Unicode Sort Order

The Unicode sort order is different from the Replication Server sort order, and must be set independently.

To set the Unicode sort order, use a text editor to add the following line to the Replication Server configuration file:

RS_unicode_sort_order=unicode_sort_order

Make sure that you suspend the connection to the data server and shut down Replication Server before changing the Unicode sort order.

To change the current sort order, use the procedure described in “Changing the character set and sort order.”

Supported Unicode Sort Orders
Name	Description
defaultml	UTF-16 default ML
altnoacc	CP850 alt: no accent
altdict	cp850 alt: lowercase first
altnocsp	CP850 alt: no case preference
scandict	CP850 Scandinavian dictionary
scannocp	CP850 Scandinavian, no case preference
binary	(Default) UTF-16 binary
dict	Latin-1 English dictionary
nocase	Latin-1 English, no case
nocasep	Latin-1 English, no case preference
noaccent	Latin-1 English, no accent
espdict	Latin-1 Spanish dictionary
espnocs	Latin-1 Spanish no case
espnoac	Latin-1 Spanish, no accent
rusnocs	8859-5 Russian, no case
cyrnocs	8859-5 Cyrillic, no case
elldict	8859-7 Greek dictionary
hundict	8859-2 Hungarian dictionary
hunnoac	8859-2 Hungarian, no accents
hunnocs	8859-2 Hungarian, no case
turknoac	8859-9 Turkish, no accent
turknocs	8859-9 Turkish, no case
thaidict	CP874 Thai dictionary
utf8bin	UTF-16 ordering matching UTF-8

You can also specify a Unicode sort order for rs_subcmp. See Replication Server Reference Manual > rs_subcmp.