This function translates a string from a given source charset to a destination charset. It provides a generic way of recoding string entities.
The src_charset may be a narrow or a wide string. If it's a narrow string (VARCHAR) then the src_charset is taken into account and defines the current encoding of the src_string. In any other case src_charset is ignored.
src_charset and dst_charset are names of system-defined 8 bit charset tables. Use charsets_list to obtain a list of currently defined character sets and aliases. If either of these is null, then the charset in effect is used. There are two special character set names - "UTF-8" and "_WIDE_" - that are recognized by this function. These represent UTF-8 encoding of characters and wide string (NVARCHAR).
select cast (charset_recode ('\xA9', 'ISO-8859-1', 'UTF-8') as varbinary) -- converts "Copyright sign" to UTF-8 (output 0xC2A9) select cast (charset_recode ('\xC0', 'WINDOWS-1251', 'ISO-8859-5') as varbinary) -- converts "Cyrillic A" from WINDOWS-1251 charset to ISO-8859-5 (output 0xB0). select cast (charset_recode (N'\x410', '_WIDE_', 'WINDOWS-1251') as varbinary) -- converts "Cyrillic A" from Unicode to WINDOWS-1251 charset (result '\xC0'). select charset_recode (N'\x410', '_WIDE_', 'ISO-8859-1') -- converts "Cyrillic A" from Unicode to ISO-8859-1 charset (Not available : result '?').