Click here to view and discuss this page in DocCommentXchange. In the future, you will be sent there automatically.

SQL Anywhere 11.0.1 (Français) » SQL Anywhere Server - Database Administration » Configuring Your Database » International languages and character sets » Character set and collation reference information » Turkish character sets and collations


Alternative Turkish collation 1254TRKALT

For some application developers, the Turkish letter I problem can cause significant problems. While the correct solution is to ensure that all object references are in the proper case or that the proper form of the letter I is used, in some cases it may be more expedient to make a decision to violate the Turkish rules in favor of the ANSI rules.

SQL Anywhere provides the collation 1254TRKALT, which is identical to 1254TRK, except that it makes I-dot and I-no-dot equivalent characters.

It is important to understand the consequences of this change. In a 1254TRKALT database, the following strings are equal:

The word Fig written with lowercase I-dot and I-no-dot.

This is not correct for a Turkish user, but may be acceptable in some cases.

The second issue appears when using ORDER BY. Consider the following strings:

Strings in the order I-dot a, I-no-dot a, I-no-dot s, and I-dot s.

In a 1254TRK database, an ORDER BY of the strings would produce the following:

Strings sorted in the order I-no-dot a, I-no-dot s, I-dot a, I-dot s.

because I-no-dot is less than I-dot. In a 1254TRKALT database, the order would be

The strings sorted in the order: I-dot a, I-no-dot a, I-no-dot s, I-dot s.

because I-no-dot is equal to I-dot.