Transforming General Coordinate Data

As well as providing routines for applying compiled mappings to 1- and 2-dimensional data (Sections  & ), the TRANSFORM facility also has a set of routines for applying more general mappings. These are appropriate, for instance, when the number of input or output data coordinates is greater than 2 (or when these numbers are unequal) or when the number of coordinates is not known in advance. In such cases, all the input and output coordinates must each reside in single data arrays.

These general routines have names of the form TRN_TRNx, where x is either I, R or D according to whether the data are specified by integer, real or double precision values respectively. Thus, to transform a general set of data points specified by an array of real coordinates, the routine TRN_TRNR would be used, thus:

CALL TRN_TRNR( BAD, ND1, NCIN, NDAT, DATA, ID, NR1, NCOUT, RESULT, STATUS )

where:

• BAD is a logical value specifying whether the input coordinates may contain bad values (see Section );

• ND1 specifies the first dimension of the DATA array;

• NCIN specifies the number of coordinates per input data point;

• NDAT specifies the number of data points to be transformed;

• DATA is a 2-dimensional array containing a list of coordinates for the input data points;

• ID is an identifier associated with the compiled { $NCIN \rightarrow NCOUT$} mapping to be applied;

• NR1 specifies the first dimension of the RESULT array;

• NCOUT specifies the number of coordinates per output data point;

• RESULT is a 2-dimensional array to receive a list of (transformed) coordinates for the output data points;

• STATUS is an inherited error status variable.