python-oxweb/ox/impawards.py

# vi:si:et:sw=4:sts=4:ts=4
# encoding: utf-8
import re

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


def getMovieData(title = '', director = '', imdbId = ''):
    data = {'posterUrls': []}
    if not imdbId:
        imdbId = imdb.getMovieId(title, director)
    print imdbId
    html = getUrlUnicode('http://impawards.com/archives/latest.html', timeout = 0)
    pages = int(findRe(html, '<a href = page(.*?).html>'))
    for page in range(pages + 1, 0, -1):
        print page
        if page <= pages:
            html = getUrlUnicode('http://impawards.com/archives/page%s.html' % page, timeout = -1)
        urls = parseArchivePage(html)
        print urls
        for url in urls:
            html = getUrlUnicode(url)
            d = parseMoviePage(html)
            print d
            if d['imdbId'] == imdbId:
                data['posterUrls'].append(d['posterUrl'])
                print d['posterUrl']
    data['posterUrls'].sort()
    return data
            
def parseArchivePage(html):
    urls = []
    results = re.compile('<a href = \.\./(.*?)>', re.DOTALL).findall(html)
    for result in results:
        urls.append('http://impawards.com/%s' % result)
    return urls

def parseMoviePage(html):
    data = {}
    data['imdbId'] = findRe(html, 'imdb.com/title/tt(.*?) ')
    data['title'] = stripTags(findRe(html, '<table WIDTH="400" BGCOLOR="#222222">(.*?) \(<a href="eligible.html">'))
    data['year'] = findRe(html, '\(<a href="eligible.html">(.*?)</a>\)')
    result = findRe(html, '<a href = (\w*?_xlg.html) target= _blank>')
    if result:
        url = 'http://impawards.com/%s/%s' % (data['year'], result)
        html = getUrlUnicode(url, timeout = -1)
        d = parsePosterPage(html, data['year'])
        data['posterUrl'] = d['posterUrl']
    else:
        data['posterUrl'] = 'http://impawards.com/%s/%s' % (data['year'], findRe(html, '<td align=center><br><img SRC="(.*?)"'))
    return data

def parsePosterPage(html, year):
    data = {}
    data['posterUrl'] = 'http://impawards.com/%s/%s' % (year, findRe(html, '<img SRC="(.*?)"'))
    return data

def archivePosters():
    import os
    from oxutils.net import getUrl
    pathname = '/Volumes/Rolux Home/Desktop/Data/impawards.com'
    html = getUrlUnicode('http://impawards.com/archives/latest.html', timeout = 0)
    pages = int(findRe(html, '<a href = page(.*?).html>'))
    for page in range(pages + 1, 0, -1):
        if page <= pages:
            html = getUrlUnicode('http://impawards.com/archives/page%s.html' % page, timeout = -1)
        urls = parseArchivePage(html)
        print urls
        for url in urls:
            html = getUrlUnicode(url)
            data = parseMoviePage(html)
            dirname = '%s/%s/%s' % (pathname, data['imdbId'][:4], data['imdbId'])
            filename = '%s/%s' % (dirname, os.path.split(data['posterUrl'])[1])
            if not os.path.exists(filename):
                jpg = getUrl(data['posterUrl'])
                if not os.path.exists(dirname):
                    os.makedirs(dirname)
                f = open(filename, 'w')
                f.write(jpg)
                f.close()
    

if __name__ == '__main__':
    archivePosters()
    getMovieData('Brick', 'Rian Johnson')