MIF files contain a tokenized header, followed by a DATA section. The comma delimiter is required. If a DELIMITER token is present in the file, the file is not imported.
The next part of the file header contains the definitions of the columns in the MIF file. Feature names, as well as numeric data about the features, are stored here. One of these columns is used as the identifier to store with the object.
The column definitions begin with the COLUMN token, followed by the number of columns in the MIF file. Following the COLUMN token is a single line entry for each column. The first entry on the line is the column name, the second is the column type. The following example indicates three columns in the MIF file:
COLUMN 3
FEAT_NAME char (15)
POPULATION integer
AREA decimal (8,4)Any column type can be used as the name, but char is the most common, since that represents a string. The number in parentheses following char indicates the largest string (which cannot exceed 127 characters). For a complete listing of column types, contact MapInfo.
The token DATA indicates that the header is complete, and the listing of objects follows. For each object, a number of attributes may be listed, such as PEN and BRUSH. These settings are ignored. MapInfo supports a variety of object types, though only a few are used to represent geographical objects. Three object types are supported—Region, Polyline, and Point.
In a MIF file, latitude/longitude coordinates must be represented in decimal degrees, with up to six decimals of precision. Longitude is the x coordinate, and latitude is the y coordinate. Western longitudes and southern latitudes are always negated. This means that for the U.S., longitudes are negated, while latitudes are positive. User coordinates are entered as whole numbers, as the decimal portion is ignored.