INF: Using Double-byte Character Sets with SQL Server

  - Microsoft SQL Server version 4.2 for OS/2
  - Microsoft SQL Server version 4.2

SUMMARY

SQL Server version 4.2 for Windows NT or OS/2 does not utilize the double- byte or Unicode character sets. However, this does not prevent the storing of a numeric representation of a non-Roman character in a SQL datatype that is large enough to hold it.

There are some limitations to storing double-byte characters. One case would be if the second byte of a double-byte character were to collide with a special character such as a "%" or "'". With kanji, this would occur for about 2% of the kanji double-byte characters. Those characters could not be used for object names and could also cause problems when formulating Transact-SQL statements. You would also have problems with LIKE--again what if the second byte of a double-byte were the same numeric representation as the "%" wildcard used by LIKE.

MORE INFORMATION

For the reasons stated above, to handle dbcs, it is necessary to have the parser look ahead to see if the character is a single- or double-byte.