MySQL collation for Kyrgyz: Difference between revisions
From FirespeakerWiki
Jump to navigationJump to search
Firespeaker (talk | contribs) mNo edit summary |
Firespeaker (talk | contribs) нңо, not ноң |
||
| Line 7: | Line 7: | ||
<collation name="utf8_cyr_kir" id="244"> | <collation name="utf8_cyr_kir" id="244"> | ||
<rules> | <rules> | ||
<reset before="primary">\ | <reset before="primary">\u04A3</reset> | ||
<p>\ | <p>\u043E</p> | ||
<reset before="primary">\u043F</reset> | <reset before="primary">\u043F</reset> | ||
<p>\u04E9</p> | <p>\u04E9</p> | ||
Revision as of 08:49, 31 December 2009
This page provides a UCA collation for Kyrgyz in mysql. The UCA itself was built using guidelines found on mysql's webiste, links from there, and some bugging of people on IRC (thanks), and a certain amount of experimenting.
The collation should go in the unicode character set definition in /usr/share/mysql/charsets/Index.xml:
<charset name="utf8">
...
<collation name="utf8_cyr_kir" id="244">
<rules>
<reset before="primary">\u04A3</reset>
<p>\u043E</p>
<reset before="primary">\u043F</reset>
<p>\u04E9</p>
<reset before="primary">\u0444</reset>
<p>\u04AF</p>
<reset before="primary">\u0451</reset>
<p>\u0435</p>
</rules>
</collation>
</charset>
The id can be anything from 0 to 255, but it should be unique. The mysql process will have to be restarted after modifying the file. The collation can then be applied to a table.