#!/usr/bin/python
# -*- coding: iso-8859-1 -*-

# Script to get book info from HelMet Library database and bookplus.fi
# Net bookshop for Tellico. GPL-2
# COpyright (c) 2006 by Petri Damsten

import sys, os
import re
import urllib2
import time
import base64
import Image

LANGUAGES = ['suomi', 'englanti']

if(len(sys.argv) < 2):
    sys.exit(1)

def clean(s):
    return re.sub('<.*?>', '', s).strip()

def getPage(url, isbn):
    page = urllib2.urlopen(url % isbn)
    #page = open("./kirjasto2.html")
    #page = open("./bookplus.html")
    content = page.read()
    page.close()
    return content

def tryre(regex, s):
    try:
        return re.findall(regex, s)[0].strip()
    except:
        return ''

def common(array1, array2):
    for s in array1:
        if s in array2:
            return s
    return ''

def helmetDataMulti(content, key):
    try:
        s = clean(re.findall(r'(?ims)>' + key +
                '<.*?bibInfoData">(.*?bibInfoLabel">)', content)[0])
        s = re.split('\n+', s)
        return s
    except:
        return []

def helmetData(content, key):
    try:
        return clean(re.findall(r'(?ims)>' + key +
                                '<.*?bibInfoData">(.*?)</td>', content)[0])
    except:
        return ''

def helmet(isbn):
    data = {}
    url = 'http://www.helmet.fi/search*fin/?searchtype=i&searcharg=%s'
    content = getPage(url, isbn)
    data['author'] = helmetData(content, 'Tekijä')
    s = helmetData(content, 'Teos')
    data['title'] = tryre(r'(.*?)[/:]', s)
    data['subtitle'] = helmetData(content, 'Alkuteos')
    s = helmetData(content, 'Julktiedot')
    data['publisher'] = tryre(r'.*:(.*?),', s)
    data['pub_year'] = tryre(r',\s*(\d+)', s)
    data['language'] = common(helmetDataMulti(content, 'Huomautus'),
                              LANGUAGES)
    s = helmetData(content, 'Ulkoasu')
    data['pages'] = tryre(r'(\d+)', s)
    s = helmetData(content, 'ISBN')
    data['isbn'] = tryre(r'(.*?)\(', s)
    s = tryre(r'\((.*?)\)', s)
    if s == 'sid.':
        data['binding'] = 'Hardback'
    elif s == 'nid.':
        data['binding'] = 'Paperback'
    else:
        data['binding'] = '?'
    data['keywords'] = helmetDataMulti(content, 'Asiasana')
    data['comments'] = ''
    data['image'] = ''
    data['image_file'] = ''
    return data

def bookplus(isbn):
    data = {}
    url = 'http://www.bookplus.fi/product.php?isbn=%s'
    content = getPage(url, isbn)

    card = tryre(r'(?ims)<!-- tuotekortin tiedot alkaa -->' +
                 '(.*?)<!-- tuotekortin tiedot päättyy -->', content)
    data['title'] = tryre(r'(?ims)greentitle13">(.*?)</font>', card)
    data['subtitle'] = ''
    data['keywords'] = ''
    data['author'] = tryre(r'(?ims)Author.*?>(.*?)</a>', card)
    data['publisher'] = tryre(r'(?ims)Publisher.*?>(.*?)</a>', card)
    data['pub_year'] = tryre(r'(?ims)vuosi: </b></font>(.*?)<br>', card)
    data['language'] = tryre(r'(?ims)Kieli: </b></font>(.*?)<br>', card)
    data['pages'] = tryre(r'(?ims)Sivuja: </b></font>(.*?)<br>', card)
    data['isbn'] = tryre(r'(?ims)ISBN: </b></font>(.*?)<br>', card)
    binding = tryre(r'(?ims)Tuotemuoto: </b></font>(.*?)<br>', card)
    if binding[:6] == 'Pehmeä':
        data['binding'] = 'Paperback'
    else:
        data['binding'] = 'Hardback'

    image = tryre(r'(?ims)<img.*?"(.*?)"', card)
    if(image != ''):
        data['image_file'] = image[image.rfind('/') + 1:]
        obj = urllib2.urlopen("http://www.bookplus.fi/" + image)
        img = obj.read()
        obj.close()
        f = open('/tmp/' + data['image_file'], 'w')
        f.write(img)
        f.close
        data['image'] = base64.encodestring(img)
    else:
        data['image_file'] = ''
        data['image'] = ''

    data['comments'] = tryre(r'(?ims)<!-- tuotekortin annotaatio -->.*?' +
            '</div>(.*?)</td>.*?<!-- tuotteet lisätiedot päättyy -->',
            content)
    return data

def printData(data):
    print 'Author    : ' + data['author']
    print 'Title     : ' + data['title']
    print 'Subtitle  : ' + data['subtitle']
    print 'Publisher : ' + data['publisher']
    print 'Pub Year  : ' + data['pub_year']
    print 'Language  : ' + data['language']
    print 'Pages     : ' + data['pages']
    print 'ISBN      : ' + data['isbn']
    print 'Keywords  : ' + data['keywords']
    print 'Binding   : ' + data['binding']
    print '-------------------------------------------'
    print data['comments']
    print '-------------------------------------------'
    print data['image']

def printXML(data):
    try:
        i = Image.open('/tmp/' + data['image_file'])
    except:
        i = None

    s = ''
    s += '<?xml version="1.0" encoding="UTF-8"?>\n'
    s += '<!DOCTYPE tellico ' + \
          'PUBLIC "-//Robby Stephenson/DTD Tellico V9.0//EN" ' + \
          '"http://periapsis.org/tellico/dtd/v9/tellico.dtd">\n'
    s += '<tellico xmlns="http://periapsis.org/tellico/" syntaxVersion="9" >\n'
    s += ' <collection title="My Books" type="2" >\n'
    s += '  <entry id="0" >\n'
    s += '   <title>%s</title>\n' % data['title'].decode('iso-8859-1')
    s += '   <subtitle>%s</subtitle>\n' % data['subtitle'].decode('iso-8859-1')
    s += '   <authors>\n'
    s += '    <author>%s</author>\n' % data['author'].decode('iso-8859-1')
    s += '   </authors>\n'
    s += '   <binding>%s</binding>\n' % data['binding'].decode('iso-8859-1')
    s += '   <publisher>%s</publisher>\n' % data['publisher'].decode('iso-8859-1')
    s += '   <pub_year>%s</pub_year>\n' % data['pub_year']
    s += '   <isbn>%s</isbn>\n' % data['isbn']
    s += '   <pages>%s</pages>\n' % data['pages']
    s += '   <languages>\n    <language>%s</language>\n</languages>\n' % \
            data['language'].decode('iso-8859-1')
    if len(data['keywords']) > 0:
        s += '   <keywords>\n'
        for word in data['keywords']:
            s += '    <keyword>%s</keyword>\n' % word.decode('iso-8859-1')
        s += '   </keywords>\n'
    if i != None:
        s += '   <cover>%s</cover>\n' % data['image_file']
    if data['comments'] != '':
        comment = data['comments'].replace('<br />', '\n')
        comment = clean(comment)
        s += '   <comments>%s</comments>\n' % comment.decode('iso-8859-1')
    s += '  </entry>\n'

    if i != None:
        s += '  <images>\n'
        s += '   <image width="%d" format="JPEG" height="%d" id="%s">\n' % \
                (i.size[0], i.size[1], data['image_file'])
        s += data['image'] + '\n'
        s += '   </image>\n'
        s += '  </images>\n'
    s += ' </collection>\n'
    s += '</tellico>\n'
    print s.encode('utf-8')

def combine(data1, data2):
    data = {}
    for key in data1.keys():
        if(data1[key] != ''):
            data[key] = data1[key]
        else:
            data[key] = data2[key]
    return data

def cleanIsbn(isbn):
    return re.sub('[^0-9]', '', isbn)

isbn = cleanIsbn(sys.argv[1])
data1 = helmet(isbn)
data2 = bookplus(isbn)
data = combine(data1, data2)
printXML(data)