def get_data(piratebayId):
_key_map = {
'spoken language(s)': u'language',
'texted language(s)': u'subtitle language',
'by': u'uploader',
'leechers': 'leecher',
'seeders': 'seeder',
}
piratebayId = get_id(piratebayId)
torrent = dict()
torrent[u'id'] = piratebayId
torrent[u'domain'] = 'thepiratebay.org'
torrent[u'comment_link'] = 'http://thepiratebay.org/torrent/%s' % piratebayId
data = read_url(torrent['comment_link'], unicode=True)
torrent[u'title'] = find_re(data, '<title>(.*?) \(download torrent\) - TPB</title>')
if not torrent[u'title']:
return None
torrent[u'title'] = decode_html(torrent[u'title']).strip()
torrent[u'imdbId'] = find_re(data, 'title/tt(\d{7})')
title = quote(torrent['title'].encode('utf-8'))
torrent[u'torrent_link']="http://torrents.thepiratebay.org/%s/%s.torrent" % (piratebayId, title)
for d in re.compile('dt>(.*?):</dt>.*?<dd.*?>(.*?)</dd>', re.DOTALL).findall(data):
key = d[0].lower().strip()
key = _key_map.get(key, key)
value = decode_html(strip_tags(d[1].strip()))
torrent[key] = value
torrent[u'description'] = find_re(data, '<div class="nfo">(.*?)</div>')
if torrent[u'description']:
torrent['description'] = normalize_newlines(decode_html(strip_tags(torrent['description']))).strip()
t = read_url(torrent[u'torrent_link'])
torrent[u'torrent_info'] = get_torrent_info(t)
return torrent
def get_data(mininovaId):
_key_map = {
'by': u'uploader',
}
mininovaId = get_id(mininovaId)
torrent = dict()
torrent[u'id'] = mininovaId
torrent[u'domain'] = 'mininova.org'
torrent[u'comment_link'] = "http://www.mininova.org/tor/%s" % mininovaId
torrent[u'torrent_link'] = "http://www.mininova.org/get/%s" % mininovaId
torrent[u'details_link'] = "http://www.mininova.org/det/%s" % mininovaId
data = read_url(torrent['comment_link'], unicode=True) + read_url(torrent['details_link'], unicode=True)
if '<h1>Torrent not found...</h1>' in data:
return None
for d in re.compile('<p>.<strong>(.*?):</strong>(.*?)</p>', re.DOTALL).findall(data):
key = d[0].lower().strip()
key = _key_map.get(key, key)
value = decode_html(strip_tags(d[1].strip()))
torrent[key] = value
torrent[u'title'] = find_re(data, '<title>(.*?):.*?</title>')
torrent[u'imdbId'] = find_re(data, 'title/tt(\d{7})')
torrent[u'description'] = find_re(data, '<div id="description">(.*?)</div>')
if torrent['description']:
torrent['description'] = normalize_newlines(decode_html(strip_tags(torrent['description']))).strip()
t = read_url(torrent[u'torrent_link'])
torrent[u'torrent_info'] = get_torrent_info(t)
return torrent
def get_data(isbn):
r = {}
url = '%s/Search/Book/%s/1' % (base, isbn)
data = read_url(url).decode('utf-8')
m = re.compile('href="(/Lookup/Book/[^"]+?)"').findall(data)
if m:
ids = m[0].split('/')
r['isbn'] = ids[-2]
r['asin'] = ids[-3]
url = '%s%s' % (base, m[0])
data = read_url(url).decode('utf-8')
r["title"] = find_re(data, "<h2>(.*?)</h2>")
keys = {
'author': 'Author(s)',
'publisher': 'Publisher',
'date': 'Publication date',
'edition': 'Edition',
'binding': 'Binding',
'volume': 'Volume(s)',
'pages': 'Pages',
}
for key in keys:
r[key] = find_re(data, '<span class="title">%s:</span>(.*?)</li>'% re.escape(keys[key]))
if r[key] == '--':
r[key] = ''
if key == 'pages' and r[key]:
r[key] = int(r[key])
desc = find_re(data, '<h2>Description:<\/h2>(.*?)<div ')
desc = desc.replace('<br /><br />', ' ').replace('<br /> ', ' ').replace('<br />', ' ')
r['description'] = strip_tags(desc).strip()
if r['description'] == u'Description of this item is not available at this time.':
r['description'] = ''
r['cover'] = find_re(data, '<img src="(.*?)" alt="Book cover').replace('._SL160_', '')
return r
def get_show_data(url):
data = read_url(url, unicode=True)
r = {}
r['title'] = strip_tags(find_re(data, '<h1>(.*?)</h1>'))
r['imdb'] = find_re(data, '<h1><a href=".*?/title/tt(\d.*?)">.*?</a></h1>')
r['episodes'] = {}
#1. 1- 1 1001 7 Aug 05 You Can't Miss the Bear
for episode in re.compile('(\d+?)\..*?(\d+?-.\d.*?) .*?(\d+?) .*?(.*?) <a target="_blank" href="(.*?)">(.*?)</a>').findall(data):
air_date = episode[3].strip()
#'22 Sep 04' -> 2004-09-22
try:
air_date = time.strftime('%Y-%m-%d', time.strptime(air_date, '%d %b %y'))
except:
pass
s = episode[1].split('-')[0].strip()
e = episode[1].split('-')[-1].strip()
try:
r['episodes']['S%02dE%02d' % (int(s), int(e))] = {
'prod code': episode[2],
'air date': air_date,
'url': episode[4],
'title':episode[5],
}
except:
print "oxweb.epguides failed,", url
return r
def get_id(piratebayId):
if piratebayId.startswith('http://torrents.thepiratebay.org/'):
piratebayId = piratebayId.split('org/')[1]
d = find_re(piratebayId, "tor/(\d+)")
if d:
piratebayId = d
d = find_re(piratebayId, "torrent/(\d+)")
if d:
piratebayId = d
return piratebayId
def get_data(id):
url = "http://www.amazon.com/title/dp/%s/" % id
data = read_url(url, unicode=True)
def find_data(key):
return find_re(data, '<li><b>%s:</b>(.*?)</li>'% key).strip()
r = {}
r['amazon'] = url
r['title'] = find_re(data, '<span id="productTitle" class="a-size-large">(.*?)</span>')
r['authors'] = []
doc = lxml.html.document_fromstring(data)
for e in doc.xpath("//span[contains(@class, 'author')]"):
print e
for secondary in e.xpath(".//span[contains(@class, 'a-color-secondary')]"):
if 'Author' in secondary.text:
author = e.xpath(".//span[contains(@class, 'a-size-medium')]")
if author:
r['authors'].append(author[0].text.strip())
else:
r['authors'].append(e.xpath('.//a')[0].text.strip())
break
elif 'Translator' in secondary.text:
r['translator'] = [e.xpath('.//a')[0].text]
break
r['publisher'] = find_data('Publisher')
r['language'] = find_data('Language')
r['isbn-10'] = find_data('ISBN-10')
r['isbn-13'] = find_data('ISBN-13').replace('-', '')
r['dimensions'] = find_re(data, '<li><b>.*?Product Dimensions:.*?</b>(.*?)</li>')
r['pages'] = find_data('Paperback')
if not r['pages']:
r['pages'] = find_data('Hardcover')
r['review'] = strip_tags(find_re(data, '<h3 class="productDescriptionSource">Review</h3>.*?<div class="productDescriptionWrapper">(.*?)</div>').replace('<br />', '\n')).strip()
for e in doc.xpath('//noscript'):
for c in e.getchildren():
if c.tag == 'div':
r['description'] = strip_tags(decode_html(lxml.html.tostring(c))).strip()
break
r['cover'] = re.findall('src="(.*?)" id="prodImage"', data)
if r['cover']:
r['cover'] = r['cover'][0].split('._BO2')[0]
if not r['cover'].endswith('.jpg'):
r['cover'] = r['cover'] + '.jpg'
if 'no-image-avail-img' in r['cover']:
del r['cover']
else:
del r['cover']
return r
def __init__(self, id, timeout=-1):
url = "http://ids.freebaseapps.com/get_ids?id=/authority/imdb/title/tt%s" % id
'''
"http://graph.freebase.com/imdb.title.tt%s" % id
might also be of interest at some point, right now not much info
'''
data = read_url(url, unicode=True)
try:
data = json.loads(data)
except ValueError:
return
'''
for key in data:
self[key] = data[key]
'''
for key in ('id', 'guid', 'name'):
self[key] = data[key]
keys = {
'wikipedia': '/wikipedia/en',
'netflix': '/authority/netflix/movie',
'nytimes': '/source/nytimes/movie',
'metacritic': '/source/metacritic/movie',
}
for key in keys:
links = filter(lambda x: x['namespace'] == keys[key],data['ids'])
if links:
self[key] = links[0]['uri']
if 'nytimes' in self:
self['nytimes'] = self['nytimes'].replace('_/overview', '%s/overview' % self['name'].replace(' ', '-'))
self['amgId'] = find_re(self['nytimes'], 'movie/(\d+)/')
def get_book(id):
if isinstance(id, basestring) and id.startswith('http'):
url = id
else:
url = get_url(id)
html = ox.cache.read_url(url, unicode=True)
data = {}
data['url'] = url
pages = []
page = get_page(url)
pages.append(page)
data['base'], data['images'] = get_images(page, html, True)
info = ox.find_re(html, '<table>.*?</table>')
for i in re.compile('<tr.*?>(.*?)</tr>').findall(info):
key, value = i.split('</td><td>')
data[ox.strip_tags(key)] = ox.strip_tags(value)
links = re.compile('<a style="float: right;" href="(/buch/.*?)">').findall(html)
while links:
for l in links:
l = 'http://gutenberg.spiegel.de' + l
html = ox.cache.read_url(l)
links = re.compile('<a style="float: right;" href="(/buch/.*?)">').findall(html)
page = get_page(l)
pages.append(page)
data['images'] += get_images(page, html)
data['pages'] = pages
return data
def get_url(id=None, imdb=None):
if imdb:
url = "http://www.imdb.com/title/tt%s/criticreviews" % imdb
data = read_url(url)
metacritic_url = find_re(data, '"(http://www.metacritic.com/movie/.*?)"')
return metacritic_url or None
return 'http://www.metacritic.com/movie/%s' % id
def parse_table(html):
return [
[
strip_tags(r).strip().replace(' ', '')
for r in x.split('<td width="305">-')
]
for x in find_re(html, '<div id="results-table">(.*?)</table>').split('</tr>')[:-1]
]
def lookup(id):
logger.debug('lookup %s', id)
r = {'asin': [id]}
url = '%s/Lookup/Book/%s/%s/1' % (base, id, id)
logger.debug('%s', url)
data = read_url(url).decode('utf-8')
r["title"] = find_re(data, "<h2>(.*?)</h2>")
if r["title"] == 'Error!':
return {}
keys = {
'author': 'Author(s)',
'publisher': 'Publisher',
'date': 'Publication date',
'edition': 'Edition',
'binding': 'Binding',
'volume': 'Volume(s)',
'pages': 'Pages',
}
for key in keys:
r[key] = find_re(
data,
'<span class="title">%s:</span>(.*?)</li>' % re.escape(keys[key]))
if r[key] == '--' or not r[key]:
del r[key]
if key == 'pages' and key in r:
r[key] = int(r[key])
desc = find_re(data, '<h2>Description:<\/h2>(.*?)<div ')
desc = desc.replace('<br /><br />',
' ').replace('<br /> ', ' ').replace('<br />', ' ')
r['description'] = decode_html(strip_tags(desc))
r['cover'] = find_re(data, '<img src="(.*?)" alt="Book cover').replace(
'._SL160_', '')
for key in r:
if isinstance(r[key], str):
r[key] = decode_html(strip_tags(r[key])).strip()
if 'author' in r and isinstance(r['author'], str) and r['author']:
r['author'] = [r['author']]
else:
r['author'] = []
if not r['author'] or r['author'][0].isupper():
del r['author']
if r['description'].lower(
) == 'Description of this item is not available at this time.'.lower():
r['description'] = ''
return r
def findISBN(title, author):
q = '%s %s' % (title, author)
url = "http://www.amazon.com/s/ref=nb_sb_noss?url=search-alias%3Dstripbooks&field-keywords=" + "%s&x=0&y=0" % quote(q)
data = read_url(url, unicode=True)
links = re.compile('href="(http://www.amazon.com/.*?/dp/.*?)"').findall(data)
id = find_re(re.compile('href="(http://www.amazon.com/.*?/dp/.*?)"').findall(data)[0], '/dp/(.*?)/')
data = get_data(id)
if author in data['authors']:
return data
return {}
def get_id(mininovaId):
mininovaId = unicode(mininovaId)
d = find_re(mininovaId, "/(\d+)")
if d:
return d
mininovaId = mininovaId.split('/')
if len(mininovaId) == 1:
return mininovaId[0]
else:
return mininovaId[-1]
def lookup(id):
logger.debug('lookup %s', id)
r = {
'asin': [id]
}
url = '%s/Lookup/Book/%s/%s/1' % (base, id, id)
logger.debug('%s', url)
data = read_url(url).decode('utf-8')
r["title"] = find_re(data, "<h2>(.*?)</h2>")
if r["title"] == 'Error!':
return {}
keys = {
'author': 'Author(s)',
'publisher': 'Publisher',
'date': 'Publication date',
'edition': 'Edition',
'binding': 'Binding',
'volume': 'Volume(s)',
'pages': 'Pages',
}
for key in keys:
r[key] = find_re(data, '<span class="title">%s:</span>(.*?)</li>'% re.escape(keys[key]))
if r[key] == '--' or not r[key]:
del r[key]
if key == 'pages' and key in r:
r[key] = int(r[key])
desc = find_re(data, '<h2>Description:<\/h2>(.*?)<div ')
desc = desc.replace('<br /><br />', ' ').replace('<br /> ', ' ').replace('<br />', ' ')
r['description'] = decode_html(strip_tags(desc))
r['cover'] = find_re(data, '<img src="(.*?)" alt="Book cover').replace('._SL160_', '')
for key in r:
if isinstance(r[key], str):
r[key] = decode_html(strip_tags(r[key])).strip()
if 'author' in r and isinstance(r['author'], str) and r['author']:
r['author'] = [r['author']]
else:
r['author'] = []
if not r['author'] or r['author'][0].isupper():
del r['author']
if r['description'].lower() == 'Description of this item is not available at this time.'.lower():
r['description'] = ''
return r
def get_images(page, html, b=False):
img = []
base = ''
if '<img' in page:
base = ox.find_re(html, '<base href="(.*?)"')
for url in re.compile('<img.*?src="(.*?)"').findall(page):
url = base + url
img.append(url)
if b:
return base, img
return img
def get_data(url):
data = read_url(url)
r = {}
r['title'] = find_re(data, '<h1 class="movie_title">(.*?)</h1>')
if '(' in r['title']:
r['year'] = find_re(r['title'], '\((\d*?)\)')
r['title'] = strip_tags(re.sub('\((\d*?)\)', '', r['title'])).strip()
r['summary'] = strip_tags(find_re(data, '<p id="movieSynopsis" class="movie_synopsis" itemprop="description">(.*?)</p>')).strip()
r['summary'] = r['summary'].replace('\t', ' ').replace('\n', ' ').replace(' ', ' ').replace(' ', ' ')
if not r['summary']:
r['summary'] = get_og(data, 'description')
meter = re.compile('<span id="all-critics-meter" class="meter(.*?)">(.*?)</span>').findall(data)
meter = filter(lambda m: m[1].isdigit(), meter)
if meter:
r['tomatometer'] = meter[0][1]
r['rating'] = find_re(data, 'Average Rating: <span>([\d.]+)/10</span>')
r['user_score'] = find_re(data, '<span class="meter popcorn numeric ">(\d+)</span>')
r['user_rating'] = find_re(data, 'Average Rating: ([\d.]+)/5')
poster = get_og(data, 'image')
if poster and not 'poster_default.gif' in poster:
r['posters'] = [poster]
for key in r.keys():
if not r[key]:
del r[key]
return r
def get_episode_data(url):
'''
prases informatin on tvcom episode pages
returns dict with title, show, description, score
example:
get_episode_data('http://www.tv.com/lost/do-no-harm/episode/399310/summary.html')
'''
data = read_url(url, unicode=True)
r = {}
r['description'] = strip_tags(find_re(data, 'div id="main-col">.*?<div>(.*?)</div').split('\r')[0])
r['show'] = find_re(data, '<h1>(.*?)</h1>')
r['title'] = find_re(data, '<title>.*?: (.*?) - TV.com </title>')
#episode score
r['episode score'] = find_re(data, '<span class="f-28 f-bold mt-10 mb-10 f-FF9 db lh-18">(.*?)</span>')
match = re.compile('Episode Number: (\d*?) Season Num: (\d*?) First Aired: (.*?)  ').findall(data)
if match:
r['season'] = int(match[0][1])
r['episode'] = int(match[0][0])
#'Wednesday September 29, 2004' -> 2004-09-29
r['air date'] = time.strftime('%Y-%m-%d', time.strptime(match[0][2], '%A %B %d, %Y'))
return r
def get_posters(url, group=True, timeout=-1):
posters = []
html = read_url(url, timeout=timeout, unicode=True)
if url in html:
if group:
results = re.compile('<a href="(http://www.movieposterdb.com/group/.+?)\??">', re.DOTALL).findall(html)
for result in results:
posters += get_posters(result, False)
results = re.compile('<a href="(http://www.movieposterdb.com/poster/.+?)">', re.DOTALL).findall(html)
for result in results:
html = read_url(result, timeout=timeout, unicode=True)
posters.append(find_re(html, '"(http://www.movieposterdb.com/posters/.+?\.jpg)"'))
return posters
def get_movie_poster(imdbId):
'''
>>> get_movie_poster('0133093')
'http://ia.media-imdb.com/images/M/MV5BMjEzNjg1NTg2NV5BMl5BanBnXkFtZTYwNjY3MzQ5._V1._SX338_SY475_.jpg'
>>> get_movie_poster('0994352')
'http://ia.media-imdb.com/images/M/MV5BMjA3NzMyMzU1MV5BMl5BanBnXkFtZTcwNjc1ODUwMg@@._V1._SX594_SY755_.jpg'
'''
info = ImdbCombined(imdbId)
if 'posterId' in info:
url = "http://www.imdb.com/rg/action-box-title/primary-photo/media/rm%s/tt%s" % (info['posterId'], imdbId)
data = read_url(url)
poster = find_re(data, 'img id="primary-img".*?src="(.*?)"')
return poster
elif 'series' in info:
return get_movie_poster(info['series'])
return ''
def find_subtitles(imdb, parts = 1, language = "eng"):
if len(language) == 2:
language = langCode2To3(language)
elif len(language) != 3:
language = langTo3Code(language)
url = "http://www.opensubtitles.org/en/search/"
if language:
url += "sublanguageid-%s/" % language
url += "subsumcd-%s/subformat-srt/imdbid-%s/rss_2_00" % (parts, imdb)
data = read_url(url)
if "title>opensubtitles.com - search results</title" in data:
fd = feedparser.parse(data)
opensubtitleId = None
if fd.entries:
link = fd.entries[0]['links'][0]['href']
opensubtitleId = re.compile('subtitles/(.*?)/').findall(link)
if opensubtitleId:
opensubtitleId = opensubtitleId[0]
else:
opensubtitleId = find_re(data, '/en/subtitles/(.*?)/')
return opensubtitleId
def get_data(id, language='en'):
if language == 'de':
url = 'http://films.arsenal-berlin.de/index.php/Detail/Object/Show/object_id/%d/lang/de_DE' % id
else:
url = 'http://films.arsenal-berlin.de/index.php/Detail/Object/Show/object_id/%d' % id
html = read_url(url, unicode=True)
if 'ID does not exist' in html:
return None
if 'Willkommen in der Datenbank des Arsenal' in html:
return None
data = {}
data[u'id'] = id
data[u'url'] = url
m = re.compile('<h1>(.*?)</h1>').findall(html)
if m:
data[u'title'] = m[0]
m = re.compile("<b>Director: </b><a href='.*?'>(.*?)</a>").findall(html)
if m:
data[u'director'] = m[0]
m = re.compile("caUI.initImageScroller\(\[\{url:'(.*?)'").findall(html)
if m:
data[u'image'] = m[0]
units = re.compile("<div class='unit'>(.*?)</div>", re.DOTALL).findall(html)
for x in map(re.compile('<b>(.*?)</b>: (.*)', re.DOTALL).findall, units):
if x:
#data[x[0][0].lower()] = strip_tags(x[0][1])
key = x[0][0].lower()
data[key] = x[0][1]
if key == "forum catalogue pdf":
data[key] = find_re(data[key], '"(http:.*?)"')
else:
data[key] = strip_tags(data[key])
if "running time (minutes)" in data:
data[u'runtime'] = float(data.pop("running time (minutes)").replace(',', '.')) * 60
for key in ('year', 'length in metres', 'forum participation year', 'number of reels'):
if key in data and data[key].isdigit():
data[key] = int(data[key])
return data
def get_data(id):
'''
>>> get_data('129689')['cast'][1][1]
u'Marianne'
>>> get_data('129689')['credits'][0][0]
u'Jean-Luc Godard'
>>> get_data('129689')['posters'][0]
u'http://image.allmusic.com/00/adg/cov200/dru800/u812/u81260bbffr.jpg'
>>> get_data('129689')['rating']
u'4.5'
'''
if id.startswith('http'):
id = get_id(id)
data = {
"url": get_url(id)
}
html = read_url(data["url"], unicode=True)
data['aka'] = parse_list(html, 'AKA')
data['category'] = find_re(html, '<dt>category</dt>.*?<dd>(.*?)</dd>')
data['countries'] = parse_list(html, 'countries')
data['director'] = parse_entry(html, 'directed by')
data['genres'] = parse_list(html, 'genres')
data['keywords'] = parse_list(html, 'keywords')
data['posters'] = [find_re(html, '<img src="(http://cps-.*?)"')]
data['produced'] = parse_list(html, 'produced by')
data['rating'] = find_re(html, 'Stars" title="(.*?) Stars"')
data['released'] = parse_entry(html, 'released by')
data['releasedate'] = parse_list(html, 'release date')
data['runtime'] = parse_entry(html, 'run time').replace('min.', '').strip()
data['set'] = parse_entry(html, 'set in')
data['synopsis'] = strip_tags(find_re(html, '<div class="toggle-text" itemprop="description">(.*?)</div>')).strip()
data['themes'] = parse_list(html, 'themes')
data['types'] = parse_list(html, 'types')
data['year'] = find_re(html, '<span class="year">.*?(\d+)')
#data['stills'] = [re.sub('_derived.*?/', '', i) for i in re.compile('<a href="#" title="movie still".*?<img src="(.*?)"', re.DOTALL).findall(html)]
data['stills'] = re.compile('<a href="#" title="movie still".*?<img src="(.*?)"', re.DOTALL).findall(html)
#html = read_url("http://allmovie.com/work/%s/cast" % id, unicode=True)
#data['cast'] = parse_table(html)
#html = read_url("http://allmovie.com/work/%s/credits" % id, unicode=True)
#data['credits'] = parse_table(html)
html = read_url("http://allmovie.com/work/%s/review" % id, unicode=True)
data['review'] = strip_tags(find_re(html, '<div class="toggle-text" itemprop="description">(.*?)</div>')).strip()
return data
def parse_list(html, title):
html = find_re(html, '<dt>%s</dt>.*?<dd>(.*?)</dd>' % title.lower())
r = map(strip_tags, re.compile('<li>(.*?)</li>', re.DOTALL).findall(html))
if not r and html:
r = [strip_tags(html)]
return r
def get_movie_id(title, director='', year='', timeout=-1):
'''
>>> get_movie_id('The Matrix')
u'0133093'
>>> get_movie_id('2 or 3 Things I Know About Her', 'Jean-Luc Godard')
u'0060304'
>>> get_movie_id('2 or 3 Things I Know About Her', 'Jean-Luc Godard', '1967')
u'0060304'
>>> get_movie_id(u"Histoire(s) du cinema: Le controle de l'univers", 'Jean-Luc Godard')
u'0179214'
>>> get_movie_id(u"Histoire(s) du cinéma: Le contrôle de l'univers", 'Jean-Luc Godard')
u'0179214'
'''
imdbId = {
(u'Le jour se l\xe8ve', u'Marcel Carn\xe9'): '0031514',
(u'Wings', u'Larisa Shepitko'): '0061196',
(u'The Ascent', u'Larisa Shepitko'): '0075404',
(u'Fanny and Alexander', u'Ingmar Bergman'): '0083922',
(u'Torment', u'Alf Sj\xf6berg'): '0036914',
(u'Crisis', u'Ingmar Bergman'): '0038675',
(u'To Joy', u'Ingmar Bergman'): '0043048',
(u'Humain, trop humain', u'Louis Malle'): '0071635',
(u'Place de la R\xe9publique', u'Louis Malle'): '0071999',
(u'God\u2019s Country', u'Louis Malle'): '0091125',
(u'Flunky, Work Hard', u'Mikio Naruse'): '0022036',
(u'The Courtesans of Bombay', u'Richard Robbins') : '0163591',
(u'Je tu il elle', u'Chantal Akerman') : '0071690',
(u'Hotel Monterey', u'Chantal Akerman') : '0068725',
(u'No Blood Relation', u'Mikio Naruse') : '023261',
(u'Apart from You', u'Mikio Naruse') : '0024214',
(u'Every-Night Dreams', u'Mikio Naruse') : '0024793',
(u'Street Without End', u'Mikio Naruse') : '0025338',
(u'Sisters of the Gion', u'Kenji Mizoguchi') : '0027672',
(u'Osaka Elegy', u'Kenji Mizoguchi') : '0028021',
(u'Blaise Pascal', u'Roberto Rossellini') : '0066839',
(u'Japanese Girls at the Harbor', u'Hiroshi Shimizu') : '0160535',
(u'The Private Life of Don Juan', u'Alexander Korda') : '0025681',
(u'Last Holiday', u'Henry Cass') : '0042665',
(u'A Colt Is My Passport', u'Takashi Nomura') : '0330536',
(u'Androcles and the Lion', u'Chester Erskine') : '0044355',
(u'Major Barbara', u'Gabriel Pascal') : '0033868',
(u'Come On Children', u'Allan King') : '0269104',
(u'Jimi Plays Monterey & Shake! Otis at Monterey', u'D. A. Pennebaker and Chris Hegedus') : '',
(u'Martha Graham: Dance on Film', u'Nathan Kroll') : '',
(u'Carmen', u'Carlos Saura'): '0085297',
(u'The Story of a Cheat', u'Sacha Guitry'): '0028201',
(u'Weekend', 'Andrew Haigh'): '1714210',
}.get((title, director), None)
if imdbId:
return imdbId
params = {'s':'tt','q': title}
if director:
params['q'] = u'"%s" %s' % (title, director)
if year:
params['q'] = u'"%s (%s)" %s' % (title, year, director)
google_query = "site:imdb.com %s" % params['q']
if isinstance(params['q'], unicode):
try:
params['q'] = unicodedata.normalize('NFKC', params['q']).encode('latin-1')
except:
params['q'] = params['q'].encode('utf-8')
params = urllib.urlencode(params)
url = "http://akas.imdb.com/find?" + params
#print url
data = read_url(url, timeout=timeout, unicode=True)
#if search results in redirect, get id of current page
r = '<meta property="og:url" content="http://www.imdb.com/title/tt(\d{7})/" />'
results = re.compile(r).findall(data)
if results:
return results[0]
#otherwise get first result
r = '<td valign="top">.*?<a href="/title/tt(\d{7})/"'
results = re.compile(r).findall(data)
if results:
return results[0]
#print (title, director), ": '',"
#print google_query
#results = google.find(google_query, timeout=timeout)
results = duckduckgo.find(google_query, timeout=timeout)
if results:
for r in results[:2]:
imdbId = find_re(r[1], 'title/tt(\d{7})')
if imdbId:
return imdbId
#or nothing
return ''
def parse_text(html, title):
return strip_tags(find_re(html, '%s</td>.*?<td colspan="2"><p>(.*?)</td>' % title)).strip()
def amazon_lookup(asin):
url = 'http://www.amazon.com/dp/%s' % asin
html = read_url(url, timeout=-1).decode('utf-8', 'ignore')
return list(set(find_isbns(find_re(html, 'Formats</h3>.*?</table'))))
def get_data(id):
base = 'http://www.istockphoto.com'
url = base + '/stock-photo-%s.php' % id
id = find_re(id, '\d+')
data = ox.cache.read_url(url, timeout=-1)
info = {}
info['title'] = ox.find_re(data, '<title>(.*?) \|')
info['thumbnail'] = base + ox.find_re(data, 'src="(/file_thumbview_approve/%s.*?)"'%id)
info['views'] = ox.find_re(data, '<tr><td>Views:</td><td>(\d+)</td>')
info['collections'] = strip_tags(ox.find_re(data, '<td>Collections:</td><td>(.*?)</td>')).split(', ')
info['collections'] = filter(lambda x: x.strip(), info['collections'])
info['keywords'] = map(lambda k: k.strip(), strip_tags(ox.find_re(data, '<td>Keywords:</td>.*?<td>(.*?)\.\.\.<')).split(', '))
info['keywords'] = ox.find_re(data, '<meta name="keywords" content="(.*?), stock image').split(', ')
info['keywords'].sort()
info['uploaded'] = ox.find_re(data, '<td>Uploaded on:</td>.*?<td>([\d\-]+)')
info['downloads'] = ox.find_re(data, '<span class="fl">.*?(\d+) </span>')
info['contributor'] = ox.find_re(data, '<td class="m">Contributor:</td>.*?<a href="user_view.php\?id=.*?">.*?alt="(.*?)"')
info['description'] = strip_tags(ox.find_re(data, 'artistsDescriptionData = \["(.*?)<br'))
info['description'] = info['description'].split('CLICK TO SEE')[0].strip()
info['similar'] = re.compile('size=1\&id=(\d+)').findall(data)
return info
def amazon_lookup(asin):
url = 'http://www.amazon.com/dp/%s' % asin
html = read_url(url, timeout=-1).decode('utf-8', 'ignore')
return list(set(find_isbns(find_re(html, 'Formats</h3>.*?</table'))))
def find_data(key):
return find_re(data, '<li><b>%s:</b>(.*?)</li>'% key).strip()
def info(key, value):
if key not in ('isbn',):
raise IOError('unknwon key %s' % key)
if len(value) == 13:
value = stdnum.isbn.to_isbn10(value)
if len(value) != 10:
raise IOError('invalid isbn %s' % value)
url = 'http://www.amazon.com/dp/' + value
data = read_url(url).decode()
doc = lxml.html.document_fromstring(data)
info = {}
if '<title>404 - Document Not Found</title>' in data:
return info
if 'To discuss automated access to Amazon data please' in data:
return info
for l in doc.xpath('//link[@rel="canonical" and @href]'):
info['asin'] = [l.get('href').rpartition('/')[-1]]
break
info['title'] = strip_tags(decode_html(doc.xpath('//span[@id="productTitle"]')[0].text))
info['title'] = re.sub(' \([^\)]+? Classics\)', '', info['title'])
info['title'] = re.sub(' \([^\)]+? Collection\)', '', info['title'])
info['description'] = strip_tags(decode_html(unquote(re.compile('encodedDescription\' : "(.*?)",').findall(data)[0])))
info['description'] = fix_bad_unicode(info['description'])
content = doc.xpath('//div[@class="content"]')[0]
content_info = {}
for li in content.xpath('.//li'):
v = li.text_content()
if ': ' in v:
k, v = li.text_content().split(': ', 1)
content_info[k.strip()] = v.strip()
if 'Language' in content_info:
info['language'] = content_info['Language']
if 'Publisher' in content_info:
if ' (' in content_info['Publisher']:
info['date'] = find_re(content_info['Publisher'].split(' (')[-1], '\d{4}')
info['publisher'] = content_info['Publisher'].split(' (')[0]
if '; ' in info['publisher']:
info['publisher'], info['edition'] = info['publisher'].split('; ', 1)
if 'ISBN-13' in content_info:
if not 'isbn' in info: info['isbn'] = []
info['isbn'].append(content_info['ISBN-13'].replace('-', ''))
if 'ISBN-10' in content_info:
if not 'isbn' in info: info['isbn'] = []
info['isbn'].append(content_info['ISBN-10'])
a = doc.xpath('//span[@class="a-size-medium"]')
if a:
for span in a:
r = span.getchildren()[0].text.strip()
role = get_role(r)
if not role in info: info[role] = []
info[role].append(span.text.strip())
else:
for span in doc.xpath('//span[@class="author notFaded"]'):
author = [x.strip() for x in span.text_content().strip().split('\n') if x.strip()]
role = get_role(author[-1])
if not role in info: info[role] = []
info[role].append(author[0])
covers = re.compile('data-a-dynamic-image="({.+?})"').findall(data)[0]
covers = json.loads(decode_html(covers))
last = [0,0]
for url in covers:
if covers[url] > last:
last = covers[url]
info['cover'] = re.sub('(\._SX.+?_\.)', '.', url)
return info
def info(key, value):
if key not in ('isbn', ):
raise IOError('unknwon key %s' % key)
if len(value) == 13:
value = stdnum.isbn.to_isbn10(value)
if len(value) != 10:
raise IOError('invalid isbn %s' % value)
url = 'http://www.amazon.com/dp/' + value
data = read_url(url).decode()
doc = lxml.html.document_fromstring(data)
info = {}
if '<title>404 - Document Not Found</title>' in data:
return info
if 'To discuss automated access to Amazon data please' in data:
return info
for l in doc.xpath('//link[@rel="canonical" and @href]'):
info['asin'] = [l.get('href').rpartition('/')[-1]]
break
info['title'] = strip_tags(
decode_html(doc.xpath('//span[@id="productTitle"]')[0].text))
info['title'] = re.sub(' \([^\)]+? Classics\)', '', info['title'])
info['title'] = re.sub(' \([^\)]+? Collection\)', '', info['title'])
info['description'] = strip_tags(
decode_html(
unquote(
re.compile('encodedDescription\' : "(.*?)",').findall(data)
[0])))
info['description'] = fix_bad_unicode(info['description'])
content = doc.xpath('//div[@class="content"]')[0]
content_info = {}
for li in content.xpath('.//li'):
v = li.text_content()
if ': ' in v:
k, v = li.text_content().split(': ', 1)
content_info[k.strip()] = v.strip()
if 'Language' in content_info:
info['language'] = content_info['Language']
if 'Publisher' in content_info:
if ' (' in content_info['Publisher']:
info['date'] = find_re(content_info['Publisher'].split(' (')[-1],
'\d{4}')
info['publisher'] = content_info['Publisher'].split(' (')[0]
if '; ' in info['publisher']:
info['publisher'], info['edition'] = info['publisher'].split(
'; ', 1)
if 'ISBN-13' in content_info:
if not 'isbn' in info: info['isbn'] = []
info['isbn'].append(content_info['ISBN-13'].replace('-', ''))
if 'ISBN-10' in content_info:
if not 'isbn' in info: info['isbn'] = []
info['isbn'].append(content_info['ISBN-10'])
a = doc.xpath('//span[@class="a-size-medium"]')
if a:
for span in a:
r = span.getchildren()[0].text.strip()
role = get_role(r)
if not role in info: info[role] = []
info[role].append(span.text.strip())
else:
for span in doc.xpath('//span[@class="author notFaded"]'):
author = [
x.strip() for x in span.text_content().strip().split('\n')
if x.strip()
]
role = get_role(author[-1])
if not role in info: info[role] = []
info[role].append(author[0])
covers = re.compile('data-a-dynamic-image="({.+?})"').findall(data)[0]
covers = json.loads(decode_html(covers))
last = [0, 0]
for url in covers:
if covers[url] > last:
last = covers[url]
info['cover'] = re.sub('(\._SX.+?_\.)', '.', url)
return info
def __init__(self, id, timeout=-1):
#use akas.imdb.com to always get original title:
#http://www.imdb.com/help/show_leaf?titlelanguagedisplay
self.baseUrl = "http://akas.imdb.com/title/tt%s/" % id
super(Imdb, self).__init__(timeout)
url = self.baseUrl + 'combined'
page = self.read_url(url, timeout=-1)
if '<title>IMDb: Page not found</title>' in page \
or 'The requested URL was not found on our server.' in page:
return
if "<p>We're sorry, something went wrong.</p>" in page:
time.sleep(1)
super(Imdb, self).__init__(0)
if 'alternativeTitles' in self:
if len(self['alternativeTitles']) == 2 and \
isinstance(self['alternativeTitles'][0], basestring):
self['alternativeTitles'] = [self['alternativeTitles']]
#normalize country names
if 'country' in self:
self['country'] = [normalize_country_name(c) or c for c in self['country']]
if 'sound' in self:
self['sound'] = list(set(self['sound']))
types = {}
stop_words = [
'alternative spelling',
'alternative title',
'alternative transliteration',
'closing credits title',
'complete title',
'IMAX version',
'informal short title',
'International (Spanish title)',
'Japan (imdb display title)',
'longer version',
'new title',
'original subtitled version',
'pre-release title',
'promotional abbreviation',
'recut version',
'reissue title',
'restored version',
'script title',
'short title',
'(subtitle)',
'TV title',
'working title',
'World-wide (Spanish title)',
]
#ignore english japanese titles
#for movies that are not only from japan
if ['Japan'] != self.get('country', []):
stop_words += [
'Japan (English title)'
]
for t in self.get('alternativeTitles', []):
for type in t[0].split('/'):
type = type.strip()
stop_word = False
for key in stop_words:
if key in type:
stop_word = True
break
if not stop_word:
if not type in types:
types[type] = []
types[type].append(t[1])
titles = {}
for type in types:
for title in types[type]:
if not title in titles:
titles[title] = []
titles[title].append(type)
def select_title(type):
title = types[type][0]
count = 0
if len(types[type]) > 1:
for t in types[type]:
if len(titles[t]) > count:
count = len(titles[t])
title = t
return title
#FIXME: does work in python2.6, possible to import from __future__?
#types = {type: select_title(type) for type in types}
_types = {}
for type in types:
_types[type] = select_title(type)
types = _types
regexps = [
"^.+ \(imdb display title\) \(English title\)$",
"^USA \(imdb display title\)$",
"^International \(English title\)$",
"^International \(English title\)$",
"^UK \(imdb display title\)$",
"^International \(.+\) \(English title\)$",
"^World-wide \(English title\)$",
]
if 'Hong Kong' in self.get('country', []):
regexps += [
"Hong Kong \(English title\)"
]
english_countries = (
'USA', 'UK', 'United States', 'United Kingdom',
'Australia', 'New Zealand'
)
if not filter(lambda c: c in english_countries, self.get('country', [])):
regexps += [
"^[^(]+ \(English title\)$",
"^.+ \(.+\) \(English title\)$",
"^USA$",
"^UK$",
"^USA \(.+\)$",
"^UK \(.+\)$",
"^Australia \(.+\)$",
"World-wide \(English title\)",
"\(literal English title\)",
"^International \(.+ title\)$",
"^International \(.+\) \(.+ title\)$",
]
for regexp in regexps:
for type in types:
if re.compile(regexp).findall(type):
#print types[type], type
self['internationalTitle'] = types[type]
break
if 'internationalTitle' in self:
break
def cleanup_title(title):
if title.startswith('"') and title.endswith('"'):
title = title[1:-1]
if title.startswith("'") and title.endswith("'"):
title = title[1:-1]
title = re.sub('\(\#[.\d]+\)', '', title)
return title.strip()
for t in ('title', 'internationalTitle'):
if t in self:
self[t] = cleanup_title(self[t])
if 'internationalTitle' in self and \
self.get('title', '').lower() == self['internationalTitle'].lower():
del self['internationalTitle']
if 'alternativeTitles' in self:
alt = {}
for t in self['alternativeTitles']:
title = cleanup_title(t[1])
if title not in (self.get('title'), self.get('internationalTitle')):
if title not in alt:
alt[title] = []
for c in t[0].split('/'):
if not '(working title)' in c:
c = c.replace('International', '').replace('World-wide', '').split('(')[0].strip()
if c:
alt[title].append(c)
self['alternativeTitles'] = []
for t in sorted(alt, lambda a, b: cmp(sorted(alt[a]), sorted(alt[b]))):
if alt[t]:
countries = sorted([normalize_country_name(c) or c for c in alt[t]])
self['alternativeTitles'].append((t, countries))
if not self['alternativeTitles']:
del self['alternativeTitles']
if 'internationalTitle' in self:
self['originalTitle'] = self['title']
self['title'] = self.pop('internationalTitle')
if 'runtime' in self and self['runtime']:
if 'min' in self['runtime']: base=60
else: base=1
self['runtime'] = int(find_re(self['runtime'], '([0-9]+)')) * base
if 'runtime' in self and not self['runtime']:
del self['runtime']
if 'votes' in self: self['votes'] = self['votes'].replace(',', '')
if 'cast' in self:
if isinstance(self['cast'][0], basestring):
self['cast'] = [self['cast']]
self['actor'] = [c[0] for c in self['cast']]
def cleanup_character(c):
c = c.replace('(uncredited)', '').strip()
return c
self['cast'] = [{'actor': x[0], 'character': cleanup_character(x[1])}
for x in self['cast']]
if 'connections' in self:
cc={}
if len(self['connections']) == 3 and isinstance(self['connections'][0], basestring):
self['connections'] = [self['connections']]
for rel, data, _ in self['connections']:
#cc[unicode(rel)] = re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>').findall(data)
def get_conn(c):
r = {
'id': c[0],
'title': cleanup_title(c[1]),
}
description = c[2].split('<br />')
if len(description) == 2 and description[-1].strip() != '-':
r['description'] = description[-1].strip()
return r
cc[unicode(rel)] = map(get_conn, re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>(.*?)<\/div', re.DOTALL).findall(data))
self['connections'] = cc
for key in ('country', 'genre'):
if key in self:
self[key] = filter(lambda x: x.lower() != 'home', self[key])
#0092999
if '_director' in self:
if 'series' in self or 'isSeries' in self:
self['creator'] = self.pop('_director')
else:
del self['_director']
if 'isSeries' in self:
del self['isSeries']
self['isSeries'] = True
if 'episodeTitle' in self:
self['episodeTitle'] = re.sub('Episode \#\d+\.\d+', '', self['episodeTitle'])
if 'series' in self:
series = Imdb(self['series'], timeout=timeout)
self['seriesTitle'] = series['title']
if 'episodeTitle' in self:
self['seriesTitle'] = series['title']
if 'season' in self and 'episode' in self:
self['title'] = "%s (S%02dE%02d) %s" % (
self['seriesTitle'], self['season'], self['episode'], self['episodeTitle'])
else:
self['title'] = "%s (S01) %s" % (self['seriesTitle'], self['episodeTitle'])
self['season'] = 1
self['title'] = self['title'].strip()
if 'director' in self:
self['episodeDirector'] = self['director']
if not 'creator' in series and 'director' in series:
series['creator'] = series['director']
if len(series['creator']) > 10:
series['creator'] = series['director'][:1]
for key in ['creator', 'country']:
if key in series:
self[key] = series[key]
if 'year' in series:
self['seriesYear'] = series['year']
if not 'year' in self:
self['year'] = series['year']
if 'year' in self:
self['episodeYear'] = self['year']
if 'creator' in self:
self['seriesDirector'] = self['creator']
if 'originalTitle' in self:
del self['originalTitle']
else:
for key in ('seriesTitle', 'episodeTitle', 'season', 'episode'):
if key in self:
del self[key]
if 'creator' in self:
if 'director' in self:
self['episodeDirector'] = self['director']
self['director'] = self['creator']
#make lists unique but keep order
for key in ('director', 'language'):
if key in self:
self[key] = [x for i,x in enumerate(self[key])
if x not in self[key][i+1:]]
for key in ('actor', 'writer', 'producer', 'editor', 'composer'):
if key in self:
if isinstance(self[key][0], list):
self[key] = [i[0] for i in self[key] if i]
self[key] = sorted(list(set(self[key])),
lambda a, b: self[key].index(a) - self[key].index(b))
if 'budget' in self and 'gross' in self:
self['profit'] = self['gross'] - self['budget']
if 'releasedate' in self:
def parse_date(d):
try:
d = datetime.strptime(d, '%d %B %Y')
except:
try:
d = datetime.strptime(d, '%B %Y')
except:
return 'x'
return '%d-%02d-%02d' % (d.year, d.month, d.day)
self['releasedate'] = min([
parse_date(d) for d in self['releasedate']
])
if self['releasedate'] == 'x':
del self['releasedate']
if 'summary' in self:
if isinstance(self['summary'], list):
self['summary'] = self['summary'][0]
self['summary'] = self['summary'].split('</p')[0].strip()
def parse_entry(html, title):
html = find_re(html, '<dt>%s</dt>.*?<dd>(.*?)</dd>' % title)
return strip_tags(html).strip()
