Blocked fraction as function of domain length

blocked

#!/usr/bin/octave

pkg load image;
pkg load statistics;

pop = load("ru.hist");
obs = load("backorder.hist");
obs = padarray(obs, max(0, size(pop)-size(obs)), "post");
[_ var] = binostat(pop, obs./pop);

errorbar(obs./pop, sqrt(var)./pop, "xk");
xlabel("Domain length");
ylabel("Blocked fraction");
ylim([0 0.01]);

printf("png... "); print -dpng "-S1200,600" blocked.png;
printf("svg... "); print -dsvg "-S1200,600" blocked.svg;
printf("done\n");

hist

#!/bin/awk -f

BEGIN {
	maxlen = 0;
}

{
	hist[length($1)]++
	if (maxlen < length($1)) maxlen = length($1)
}

END {
	for (i = 1; i <= maxlen; i++) { hist[i] += 0; print hist[i]; }
}

Makefile

.PHONY: all clean

all: blocked.png blocked.svg
clean:
	rm -f backorder.hist ru.hist blocked.png blocked.svg

blocked.png blocked.svg: blocked ru.hist backorder.hist
	./$<

backorder.hist: backorder.csv hist
	tail +2 $< | cut -d';' -f1 | \
	sed -Ene 's/^([-0-9A-Za-z]*)\.ru/\1/p' | \
	tr A-Z a-z | uniq | ./hist >$@

ru.hist: ru.zone.txt.xz hist
	pv $< | xzcat | \
	sed -Ene 's/^([-0-9A-Za-z]*)\.[Rr][Uu]\.\s+[0-9]+\s+IN.*/\1/p' | \
	tr A-Z a-z | uniq | ./hist >$@