> Sort order depends on the locale used in initdb. If you want data sorted
> by the codes used to represent the data, then you might want to initdb
> with a locale of "C". Doing an initdb will require a dump and reload.
Bruno, thank you.
SHOW ALL command returns the following:
"client_encoding";"UNICODE"
"lc_collate";"Estonian_Estonia.1257"
"lc_ctype";"Estonian_Estonia.1257"
"lc_messages";"Estonian_Estonia.1257"
"lc_monetary";"Estonian_Estonia.1257"
"lc_numeric";"Estonian_Estonia.1257"
"lc_time";"Estonian_Estonia.1257"
Unfortunately, the sort order is incorrect:
"A"
"S"
"B"
"C"
"�"
"�"
"�"
"�"
"D"
"E"
"F"
"G"
accented charactes must be at the end of alphabet.
Why Postgres uses VERY stange sort order ? I do'nt believe that this order
exists in any locale.
Also, UPPER() function causes error
ERROR: invalid multibyte character for locale
HINT: The server's LC_CTYPE locale is probably incompatible with the
database encoding.
Is it possible to fix the sort order and use UPPER() function in this
locale without dump and reload ?
Andrus.