Unicode Collation Algorithm based collations

1. Language
2. Region
3. Script
4. Keyword

1. L[ISO 639-1 language code]
2. R[ISO 3166 country/region code]
3. Z[ISO 15924 script code]
4. K[name]

The Locale attribute is probably the most important attribute to obtain ordering that conforms to the user expectations in different countries and regions. You need to explicitly specify the Locale attribute to properly collate text for a specific language.

The Locale attribute consists of the following parts: language, region/country, script, and keyword. Not all the parts are mandatory. See Table 2 for a complete list of the valid combinations. The specification of a locale automatically presets all the other collation attributes to values that are suitable for that locale. Typically there is no need to specify additional collation attribute.

The Strength attribute determines whether accent or case is taken into account when collating or comparing text strings. In writing systems without case or accent, the Strength attribute controls similarly important features.

Almost all characters can be distinguished by the first three strength levels, therefore in most locales the default Strength attribute is set at the tertiary level. However if the Alternate attribute (described in a following row) is set to shifted, then the quaternary strength level can be used to break ties among white space characters, punctuation marks, and symbols that would otherwise be ignored. The identity strength level is used to distinguish among similar characters, such as the MATHEMATICAL BOLD SMALL A character (U+1D41A) and the MATHEMATICAL ITALIC SMALL A character (U+1D44E).

Setting the Strength attribute to higher level will slow down text string comparisons and increase the length of the sort keys.

• X (Off)
• O (On)

Setting the Case Level attribute to on and the Strength attribute to primary level will ignore accent but not case. The Case Level attribute is set to X by default in most locales. When this attributes is set to O, it will slightly affect text string comparisons performance and lengthen the sort keys.

The Case First attribute controls whether upper case characters collate before or after lower case characters, in the absence of other differences in the two text strings.

The possible values are upper case first (U), lower case first (L), and off (X). There is almost no difference between the lower case first setting and the off setting, therefore typically there is no need to use the lower case first setting.

Specifying a Case First attribute of U or L can increase the length of the sort keys.

The Alternate attribute controls the handling of variable characters in the UCA: white space, punctuation marks, and symbols.

If the Alternate attribute is set to non-ignorable (N), then differences among these variable characters are of the same importance as differences among non-variable characters such as the English alphabet. If the Alternate attribute is set to shifted (S), then these variable characters are of only minor importance. If the Alternate attribute is set to shifted and the Strength attribute is set to the quaternary level, then variable characters are considered in a comparison when all other aspects of the strings - base letters, accents, and case - are identical.

The default for most locales is non-ignorable.

If shifted is selected, performance will be slower if there are many strings that are identical except for punctuation marks. Sort key length will not be affected unless the strength level is also increased.

The Variable Top attribute controls which characters to ignore, and is only meaningful if the Alternate attribute is set to Shifted. All characters whose primary weight is equal or lower than the specified character are considered ignorable.

The character is specified as one or two UTF-16BE code units in hexadecimal notation. A Unicode supplementary character is specified using a surrogate pair. For example, if you want to ignore white space characters and not visible characters, then set the Alternate attribute to Shifted and this attribute to U+0020 (space) or U+3000 (ideographic space). Since all characters having the same primary weight are equivalent, so setting this attribute to U+0020 is equivalent to setting it to U+3000.

This attribute alone has little impact on text string comparison performance, but setting it higher makes sort keys longer.

• X (Off)
• O (On)

• X (Off)
• O (On)

The French sorts strings by examining the accents starting from the end of the string. This attribute is automatically set to O for the French locales, and has a minor performance cost for text string comparisons, but no change in the length of the sort keys.

• X (Off)
• O (On)

The Hiragana attribute determines whether to distinguish between Japanese Hiragana and Katakana characters. To conform with the Japanese JIS X 4061 standard, you need to set this attribute to O and the Strength attribute to the quaternary level. This will, however, slow down text string comparisons and increase the length of the sort keys.