLIC LETTER PALOCHKA. Eventually, after further collation tuning, all palochkas will sort together.
For supplementary characters in general collations, the weight is the weight for
0xfffd REPLACEMENT CHARACTER. For supplementary characters in UCA 4.0.0 collations, their collating weight is
0xfffd. That is, to MySQL, all supplementary characters are equal to each other, and greater than almost all BMP characters.
An example with Deseret characters and
CREATE TABLE t (s1 VARCHAR(5) CHARACTER SET utf32 COLLATE utf32_unicode_ci); INSERT INTO t VALUES (0xfffd); /* REPLACEMENT CHARACTER */ INSERT INTO t VALUES (0x010412); /* DESERET CAPITAL LETTER BEE */ INSERT INTO t VALUES (0x010413); /* DESERET CAPITAL LETTER TEE */ SELECT COUNT(DISTINCT s1) FROM t;
The result is 2 because in the MySQL
collations, the replacement character has a weight of
0x0dc6, whereas Deseret Bee and Deseret Tee both have a weight of
0xfffd. (Were the
utf32_general_cicollation used instead, the result is 1 because all three characters have a weight of
0xfffdin that collation.)
The rule that all supplementary characters are equal to each other is nonoptimal but is not expected to cause trouble. These characters are very rare, so it is very rare that a multi-character string consists entirely of supplementary characters. In Japan, since the supplementary characters are obscure Kanji ideographs, the typical user does not care what order they are in, anyway. If you really want rows sorted by the MySQL rule and secondarily by code point value, it is easy:
ORDER BY s1 COLLATE utf32_unicode_ci, s1 COLLATE utf32_bin
There is a difference between “ordering by the
character's code value” and “ordering by the
character's binary representation,” a difference that
appears only with
utf16_bin, because of
utf16_bin (the binary
utf16) was a binary
comparison “byte by byte” rather than
“character by character.” If that were so, the
order of characters in
differ from the order in
example, the following chart shows two rare characters. The
first character is in the range
FFFF, so it is
greater than a surrogate but less than a supplementary. The
second character is a supplementary.
Code point Character utf8 utf16 ---------- --------- ---- ----- 0FF9D HALFWIDTH KATAKANA LETTER N EF BE 9D FF 9D 10384 UGARITIC LETTER DELTA F0 90 8E 84 D8 00 DF 84
The two characters in the chart are in order by code point
0x10384. And they are in order by
utf8 value because
0xf0. But they are not in order by
utf16 value, if we use byte-by-byte
utf16_bin collation is not
“byte by byte.” It is “by code
point.” When MySQL sees a supplementary-character
utf16, it converts to the
character's code-point value, and then compares. Therefore,
are the same ordering. This is consistent with the SQL:2008
standard requirement for a UCS_BASIC collation:
“UCS_BASIC is a collation in which the ordering is
determined entirely by the Unicode scalar values of the
characters in the strings being sorted. It is applicable to
the UCS character repertoire. Since every character repertoire
is a subset of the UCS repertoire, the UCS_BASIC collation is
potentially applicable to every character set. NOTE 11: The
Unicode scalar value of a character is its code point treated
as an unsigned integer.”
If the character set is
ucs2, comparison is
ucs2 strings should not
contain surrogates, anyway.