Dmitry, > > In Pavel's use case "color" is a VarChar as such any value/string/variable > which is assigned to it should be cast as a VarChar, regardless of the > intermediate datatype. > > > > The current outcome is wrong! > > The SQL committee respectfully disagrees.
Actually, given that MS SQL, Oracle, DB2 and Postgres all generate the results that Pavel and I expect, I would say that it is our interpretation of the standard which is wrong. > Suggestion to describe literals as VARCHARs may fix the CASE issue but break > something else that used to work correctly. Do you care? Any change that we make can break something, that is always a possibility. It is the developers responsibility to test their application when they are looking to adopt any new version of anything (tool, database, components...). ------------------------------------------------------------------------------ Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
