python-oxweb/oxweb/criterion.py

# -*- coding: UTF-8 -*-
# vi:si:et:sw=4:sts=4:ts=4
import re

from oxlib.cache import getUrlUnicode
from oxlib.html import stripTags
from oxlib.net import getUrl
from oxlib.text import findRe, removeSpecialCharacters

import imdb


def getIds():
    ids = []
    html = getUrlUnicode("http://www.criterion.com/library/dvd")
    results = re.compile("page=(.*?)\"").findall(html)
    pages = int(results[len(results) - 2])
    for page in range(1, pages + 1):
        html = getUrlUnicode("http://www.criterion.com/library/dvd?page=" + str(page))
        results = re.compile("films/(.*?)\"").findall(html)
        for result in results:
            ids.append(result)
        results = re.compile("boxsets/(.*?)\"").findall(html)
        for result in results:
            html = getUrlUnicode("http://www.criterion.com/boxsets/" + result)
            results = re.compile("films/(.*?)\"").findall(html)
            for result in results:
                ids.append(result)
    return map(lambda id: str(id), sorted(map(lambda id: int(id), set(ids))))

def getData(id):
    '''
    >>> getData('1333')['imdbId']
    '0060304'

    >>> getData('236')['posterUrl']
    'http://criterion_production.s3.amazonaws.com/release_images/1586/ThirdManReplace.jpg'

    >>> getData('786')['posterUrl']
    'http://criterion_production.s3.amazonaws.com/product_images/185/343_box_348x490.jpg'
    '''
    data = {}
    data['id'] = id
    try:
        html = getUrlUnicode("http://www.criterion.com/films/" + id)
    except:
        html = getUrl("http://www.criterion.com/films/" + id)
    data["number"] = findRe(html, "<p class=\"spinenumber\">(.*?)</p>")
    data["title"] = findRe(html, "<h2 class=\"movietitle\">(.*?)</h2>")
    data["director"] = findRe(html, "<h2 class=\"director\">(.*?)</h2>")
    results = re.compile("<p><strong>(.*?)</strong></p>").findall(html)
    data["country"] = results[0]
    data["year"] = results[1]
    result = findRe(html, "<div class=\"synopsis contentbox lightgray\">(.*?)</div>")
    data["synopsis"] = findRe(result, "<p>(.*?)</p>")
    result = findRe(html, "<div class=\"editioninfo\">(.*?)</div>")
    if 'Blu-Ray' in result or 'Essential Art House DVD' in result:
        result = re.compile("<div class=\"editioninfo\">(.*?)</div>", re.DOTALL).findall(html)[1]
    result = findRe(result, "<a href=\"(.*?)\">")
    if not "/boxsets/" in result:
        data["posterUrl"] = result
    else:
        html_ = getUrlUnicode(result)
        result = findRe(html_, "<a href=\"http://www.criterion.com/films/%s\">(.*?)</a>" % id)
        result = findRe(result, "src=\"(.*?)\"")
        data["posterUrl"] = result.replace("_w100", "")
    result = findRe(html, "<img alt=\"Film Still\" height=\"252\" src=\"(.*?)\"")
    if result:
        data["stillUrl"] = result
        data["trailerUrl"] = ""
    else:
        data["stillUrl"] = findRe(html, "\"thumbnailURL\", \"(.*?)\"")
        data["trailerUrl"] = findRe(html, "\"videoURL\", \"(.*?)\"")
    data['imdbId'] = imdb.getMovieId(data['title'], data['director'], data['year'])
    return data

def getPosterUrl(id):
    data = getData(id)
    return data['posterUrl']

def getMovieId(title = '', director = '', year = '', imdbId = ''):
    if not imdbId:
        imdbId = imdb.getMovieId(title, director, year)
    ids = getIds()
    for id in ids:
        data = getData(id)
        if imdb.getMovieId(data['title'], data['director'], data['year'] == imdbId):
            return id
    return ''

def getMovieData(title = '', director = '', year = '', imdbId = ''):
    '''
    >>> getMovieData('Pierrot le fou', 'Jean-Luc Godard', '1965')['id']
    '149'
    '''
    data = {}
    if not imdbId:
        imdbId = imdb.getMovieId(title, director, year)
    id = getMovieId(imdbId = imdbId)
    if id:
        data_ = getData(id)
        data['id'] = data_['id']
        data['posterUrl'] = data_['posterUrl']
        data['synopsis'] = data_['synopsis']
    return data