create_test_data_set

#!/bin/bash

# CBDA validation and training set creation.

# Create a test data set for testing and debugging of script
# create-training-set.py.

# The created file has a header line and rowCount data lines, each with
# columnCount comma separated fields.
#
# The headers in the header line are of the form "hc" where "c" is the column
# ordinal. If columnCount is 3 then the header line is "h1,h2,h3".
#
# The data items in the data lines are of the form "r.c" where r is the row
# ordinal and c is the column ordinal. If columnCount is 3 then the first data
# line is "2.1,2.2,2.3", since it is the 2nd line of the file, after the header
# line. Similarly the 2nd data line is "3.1,3.2,3.3", etc.

rowCount=${1?No row count specified}
columnCount=${2?No column count specified}
outFile=${3?No output file name specified}

rm -f $outFile

# Write header line.
for (( c = 1; c <= $columnCount; ++c ))
do
	if (( c > 1 )); then
		 echo -n "," >> $outFile
	fi
	echo -n "h$c" >> $outFile
done
echo >> $outFile

# Write data lines.
# Start at 2 since the first data line is the 2nd line of the file, after the
# header line.
for (( r = 2; r <= $rowCount+1; ++r ))
do
	# Write a data line.
	for (( c = 1; c <= $columnCount; ++c ))
	do
		if (( c > 1 )); then
			 echo -n "," >> $outFile
		fi
		echo -n "$r.$c" >> $outFile
	done
	echo >> $outFile
done