DYdusnfs> IIRC, unicode derived case/accent insensitive collations use six bytes
DYdusnfs> per character encoding. This gives us 2010 bytes which is pretty near
DYdusnfs> the 1/4 page size. Given that some overhead should be taken into
DYdusnfs> account, the key length may in fact overflow the limit.

DYdusnfs> I'm not absolutely sure this is your case, but it could be.
DYdusnfs> Dmitry

If this is true, Ivan needs to update his calculator:
http://mujweb.cz/iprenosil/interbase/ip_ib_indexcalculator.htm

[]s
Carlos
Firebird Performance in Detail - http://videos.firebirddevelopersday.com
www.firebirdnews.org - www.FireBase.com.br

DYdusnfs> 30.09.2014 19:19, Geoff Worboys wrote:

>>
>> CREATE INDEX MyIndex1 ON MyTable1 (Field1, Field2);
>>
>> where Field1 is VARCHAR(80)
>> Field2 is VARCHAR(255)
>>
>> both use the database default CHARACTER SET WIN1252 with
>> collation WIN1252_NOCASE as defined like this:
>>
>> CREATE COLLATION WIN1252_UNICODE
>> FOR WIN1252^
>>
>> CREATE COLLATION WIN1252_NOCASE
>> FOR WIN1252
>> FROM WIN1252_UNICODE
>> CASE INSENSITIVE^
>>
>> As it happens, I'm happy not to have this particular index,
>> but I'd like to understand what's going on.