#!/bin/sh
#
# Generate textual and visual statistics of the debianforum.de
#
# depends on: w3m, awk, (gnuplot)
#
# markus schnalke <meillo@marmaro.de>, 2015-01, License: CC0


url=http://debianforum.de/forum/index.php 

if [ $# -lt 1 -o $# -gt 2 ] ; then
	echo "Usage: dfde-stats.sh out.txt [out.png]" >&2
	exit 1
fi

w3m -dump "$url" | awk '
	/[0-9] *Themen/ { ntopics = $1 }
	/[0-9] *Beiträge/ { nposts = $1 }
	/^    [^ ]/ {
		sub(/^ */, ""); sub(/ *$/, "")
		gsub(/ /, "_")
		name = $0
	}
	/^$/ {
		if (!ntopics) {
			next
		}
		printf("%d\t%d\t%.2f\t%s\n",
				ntopics, nposts, nposts/ntopics, name)
		ntopics = nposts = name = 0
	}
' >"$1"

test $# -ne 2 && exit

gnuplot <<!
	set terminal png size 1200,800 
	set output '$2'

	set style data histograms
	set style histogram cluster gap 1
	set style fill solid border -1
	set boxwidth 0.9

	set y2tics
	set tics nomirror
	set xtics rotate by -45
	set tics out
	set grid y
	set key left top

	plot '$1' using 1:xtic(4) ti 'threads' lc rgb "blue", \
		'' u 2 ti 'posts' lc rgb "skyblue", \
		'' u 3 ti 'ratio' axis x1y2 lc rgb "red" with lines
!