Unicode collation acts as a sort order for Unicode data. This is separate from the sort order for non-Unicode data.
A Unicode collation consists of a locale and several comparison styles. Locales are usually named after countries or cultural regions. They sort characters as is standard in that area. The Unicode collation still provides a sort order for all characters in the Unicode standard, but precedence is given to the locale specified. Here are the locales available in Microsoft® SQL Server™.
| Locale ID | Name |
|---|---|
| 1033 | General Unicode |
| 33280 | Binary Order |
| 1027 | Catalan |
| 197636 | Chinese Bopomofo (Taiwan) |
| 2052 | Chinese Punctuation |
| 133124 | Chinese Stroke Count |
| 1028 | Chinese Stroke Count (Taiwan) |
| 1050 | Croatian |
| 1029 | Czech |
| 1043 | Dutch |
| 1061 | Estonian |
| 1036 | French |
| 66615 | Georgian Modern |
| 1031 | German |
| 66567 | German Phone Book |
| 1038 | Hungarian |
| 66574 | Hungarian Technical |
| 1039 | Icelandic |
| 1040 | Italian |
| 1041 | Japanese |
| 66577 | Japanese Unicode |
| 1042 | Korean |
| 66578 | Korean Unicode |
| 1062 | Latvian |
| 1063 | Lithuanian |
| 1071 | Macedonian |
| 1044 | Norwegian/Danish |
| 1045 | Polish |
| 1046 | Portuguese |
| 1048 | Romanian |
| 1051 | Slovak |
| 1060 | Slovenian |
| 1034 | Spanish Traditional |
| 3082 | Spanish Modern |
| 1053 | Swedish/Finnish |
| 1054 | Thai |
| 2057 | UK English |
| 1058 | Ukrainian |
| 1066 | Vietnamese |
The general Unicode collation locale is compatible with many locales that are not available individually in SQL Server. Choose the general Unicode collation for any of these locales.
| Afrikaans | Faeroese | Malay |
| Albanian | Farsi | Russian |
| Arabic | Georgian | Serbian |
| Basque | Greek | Swahili |
| Bulgarian | Hebrew | Urdu |
| Belarusian | Hindi | |
| English | Indonesian |
The four comparison styles specify whether or not to ignore the differences among similar characters. If the case-insensitive comparison style is chosen, for example, the characters A and a are considered equal. The width-insensitive and Kana-insensitive comparison styles are relevant only to characters found in certain East Asian languages.
During installation, you must specify a Unicode collation. The default value is based on the character set and sort order already chosen. You may choose a collation other than the default, but exercise this option with care. When a different value is chosen:
Caution When upgrading an existing SQL Server version 6.x installation to SQL Server 7.0, always choose the default Unicode collation. The master database in SQL Server 7.0 contains Unicode columns that were non-Unicode in the 6.x installation. If a nondefault Unicode collation is chosen that sorts data differently than it was sorted in the 6.x installation, uniqueness constraints may be violated and conversion of SQL Server 6.x user objects to SQL Server 7.0 may fail.
| Unicode Data | Using Unicode Data |