AggHAR example

Two files are supplied with AggHAR as an example. The sample input file is AGGEXM.HAR (also used for the ViewHAR aggregation example) and the sample file of supplementary mapping data is AGGSUP.HAR. These will likely both be located in the same directory as AggHAR.EXE. That directory will often be C:\GP.

To try AggHAR, copy both files to a temporary directory where you have a DOS box open, and type:

AggHAR AGGEXM.HAR AGGOUT.HAR AGGSUP.HAR

A host of messages should flash by, ending with 'Finished OK'.

Note the command will fail if there is already is a file called AGGOUT.HAR [a safety feature].

It is often useful to trap the screen messages to a file, so you can examine them. Do this by typing:

del AGGOUT.HAR

AggHAR AGGEXM.HAR AGGOUT.HAR AGGSUP.HAR >AGG.LOG

Then use a text editor to examine AGG.LOG.

Use ViewHAR to examine the output and both input files. The bulk of the mapping data in AGGSUP.HAR relates to weighted aggregation. Apart from that, all that is needed are the headers:

•ASEC the new aggregated set

•MIND the mapping from old set IND to ASEC

•MCOM the mapping from old set COM to ASEC

Notice from AGG.LOG that AggHAR works out by itself how to aggregate arrays dimensioned with set MAR -- see implied subset mappings.

There are also aggregated sets and mappings in AGGEXM.HAR: that file is also used for the ViewHAR aggregation example, which requires them. They are not used by AggHAR. The only set information that AggHAR gets from the main input file is the set labelling data which is part of each real array.

Finally, experiment with the U, P, and M options. For example, type:

del AGGOUT.HAR

AggHAR AGGEXM.HAR AGGOUT.HAR AGGSUP.HAR -P >AGG.LOG

The P option causes original set names to be be used on aggregated headers.

URL of this topic: www.copsmodels.com/webhelp/viewhar/hc_aggharexm.htm

Link to full GEMPACK Manual

Link to GEMPACK homepage