#!/usr/bin/env python
# -*- coding: iso8859-1 -*-
#
# Fulhack(tm) for KTH:s Lexin
# 
# Erik Waling <erikw@acc.umu.se>

import httplib
import string
import re
import sys
    
class Lexin:

    connection = None
    allowedLang = ['sve', 'eng', 'bos', 'fin', 'gre', 'kro', 'rys', 'spa', 'tur']
    fromLang = ''
    toLang = ''
    
    def __init__(self, url):
        self.connection = httplib.HTTPConnection(url)

    def setToLanguage(self, lang):
        if lang in self.allowedLang:
            self.toLang = lang;
            return True

        return False

    def setFromLanguage(self, lang):
        if lang in self.allowedLang:
            self.fromLang = lang;
            return True

        return False
    
    def translate(self, word):

        if (len(self.toLang) == 0 or len(self.fromLang) == 0): 
            return 'Both to and from language must be set.'

        if not ((self.toLang == self.allowedLang[0] or self.fromLang == self.allowedLang[0]) and not (self.toLang == self.allowedLang[0] and self.fromLang == self.allowedLang[0])):
            return 'One of the languages must be %s.' %self.allowedLang[0]
        
        if self.toLang == self.allowedLang[0]:
            word = ':' + word
            lang = self.fromLang
        else:
            lang = self.toLang
	
	word = word.replace(' ', '%20')
        
	self.connection.request('GET', '/cgi-bin/sve-%s?%s' %(lang, word))    

        r = self.connection.getresponse()
        if not (r.status == 200):
            return 'Error retrieving data.'    
        
        return self.__parse(r.read())

    def __parse(self, data):
        
        data = data.splitlines()

        reg = re.compile('^(<L.*|<D.*)')

        match = ''

        for line in data:
            if (reg.match(line) and not '<A HREF' in line):
                match += line + '\n'

        if len(match) == 0:
            return 'No hits.'
            
        match = match.replace('---', ' - ')
        match = match.replace('<DT><DD>', '   ')
        match = match.replace('<DT>', '\n')
        match = match.replace('<I>', '[')
        match = match.replace('</I>', ']')
        match = match.replace('<DD>', ': ')
        match = match.replace('|', '/')
        match = match.replace('<LI>', '   ')
        match = match.replace('<B>', '')
        match = match.replace('</B>', '')
        match = match.replace('<P>', '')
        match = match.replace('&lt;', '<')
	match = match.replace('&gt;', '>')
	match = match.replace('<DL>', '------------------------------------------------------------')
       	
	
        return match

    def close(self):
        self.connection.close()

def langHelp():
    print '\nEither from or to language MUST be \'sve\'\n'
    print 'sve - Swedish         eng - English          bos - Bosnian'
    print 'fin - Finnish         gre - Greek            kro - Croatian'
    print 'rys - Russian         spa - Spanish          tur - Turkish'
    

if __name__ == '__main__':
    
    if len(sys.argv) < 4:
        print 'Usage: lexin <from language> <to language> [word ...]'
        langHelp()
        sys.exit(1)
    
    lexin = Lexin('lexikon.nada.kth.se')
    
    if not (lexin.setFromLanguage(sys.argv[1])):
        print 'Invalid language: %s' %sys.argv[1]
        langHelp()
        sys.exit(1)
        
    if not (lexin.setToLanguage(sys.argv[2])):
        print 'Invalid language: %s' %sys.argv[2]
        langHelp()
        sys.exit(1)
    
    for i in range(3, len(sys.argv)):
        print lexin.translate(sys.argv[i])
    
    lexin.close()