python-oxweb/ox/criterion.py

import re

import ox.imdb as imdb
from oxutils.cache import getUrlUnicode
from oxutils.html import stripTags
from oxutils.text import findRe

def getPosterUrl(title, director):
    # imdb module is currently broken
    imdbId = imdb.getMovieId(title, director)
    html = getUrlUnicode('http://criterion.com/asp/list.asp?sort=spine', timeout = -1)
    strings = findRe(html, '<table cellspacing="0" id="browse-all-table">(.*?)</table>').split('<tr>')
    strings.pop(0)
    for string in strings:
        criterionId = findRe(string, '"release.asp\?id=(.*?)"')
        criterionTitle = findRe(string, 'class="title">(.*?)</a>')
        criterionTitle = re.sub('(?<=\\w)<br>(?=\\w)', ' / ', criterionTitle)
        criterionTitle = criterionTitle.replace('<br>', '')
        criterionDirector = stripTags(findRe(string, '</a>.*?</td>(.*?)</td>')).strip()
        # print ('%s: %s (%s)' % (criterionId, criterionTitle, criterionDirector)).encode('utf-8')
        if imdb.getMovieId(criterionTitle, criterionDirector) == imdbId:
            return 'http://criterion.com/content/images/full_boxshot/%s_box_348x490.jpg' % criterionId
    return ''

if __name__ == '__main__':
    getPosterUrl('Le mepris', 'Jean-Luc Godard')