Reference: Coda Datatypes
Coda Base Datatypes
Coda contains a small but useful number of basic types that can be used to describe data. If necessary, user-defined types are also allowed. Coda is different than SQL in that all data literals are single-quoted in DML and SELECT statements. This includes numeric types and timestamps.
Any text string that is less than 255 characters.
A whole number, either positive or negative. Specific limits are imposed by the deployment database platform.
A floating-point number, either positive or negative. Specific limits are imposed by the deployment database platform.
A bit datatype, either 0 or 1.
An arbitrary character string. Implemented as a CLOB at the deployment level, limits on the length are imposed by the deployment database platform.
A structured text field of variable length. The FILE must contain a name followed by a semicolon, MIME type followed by a semicolon, and a Base64-encoded version of the binary data of the file.
A reference to a row in another table or form. Implemented as an integer.
A Unix Timestamp.
In many settings such as CREATE TABLE or CREATE PROCEDURE statements, an optional ARRAY flag can be specified on a datatype to indicate that an array of data is expected instead of a scalar. To specify an array literal, use the format ['DATA', 'DATA2', ...].