How to determine field length in an unicode database

[Poul Møller Hansen]

I have created a db2 8.1 database with codeset UTF-8.
I thought the database was able to handle field legths as characters and not
as byte data then.
But I'm getting an error when inserting a 20 characters string containing
non ASCII characters
into a character 20 field.

Is there a good way to handle this, or do I have to estimate a field length
?
Poul

[W Gemini]

Poul Møller Hansen wrote:

I have created a db2 8.1 database with codeset UTF-8.
I thought the database was able to handle field legths as characters and not
as byte data then.
But I'm getting an error when inserting a 20 characters string containing
non ASCII characters
into a character 20 field.

Is there a good way to handle this, or do I have to estimate a field length
?
Poul

DB2 always uses bytes rather than characters for varchar. So a
varchar(20) is twenty bytes rather than 20 characters. Since one UTF-8
character can be 4 bytes long, use varchar(80) for the column definition.