tools/parse-labor-wiki-termine/parse-wiki

#!/usr/bin/env ruby


WIKIPAGE    = "Termine"
WIKIURL     = "http://www.das-labor.org:80/wiki/Spezial:Exportieren/Termine"
WIKIPREFIX  = "http://www.das-labor.org/wiki/"

XMLFILE=ARGV[0]+"/termine.xml";

WIKILINES = [
#  /\W+(?:<DATE>\d{2}.\d{2}.\d{2})\W+(?:<DAY>\w\w)\W+(?:<TEXT>.*)/,
  /^\s*\*\W+(?:<DAY>\d{2}).(?:<MONTH>\d{2}).(?:<YEAR>\d{2})\W+(?:<WDAY>\w\w)\W+(?:<HOUR>\d{1,2}):(?:<MIN>\d{2})[ ,;-]+(?:<TEXT>.*)/,
  /^\s*\*\W+(?:<WDAY>\w\w)\W+(?:<DAY>\d{2}).(?:<MONTH>\d{2}).(?:<YEAR>\d{2})\W+(?:<HOUR>\d{1,2}):(?:<MIN>\d{2})[ ,;-]+(?:<TEXT>.*)/
]

#############################################################################
# Starts Here
require 'http-access2'
require 'rexml/document'
require 'htmlentities/lib/htmlentities.rb'
require 'namedregexp.rb'

class LaborTermine 
  attr_reader "termine"

  def initialize()
     @termine = Array.new;
  end

  def getWikiContent
    proxy = ENV['HTTP_PROXY']
    clnt  = HTTPAccess2::Client.new(proxy)
    txt   = clnt.get(WIKIURL).content
    return HTMLEntities.decode_entities(txt);
  end

  def parseLine(line)
    line.gsub!("'''", "");
    WIKILINES.each { |re|
      nre = NamedRegexp.new(re);
      m = nre.match(line);
      return m if !m.nil?
    }
    return nil
   end

  def parseWikiContent(wikiXml)
    doc = REXML::Document.new(wikiXml)

    text = doc.elements["/mediawiki/page/revision"];
    text.to_s.each_line { |line|
      m = parseLine(line);
      if m then
        aterm = Hash.new;
  	m.each { |a|
		aterm[a[0].downcase] = a[1];
	}
	# lr = /\[\[\s*(.*?)\s*\]\]/; if lr.match( aterm["text"] ) then 
	if  /\[\[\s*(.+?)\s*\|\s*(.+?)\]\]/.match( aterm["text"] ) then
		wikipage=$1;
    aterm["text"] = $2;
		wikipage.gsub!( " ", "_" );
		aterm["link"] = WIKIPREFIX+wikipage;
		aterm["text"].sub!("[[", "");
		aterm["text"].sub!("]]", "");
	elsif  /\[\[\s*(.*?)\s*\]\]/.match( aterm["text"] ) then
		wikipage=$1;
		wikipage.gsub!( " ", "_" );
		aterm["link"] = WIKIPREFIX+wikipage;
		aterm["text"].sub!("[[", "");
		aterm["text"].sub!("]]", "");
	elsif  /^(.*?)\[(http:\S+)\s+(.*)\](.*)$/.match(aterm["text"]) then	 # comment
		aterm["text"] = $1+" "+$3+$4;
		aterm["link"] = $2;
	end
	@termine.push aterm;
      end
    }
  end

  # Main
  def run
    wikiXml = getWikiContent
    parseWikiContent(wikiXml)
  end
end


####
# MAIN
#

class Array
  def to_h
    Hash[*self]
  end
end


lt = LaborTermine.new;
lt.run;

File.open( XMLFILE, "w" ) { |f|
  f.puts "<termine>"
  lt.termine.each { |termin|
    puts "#{termin["day"]}.#{termin["month"]}.#{termin["year"]} #{termin["hour"]}:#{termin["min"]} #{termin["text"]}"
    f.puts "<termin>"
      termin.each { |key,val|
        f.puts "  <#{key}>#{val}</#{key}>"
    	}
    f.puts "</termin>"
  }
  f.puts "</termine>"
}