The file infostats.tar.gz contains Arc Macro Language and C programs for release 7.0 of the Environmental Research Institute's geographic information system ARC/INFO that are designed to aid in analysis of data.

The file should be unpacked with the UNIX commands:

% gunzip infostats.tar.gz
% tar -xf infostats.tar


The tar file will extract into a subdirectory "cartography" in the current directory. The use of the AMLs can be simplified by extracting the file in $ARCHOME. Adding the line "&atool $ARCHOME/cartography" to the appropriate $ARCHOME/stations file (for X windows, $ARCHOME/stations/stat_9999) will make the AMLs available to all users on the system, after the &station AML directive is given (this can be in $HOME/.arc to run everytime Arc is started by an individual, or in $ARCHOME/startup/arc.aml — if one terminal type is always used).

The AML uses a system program that is searched for in $ARCHOME/cartography/bin; if no executable is found, then the system path is used. In either case, the executable must be compiled. This can be done on a system with a c compiler (cc or gcc). Change into the cartography/bin directory, edit the Makefile (the only line that may need to be changed is the "CC =" line, it defaults to the GNU c compiler, gcc), and type make. If make crashes over 'ranlib' (this will occur on Data General Aviions, which do not have a ranlib command), then copy the do-nothing shell script, ranlib, to a $PATH directory and run make again. If the tools are installed in and make is run in $ARCHOME/cartography/bin, nothing else needs to be done. If not, then during the edit, change the "SRCDIR =" and "TRGTDIR =" lines; the TRGTDIR must be in the system PATH (set or setenv will show this) and make should be run as "make install".

This code is placed in the public domain, and support is not (in general) available. Bugs can be reported to the address below, but see the legal notice, and keep in mind I have no way of testing any fix.

I derived much of the statistical code from Press, William H., Brian P. Flannery, Saul A. Teukolsky and William T. Vetterling. 1988. Numerical Recipies in C: The Art of Scientific Computing. Cambridge University Press: New York. Changes from that code include use of doubles and longs instead of floats and ints, and correcting the array subscript error (C arrays are 0 to n-1, not 1 to n) that I found in many of the for loops.