Add AWK scripts for computing basic statistics

kgryte · kgryte · commit 576d6be2c590 · 2016-10-17T17:37:30.000-07:00
diff --git a/tools/awk/count.awk b/tools/awk/count.awk
@@ -0,0 +1,21 @@
+#!/usr/bin/env awk -f
+#
+# Computes the number of values.
+#
+# Usage: count
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - number of values
+
+BEGIN {
+	count = 0
+}
+{
+	count++
+}
+END {
+	print count
+}
diff --git a/tools/awk/max.awk b/tools/awk/max.awk
@@ -0,0 +1,24 @@
+#!/usr/bin/env awk -f
+#
+# Computes the maximum value.
+#
+# Usage: max
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - maximum value
+
+!i++ {
+	# Only for the first record:
+	max = $1
+}
+{
+	if ( $1 > max ) {
+		max = $1
+	}
+}
+END {
+	print max
+}
diff --git a/tools/awk/mean.awk b/tools/awk/mean.awk
@@ -0,0 +1,30 @@
+#!/usr/bin/env awk -f
+#
+# Computes the arithmetic mean.
+#
+# Usage: mean
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - arithmetic mean
+#
+# Notes:
+#   - Uses [Welford's method][1].
+#
+# [1]: https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Online_algorithm
+
+BEGIN {
+	N = 0
+	mean = 0
+	delta = 0
+}
+{
+	N++
+	delta = $1 - mean
+	mean += delta / N
+}
+END {
+	print mean
+}
diff --git a/tools/awk/median.awk b/tools/awk/median.awk
@@ -0,0 +1,27 @@
+#!/usr/bin/env awk -f
+#
+# Computes the median.
+#
+# Usage: median
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - median value
+
+BEGIN {
+	i = 0
+}
+{
+	a[i++] = $1
+}
+END {
+	j = i/2
+	if ( i%2 == 1 ) {
+		median = a[int(j)]
+	} else {
+		median = (a[j] + a[j-1])/2
+	}
+	print median
+}
diff --git a/tools/awk/min.awk b/tools/awk/min.awk
@@ -0,0 +1,24 @@
+#!/usr/bin/env awk -f
+#
+# Computes the minimum value.
+#
+# Usage: min
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - minimum value
+
+!i++ {
+	# Only for the first record:
+	min = $1
+}
+{
+	if ( $1 < min ) {
+		min = $1
+	}
+}
+END {
+	print min
+}
diff --git a/tools/awk/sum.awk b/tools/awk/sum.awk
@@ -0,0 +1,21 @@
+#!/usr/bin/env awk -f
+#
+# Computes the sum.
+#
+# Usage: sum
+#
+# Input:
+#   - a column of numbers
+#
+# Output:
+#   - sum
+
+BEGIN {
+	sum = 0
+}
+{
+	sum += $1
+}
+END {
+	print sum
+}
