cdfcat −− concatenate several variable slices into one, increasing the dimension of all variable fields
cdfcat test_variable start_value end_value max_no_of_slices
Cdfcat expects a common data format file on the standard input. It writes a modified common data format file to standard output containing only one variable slice. This slice consists of fields increased in dimension by concatenation of some or all of the variable slices of the input file. The new dimension is given the name test_variable.
A zero dimensional variable field named test_variable is expected in the input. For input slices to be included in the concatenation, the value of test_variable must be in the range [start_value, end_value]. The value of test_variable is assumed to increase monotonically through the input file. Max_no_of_slices is an estimate of the number of slices to be expected in the concatenation. It is for memory allocation purposes only. If the number of slices to be concatenated exceeds this parameter, the program fails with an error message.
Two parameters are added to the output header, one of the form test_variable0, the other of the form dtest_variable. These are used by programs such as cdfplot to scale axes in plots.
The field with the name test_variable is now an index field (a one dimensional field with dimension name and field name the same), and is moved from the variable slice to the static slice in the output file.
kestrel%cdfcat time 0 101 21 < infile > outfile
In this example a scalar field named "time" is assumed, and all variable slices with time between 0 and 101 are concatenated. No more than 21 slices are assumed to fall into this time interval.
If any 4 dimensional fields occur in the input, cdfcat will fail, as this would require creation of a 5 dimensional field, which is not allowed in common data format files. Since cdfcat must read many slices into memory at once, it can be very demanding of memory.