LSF Participant Extraction

tn

#!/usr/bin/env ruby
# this scripts collects participant data from the "special info" list 
# - mainly, it eliminates doublettes (students who have been rejected and applied again)
# and keeps only the most successful application

inputfile = ARGV[0]
outputfile = ARGV[1]
unless inputfile # && outputfile
	puts "usage: tn.rb inputfile [outputfile]"
	exit 1
end

# some helper methods etc

Anmeldung = Struct.new(:matrikel, :studiengang, :nachname, :vorname, :status)

#   123456 - IMI (M) ( 920 )  	   Kleinen, Barbara    	AB
#1.	123456
#2.	IMI
#3.	Kleinen
#4.	Barbara
#5.	AB
REGEX = /\s*(\d{3,6}) - (\w+)\s+\(M\)\s+\(.*\)\s+([\w,ä,ü,ö,é, ]+),\s+([\w,ä,ü,ö,é,-]+)\s+(\w+)/
def split_line(line)
	m = REGEX.match(line)
	unless m
		puts "could not match line #{line}"
		exit 1
	end
	Anmeldung.new(*m[1,5])
end

# keep only the Listing with the highest precedence:
# ZU - Zugelassen (admitted)
# AN - Angemeldet (applied)
# AB - Abgelehnt - rejected
BIGGER = ['ZU','AN','AB'] 
def bigger(s1,s2)
	BIGGER.index(s1) < BIGGER.index(s2)
end

def add_anmeldung(h,anmeldung)
	key = anmeldung.nachname
	prev = h[key]
	if (prev == nil) || bigger(anmeldung.status,prev.status)
		h[key] = anmeldung
	end
end

# actual script begins here

File.open(inputfile,'r') do | file|
	h = {}
	file.each_line do | line |
		anmeldung = split_line(line)
		add_anmeldung(h,anmeldung)
	end
	counts =  Hash.new(0)
	begin
	    if (outputfile)
	    	out = File.open(outputfile,'w') 
	    else
	    	out = $stdout
	    end
	
		h.keys.sort.each do | nachname | 
			a = h[nachname]
			out.puts "#{a.status}, #{a.nachname}, #{a.vorname}, #{a.studiengang}, #{a.matrikel}"
			counts[a.status] += 1
			counts['total'] += 1
		end
		out.puts
		out.puts counts.inspect
	rescue
		raise
	ensure
		out.close if outputfile
	end
end