def __init__(self):
self.cache = Cache('kinopoisk.db')
self.html = Clear()
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.kinopoisk.ru/level/7/'
}
def __init__(self):
self.api_key = '1D62F2F90030C444'
self.cache = Cache('tvdb.db')
self.http = HTTP()
self.headers = {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.thetvdb.com/'
}
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'profile\.php\?mode=sendpassword"')
self.re_captcha = re.compile(r'<img src="(\/\/[^\/]+/captcha/[^"]+)"')
self.re_captcha_sid = re.compile(r'<input type="hidden" name="cap_sid" value="([^"]+)">')
self.re_captcha_code = re.compile(r'<input type="text" name="(cap_code_[^"]+)"')
self.captcha_sid = None
self.captcha_code = None
self.captcha_code_value = None
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.lib/forum/index.php'
}
def __init__(self):
self.cache = Cache('kinopoisk.db')
self.html = Clear()
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.kinopoisk.ru/level/7/'
}
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'"profile\.php\?mode=sendpassword"')
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.org/forum/index.php'
}
def __init__(self):
self.api_key = '1D62F2F90030C444'
self.cache = Cache('tvdb.db')
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.thetvdb.com/'
}
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'profile\.php\?mode=sendpassword"')
self.re_captcha = re.compile(r'<img src="(\/\/[^\/]+/captcha/[^"]+)"')
self.re_captcha_sid = re.compile(r'<input type="hidden" name="cap_sid" value="([^"]+)">')
self.re_captcha_code = re.compile(r'<input type="text" name="(cap_code_[^"]+)"')
self.captcha_sid = None
self.captcha_code = None
self.captcha_code_value = None
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.nl/forum/index.php'
}
class KinoPoisk:
"""
API:
scraper - скрапер
movie - профайл фильма
search - поиск фильма
best - поиск лучших фильмов
person - поиск персон
work - информация о работах персоны
"""
def __init__(self):
self.cache = Cache('kinopoisk.db')
self.html = Clear()
self.http = HTTP()
self.headers = {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.kinopoisk.ru/level/7/'
}
# API
def scraper(self, name, year=None, trailer_quality=None):
try:
tag = 'scraper:' + urllib.quote_plus(name.encode('windows-1251'))
except:
return None
else:
if year:
tag += ':' + str(year)
id = self.cache.get(tag, self._scraper, name, year)
if not id:
return None
return self.movie(id, trailer_quality)
def movie(self, id, trailer_quality=None):
id = str(id)
if trailer_quality is None:
trailer_quality = 6
movie = self.cache.get('movie:' + id, self._movie, id)
if not movie:
return None
if movie['trailers']:
# компилируем список с нужным нам качеством
video = []
for m in movie['trailers']:
url = [x for x in m['video'] if x[0] <= trailer_quality]
if url:
m['video'] = url[-1]
video.append(m)
movie['trailers'] = video
if movie['trailers']:
# готовим главный трейлер
r = [x for x in movie['trailers'] if x['trailer']]
if r:
movie['info']['trailer'] = r[0]['video'][1]
else:
# если трейлер не найден, то отдаем что попало...
movie['info']['trailer'] = movie['trailers'][0]['video'][1]
return movie
def search(self, name, trailer_quality=None):
return self._search_movie(name)
def best(self, **kwarg):
page = kwarg.get('page', 1)
limit = kwarg.get('limit', 50)
url = 'http://www.kinopoisk.ru/top/navigator/m_act%5Bis_film%5D/on/m_act%5Bnum_vote%5D/' + str(
kwarg.get('votes', 100)) + '/'
if kwarg.get('dvd'):
url += 'm_act%5Bis_dvd%5D/on/'
if kwarg.get('decade'):
url += 'm_act%5Bdecade%5D/' + str(kwarg['decade']) + '/'
if kwarg.get('genre'):
url += 'm_act%5Bgenre%5D/' + str(GENRE[kwarg['genre']]) + '/'
if kwarg.get('country'):
url += 'm_act%5Bcountry%5D/' + str(kwarg['country']) + '/'
if kwarg.get('rate'):
url += 'm_act%5Brating%5D/' + str(kwarg['rate']) + ':/'
if kwarg.get('mpaa'):
url += 'm_act%5Bmpaa%5D/' + str(kwarg['mpaa']) + '/'
url += 'perpage/' + str(limit) + '/order/ex_rating/'
if page > 1:
url += 'page/' + str(page) + '/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
res = {'pages': (1, 0, 1, 0), 'data': []}
r = re.compile('<div class="pagesFromTo(.+?)<div class="pagesFromTo',
re.U | re.S).search(
response.body.decode('windows-1251'))
if r:
body = r.group(1)
# compile pagelist
p = re.compile('>([0-9]+)—[0-9]+[^0-9]+?([0-9]+)',
re.U).search(body)
if p:
page = (int(p.group(1)) - 1) / limit + 1
total = int(p.group(2))
pages = total / limit
if limit * pages != total:
pages += 1
res['pages'] = (pages, 0 if page == 1 else page - 1, page,
0 if page == pages else page + 1)
# end compile
for id in re.compile('<div id="tr_([0-9]+)"',
re.U | re.S).findall(body):
res['data'].append(int(id))
return res
def person(self, name):
#response = self.http.fetch('https://www.kinopoisk.ru/index.php?level=7&from=forma&result=adv&m_act%5Bfrom%5D=forma&m_act%5Bwhat%5D=actor&m_act%5Bfind%5D=' + urllib.quote_plus(name.encode('windows-1251')), headers=self.headers)
response = self.http.fetch(
'http://www.kinopoisk.ru/s/type/people/list/1/find/' +
urllib.quote_plus(name.encode('windows-1251')) +
'/order/relevant/',
headers=self.headers)
if response.error:
return None
res = []
body = re.compile(
'<div class="navigator">(.+?)<div class="navigator">',
re.U | re.S).search(response.body.decode('windows-1251'))
if body:
for block in re.compile('<p class="pic">(.+?)<div class="clear">',
re.U | re.S).findall(body.group(1)):
id, name, original, year, poster = None, None, None, None, None
r = re.compile(
'<p class="name"><a href="/name/([0-9]+)[^>]+>([^<]+)</a>',
re.U | re.S).search(block)
if r:
id = r.group(1)
name = r.group(2).strip()
if id and name:
r = re.compile('<span class="gray">([^<]+)</span>',
re.U | re.S).search(block)
if r:
original = r.group(1).strip()
if not original:
original = None
r = re.compile('<span class="year">([0-9]{4})</span>',
re.U | re.S).search(block)
if r:
year = int(r.group(1))
if block.find('no-poster.gif') == -1:
poster = 'http://st.kinopoisk.ru/images/actor/' + id + '.jpg'
res.append({
'id': int(id),
'name': name,
'originalname': original,
'year': year,
'poster': poster
})
return {'pages': (1, 0, 1, 0), 'data': res}
def work(self, id):
response = self.http.fetch('http://www.kinopoisk.ru/name/' + str(id) +
'/',
headers=self.headers)
if response.error:
return None
res = {}
r = re.compile('id="sort_block">(.+?)<div id="block_right"', re.U
| re.S).search(response.body.decode('windows-1251'))
if r:
for block in r.group(1).split(
u'<tr><td colspan="3" class="specializationBox')[1:]:
work = None
for w in ('actor', 'director', 'writer', 'producer',
'producer_ussr', 'composer', 'operator', 'editor',
'design', 'voice', 'voice_director'):
if block.find(u'id="' + w + u'"') != -1:
work = 'producer' if w == 'producer_ussr' else w
break
if work:
movies = []
for id, name in re.compile(
'<span class="name"><a href="/film/([0-9]+)/[^>]+>([^<]+?)</a>',
re.U).findall(block):
for tag in (u'(мини-сериал)', u'(сериал)'):
if name.find(tag) != -1:
break
else:
movies.append(int(id))
if movies:
res.setdefault(work, []).extend(movies)
return res
def review(self, id, query):
query_s = 'all' if query == 'stat' else query
data = self.cache.get('review:' + str(id) + ':' + query_s,
self._review, id, query_s)
if not data:
return data
return data[query]
def countries(self):
return COUNTRIES
def country(self, id, default=None):
country = [x[1] for x in COUNTRIES if x[0] == id]
return country[0] if country else default
# PRIVATE
def _search_movie(self, name, year=None):
url = 'http://www.kinopoisk.ru/s/type/film/list/1/find/' + urllib.quote_plus(
name.encode('windows-1251')) + '/order/relevant'
if year:
url += '/m_act%5Byear%5D/' + str(year)
url += '/m_act%5Btype%5D/film/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
res = []
r = re.compile('<div class="navigator">(.+?)<div class="navigator">',
re.U | re.S).search(
response.body.decode('windows-1251'))
if r:
for id in re.compile(
'<p class="name"><a href="/level/1/film/([0-9]+)',
re.U | re.S).findall(r.group(1)):
res.append(int(id))
return {'pages': (1, 0, 1, 0), 'data': res}
def _scraper(self, name, year):
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if year and year >= time.gmtime(time.time()).tm_year:
timeout = 7 * 24 * 60 * 60 #week
ids = self._search_movie(name, year)
if ids is None:
return False, None
elif not ids['data']:
# сохраняем пустой результат на 3-е суток
return 259200, None
else:
return timeout, ids['data'][0]
def _review(self, id, query):
url = 'http://www.kinopoisk.ru/film/' + str(id) + '/ord/rating/'
if query in ('good', 'bad', 'neutral'):
url += 'status/' + query + '/'
url += 'perpage/200/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return False, None
html = response.body.decode('windows-1251')
res = {
'stat': {
'all': 0,
'good': 0,
'bad': 0,
'neutral': 0
},
query: []
}
r = re.compile('<ul class="resp_type">(.+?)</ul>',
re.U | re.S).search(html)
if r:
ul = r.group(1)
for q, t in (('pos', 'good'), ('neg', 'bad'), ('neut', 'neutral')):
r = re.compile(
'<li class="' + q +
'"><a href="[^>]+>[^<]+</a><b>([0-9]+)</b></li>',
re.U).search(ul)
if r:
res['stat'][t] = int(r.group(1))
res['stat']['all'] = res['stat']['good'] + res['stat'][
'bad'] + res['stat']['neutral']
r = re.compile('<div class="navigator">(.+?)<div class="navigator">',
re.U | re.S).search(html)
if r:
for block in r.group(1).split('itemprop="reviews"'):
review = {
'nick': None,
'count': None,
'title': None,
'review': None,
'time': None
}
r = re.compile('itemprop="reviewBody">(.+?)</div>',
re.U | re.S).search(block)
if r:
text = r.group(1)
for tag1, tag2 in ((u'<=end=>', u'\n'), (u'<b>', u'[B]'),
(u'</b>', u'[/B]'), (u'<i>', u'[I]'),
(u'</i>', u'[/I]'), (u'<u>', u'[U]'),
(u'</u>', u'[/U]')):
text = text.replace(tag1, tag2)
r = self.html.text(text)
if r:
review['review'] = r
user = None
r = re.compile(
'<p class="profile_name"><s></s><a href="[^>]+>([^<]+)</a></p>'
).search(block)
if r:
user = self.html.string(r.group(1))
else:
r = re.compile('<p class="profile_name"><s></s>([^<]+)</p>'
).search(block)
if r:
user = self.html.string(r.group(1))
if user:
review['nick'] = user
r = re.compile('<p class="sub_title"[^>]+>([^<]+)</p>').search(
block)
if r:
title = self.html.string(r.group(1))
if title:
review['title'] = title
r = re.compile('<span class="date">([^<]+)</span>',
re.U | re.S).search(block)
if r:
review['time'] = r.group(1).replace(u' |', u',')
r = re.compile(u'<a href="[^>]+>рецензии \(([0-9]+)\)</a>',
re.U | re.S).search(block)
if r:
review['count'] = int(r.group(1))
if review['nick'] and review['review']:
res[query].append(review)
return 3600, res # one hour
def _movie(self, id):
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/',
headers=self.headers)
if response.error:
return False, None
html = response.body.decode('windows-1251')
res = {
'id': int(id),
'thumb': None,
'fanart': None,
'trailers': [],
'info': {}
}
# имя, оригинальное имя, девиз, цензура, год, top250
# runtime - длительность фильма (в отдельную переменную, иначе не видно размер файла)
for tag, reg, cb in (
('title', '<h1 class="moviename-big" itemprop="name">(.+?)</h1>',
self.html.string),
('originaltitle', 'itemprop="alternativeHeadline">([^<]*)</span>',
self.html.string),
('tagline',
'<td style="color\: #555">«(.+?)»</td></tr>',
self.html.string), ('mpaa', 'images/mpaa/([^\.]+).gif',
self.html.string),
('runtime',
'<td class="time" id="runtime">[^<]+<span style="color\: #999">/</span>([^<]+)</td>',
self.html.string),
('year', '<a href="/lists/m_act%5Byear%5D/([0-9]+)/"',
int), ('top250', '<a href="/level/20/#([0-9]+)', int)):
r = re.compile(reg, re.U).search(html)
if r:
value = r.group(1).strip()
if value:
res['info'][tag] = cb(value)
# режисеры, сценаристы, жанры
for tag, reg in (('director', u'<td itemprop="director">(.+?)</td>'), (
'writer',
u'<td class="type">сценарий</td><td[^>]*>(.+?)</td>'),
('genre', u'<span itemprop="genre">(.+?)</span>')):
r = re.compile(reg, re.U | re.S).search(html)
if r:
r2 = []
for r in re.compile('<a href="[^"]+">([^<]+)</a>',
re.U).findall(r.group(1)):
r = self.html.string(r)
if r and r != '...':
r2.append(r)
if r2:
res['info'][tag] = u', '.join(r2)
# описание фильма
r = re.compile(
'<span class="_reachbanner_"><div class="brand_words film-synopsys" itemprop="description">(.+?)</div></span>',
re.U).search(html)
if r:
plot = self.html.text(r.group(1).replace('<=end=>', '\n'))
if plot:
res['info']['plot'] = plot
# IMDB
r = re.compile('IMDb: ([0-9.]+) \(([0-9\s]+)\)</div>',
re.U).search(html)
if r:
res['info']['rating'] = float(r.group(1).strip())
res['info']['votes'] = r.group(2).strip()
# премьера
r = re.compile(u'премьера \(мир\)</td>(.+?)</tr>',
re.U | re.S).search(html)
if r:
r = re.compile(u'data\-ical\-date="([^"]+)"',
re.U | re.S).search(r.group(1))
if r:
data = r.group(1).split(' ')
if len(data) == 3:
i = 0
for mon in (u'января', u'февраля', u'марта', u'апреля',
u'мая', u'июня', u'июля', u'августа',
u'сентября', u'октября', u'ноября',
u'декабря'):
i += 1
if mon == data[1]:
mon = str(i)
if len(mon) == 1:
mon = '0' + mon
day = data[0]
if len(day) == 1:
day = '0' + day
res['info']['premiered'] = '-'.join(
[data[2], mon, day])
break
# постер
r = re.compile(u'onclick="openImgPopup\(([^\)]+)\)',
re.U | re.S).search(html)
if r:
poster = r.group(1).replace("'", '').strip()
if poster:
res['thumb'] = 'http://kinopoisk.ru' + poster
# актеры
r = re.compile(u'<h4>В главных ролях:</h4>(.+?)</ul>',
re.U | re.S).search(html)
if r:
actors = []
for r in re.compile(
'<li itemprop="actors"><a [^>]+>([^<]+)</a></li>',
re.U).findall(r.group(1)):
r = self.html.string(r)
if r and r != '...':
actors.append(r)
if actors:
res['info']['cast'] = actors[:]
menu = re.compile(
'<ul id="newMenuSub" class="clearfix(.+?)<!\-\- /menu \-\->',
re.U | re.S).search(html)
if menu:
menu = menu.group(1)
# фанарт
if menu.find('/film/' + id + '/wall/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' +
id + '/wall/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
fanart = re.compile(
'<a href="/picture/([0-9]+)/w_size/([0-9]+)/">',
re.U).findall(html)
if fanart:
fanart.sort(cmp=lambda (id1, size1),
(id2, size2): cmp(int(size1), int(size2)))
# пробуем взять максимально подходящее
fanart_best = [x for x in fanart if int(x[1]) <= 1280]
if fanart_best:
fanart = fanart_best
response = self.http.fetch(
'http://www.kinopoisk.ru/picture/' +
fanart[-1][0] + '/w_size/' + fanart[-1][1] + '/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile('id="image" src="([^"]+)"',
re.U | re.S).search(html)
if r:
res['fanart'] = r.group(1).strip()
if res['fanart'].startswith('//'):
res['fanart'] = 'http:' + res['fanart']
# если нет фанарта (обоев), то пробуем получить кадры
if not res['fanart'] and menu.find('/film/' + id +
'/stills/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' +
id + '/stills/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
fanart = re.compile(
'<a href="/picture/([0-9]+)/"><img src="[^<]+</a>[^<]+<b><i>([0-9]+)×([0-9]+)</i>',
re.U).findall(html)
if fanart:
fanart.sort(cmp=lambda (id1, size1, t1), (
id2, size2, t2): cmp(int(size1), int(size2)))
# пробуем взять максимально подходящее
fanart_best = [
x for x in fanart
if int(x[1]) <= 1280 and int(x[1]) > int(x[2])
]
if fanart_best:
fanart = fanart_best
response = self.http.fetch(
'http://www.kinopoisk.ru/picture/' +
fanart[-1][0] + '/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile('id="image" src="([^"]+)"',
re.U | re.S).search(html)
if r:
res['fanart'] = r.group(1).strip()
if res['fanart'].startswith('//'):
res['fanart'] = 'http:' + res['fanart']
# студии
if menu.find('/film/' + id + '/studio/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' +
id + '/studio/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile(u'<b>Производство:</b>(.+?)</table>',
re.U | re.S).search(html)
if r:
studio = []
for r in re.compile(
'<a href="/lists/m_act%5Bstudio%5D/[0-9]+/" class="all">(.+?)</a>',
re.U).findall(r.group(1)):
r = self.html.string(r)
if r:
studio.append(r)
if studio:
res['info']['studio'] = u', '.join(studio)
# трэйлеры
trailers1 = [] # русские трейлеры
trailers2 = [] # другие русские видео
trailers3 = [] # трейлеры
trailers4 = [] # другие видео
if menu.find('/film/' + id + '/video/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' +
id + '/video/',
headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
for row in re.compile(
u'<!-- ролик -->(.+?)<!-- /ролик -->',
re.U | re.S).findall(html):
# отсекаем лишние блоки
if row.find(u'>СМОТРЕТЬ</a>') != -1:
# русский ролик?
if row.find('class="flag flag2"') == -1:
is_ru = False
else:
is_ru = True
# получаем имя трейлера
r = re.compile(
'<a href="/film/' + id +
'/video/[0-9]+/[^>]+ class="all">(.+?)</a>',
re.U).search(row)
if r:
name = self.html.string(r.group(1))
if name:
trailer = {
'name': name,
'time': None,
'trailer': False,
'ru': is_ru,
'video': []
}
# трейлер или тизер?
for token in (u'Трейлер', u'трейлер',
u'Тизер', u'тизер'):
if name.find(token) != -1:
trailer['trailer'] = True
break
# получаем время трейлера
r = re.compile(
u'clock.gif"[^>]+></td>\s*<td style="color\: #777">[^0-9]*([0-9\:]+)</td>',
re.U | re.S).search(row)
if r:
trailer['time'] = r.group(1).strip()
# делим ролики по качеству
for r in re.compile(
'trailer/([1-3])a.gif"(.+?)link=([^"]+)" class="continue">.+?<td style="color\:#777">([^<]+)</td>\s*</tr>',
re.U | re.S).findall(row):
quality = int(r[0])
if r[1].find('icon-hd') != -1:
quality += 3
trailer['video'].append(
(quality, r[2].strip(), r[3]))
if id == '462754':
#raise
pass
if trailer['video']:
if trailer['ru']:
if trailer['trailer']:
trailers1.append(trailer)
else:
trailers2.append(trailer)
else:
if trailer['trailer']:
trailers3.append(trailer)
else:
trailers4.append(trailer)
# склеиваем трейлеры
res['trailers'].extend(trailers1)
res['trailers'].extend(trailers2)
res['trailers'].extend(trailers3)
res['trailers'].extend(trailers4)
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if 'year' not in res['info'] or int(
res['info']['year']) >= time.gmtime(time.time()).tm_year:
timeout = 7 * 24 * 60 * 60 #week
return timeout, res
class KinoPoisk:
"""
API:
scraper - скрапер
movie - профайл фильма
search - поиск фильма
best - поиск лучших фильмов
person - поиск персон
work - информация о работах персоны
"""
def __init__(self):
self.cache = Cache('kinopoisk.db')
self.html = Clear()
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.kinopoisk.ru/level/7/'
}
# API
def scraper(self, name, year=None, trailer_quality=None):
try:
tag = 'scraper:' + urllib.quote_plus(name.encode('windows-1251'))
except:
return None
else:
if year:
tag += ':' + str(year)
id = self.cache.get(tag, self._scraper, name, year)
if not id:
return None
return self.movie(id, trailer_quality)
def movie(self, id, trailer_quality=None):
id = str(id)
if trailer_quality is None:
trailer_quality = 6
movie = self.cache.get('movie:' + id, self._movie, id)
if not movie:
return None
if movie['trailers']:
# компилируем список с нужным нам качеством
video = []
for m in movie['trailers']:
url = [x for x in m['video'] if x[0] <= trailer_quality]
if url:
m['video'] = url[-1]
video.append(m)
movie['trailers'] = video
if movie['trailers']:
# готовим главный трейлер
r = [x for x in movie['trailers'] if x['trailer']]
if r:
movie['info']['trailer'] = r[0]['video'][1]
else:
# если трейлер не найден, то отдаем что попало...
movie['info']['trailer'] = movie['trailers'][0]['video'][1]
return movie
def search(self, name, trailer_quality=None):
return self._search_movie(name)
def best(self, **kwarg):
page = kwarg.get('page', 1)
limit = kwarg.get('limit', 50)
url = 'http://www.kinopoisk.ru/top/navigator/m_act%5Bis_film%5D/on/m_act%5Bnum_vote%5D/' + str(kwarg.get('votes', 100)) + '/'
if kwarg.get('dvd'):
url += 'm_act%5Bis_dvd%5D/on/'
if kwarg.get('decade'):
url += 'm_act%5Bdecade%5D/' + str(kwarg['decade']) + '/'
if kwarg.get('genre'):
url += 'm_act%5Bgenre%5D/' + str(GENRE[kwarg['genre']]) + '/'
if kwarg.get('country'):
url += 'm_act%5Bcountry%5D/' + str(kwarg['country']) + '/'
if kwarg.get('rate'):
url += 'm_act%5Brating%5D/' + str(kwarg['rate']) + ':/'
if kwarg.get('mpaa'):
url += 'm_act%5Bmpaa%5D/' + str(kwarg['mpaa']) + '/'
url += 'perpage/' + str(limit) + '/order/ex_rating/'
if page > 1:
url += 'page/' + str(page) + '/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
res = {'pages': (1, 0, 1, 0), 'data': []}
r = re.compile('<div class="pagesFromTo(.+?)<div class="pagesFromTo', re.U|re.S).search(response.body.decode('windows-1251'))
if r:
body = r.group(1)
# compile pagelist
p = re.compile('>([0-9]+)—[0-9]+[^0-9]+?([0-9]+)', re.U).search(body)
if p:
page = (int(p.group(1))-1)/limit + 1
total = int(p.group(2))
pages = total/limit
if limit*pages != total:
pages += 1
res['pages'] = (pages, 0 if page == 1 else page-1, page, 0 if page==pages else page+1)
# end compile
for id in re.compile('<div id="tr_([0-9]+)"', re.U|re.S).findall(body):
res['data'].append(int(id))
return res
def person(self, name):
response = self.http.fetch('http://www.kinopoisk.ru/s/type/people/list/1/find/' + urllib.quote_plus(name.encode('windows-1251')) + '/order/relevant/', headers=self.headers)
if response.error:
return None
res = []
body = re.compile('<div class="navigator">(.+?)<div class="navigator">', re.U|re.S).search(response.body.decode('windows-1251'))
if body:
for block in re.compile('<p class="pic">(.+?)<div class="clear">', re.U|re.S).findall(body.group(1)):
id, name, original, year, poster = None, None, None, None, None
r = re.compile('<p class="name"><a href="http://www\.kinopoisk\.ru/level/4/people/([0-9]+)[^>]+>([^<]+)</a>', re.U|re.S).search(block)
if r:
id = r.group(1)
name = r.group(2).strip()
if id and name:
r = re.compile('<span class="gray">([^<]+)</span>', re.U|re.S).search(block)
if r:
original = r.group(1).strip()
if not original:
original = None
r = re.compile('<span class="year">([0-9]{4})</span>', re.U|re.S).search(block)
if r:
year = int(r.group(1))
if block.find('no-poster.gif') == -1:
poster = 'http://st.kinopoisk.ru/images/actor/' + id + '.jpg'
res.append({'id': int(id), 'name': name, 'originalname': original, 'year': year, 'poster': poster})
return {'pages': (1, 0, 1, 0), 'data': res}
def work(self, id):
response = self.http.fetch('http://www.kinopoisk.ru/name/' + str(id) + '/', headers=self.headers)
if response.error:
return None
res = {}
r = re.compile('id="sort_block">(.+?)<style>', re.U|re.S).search(response.body.decode('windows-1251'))
if r:
for block in r.group(1).split(u'<table cellspacing="0" cellpadding="0" border="0" width="100%">'):
work = None
for w in ('actor', 'director', 'writer', 'producer', 'producer_ussr', 'composer', 'operator', 'editor', 'design', 'voice', 'voice_director'):
if block.find(u'id="' + w + u'"') != -1:
work = 'producer' if w == 'producer_ussr' else w
break
if work:
movies = []
for id, name in re.compile('<span class="name"><a href="/film/([0-9]+)/" >([^<]+?)</a>', re.U).findall(block):
for tag in (u'(мини-сериал)', u'(сериал)'):
if name.find(tag) != -1:
break
else:
movies.append(int(id))
if movies:
res.setdefault(work, []).extend(movies)
return res
def review(self, id, query):
query_s = 'all' if query == 'stat' else query
data = self.cache.get('review:' + str(id) + ':' + query_s, self._review, id, query_s)
if not data:
return data
return data[query]
def countries(self):
return COUNTRIES
def country(self, id, default=None):
country = [x[1] for x in COUNTRIES if x[0] == id]
return country[0] if country else default
# PRIVATE
def _search_movie(self, name, year=None):
url = 'http://www.kinopoisk.ru/s/type/film/list/1/find/' + urllib.quote_plus(name.encode('windows-1251')) + '/order/relevant'
if year:
url += '/m_act%5Byear%5D/' + str(year)
url += '/m_act%5Btype%5D/film/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
res = []
r = re.compile('<div class="navigator">(.+?)<div class="navigator">', re.U|re.S).search(response.body.decode('windows-1251'))
if r:
for id in re.compile('<p class="name"><a href="/level/1/film/([0-9]+)', re.U|re.S).findall(r.group(1)):
res.append(int(id))
return {'pages': (1, 0, 1, 0), 'data': res}
def _scraper(self, name, year):
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if year and year >= time.gmtime(time.time()).tm_year:
timeout = 7*24*60*60 #week
ids = self._search_movie(name, year)
if ids is None:
return False, None
elif not ids['data']:
# сохраняем пустой результат на 3-е суток
return 259200, None
else:
return timeout, ids['data'][0]
def _review(self, id, query):
url = 'http://www.kinopoisk.ru/film/' + str(id) + '/ord/rating/'
if query in ('good', 'bad', 'neutral'):
url += 'status/' + query + '/'
url += 'perpage/200/'
response = self.http.fetch(url, headers=self.headers)
if response.error:
return False, None
html = response.body.decode('windows-1251')
res = {
'stat': {'all': 0, 'good': 0, 'bad': 0, 'neutral': 0},
query: []
}
r = re.compile('<ul class="resp_type">(.+?)</ul>', re.U|re.S).search(html)
if r:
ul = r.group(1)
for q, t in (('pos', 'good'), ('neg', 'bad'), ('neut', 'neutral')):
r = re.compile('<li class="' + q + '"><a href="[^>]+>[^<]+</a><b>([0-9]+)</b></li>', re.U).search(ul)
if r:
res['stat'][t] = int(r.group(1))
res['stat']['all'] = res['stat']['good'] + res['stat']['bad'] + res['stat']['neutral']
r = re.compile('<div class="navigator">(.+?)<div class="navigator">', re.U|re.S).search(html)
if r:
for block in r.group(1).split('itemprop="reviews"'):
review = {
'nick': None,
'count': None,
'title': None,
'review': None,
'time': None
}
r = re.compile('itemprop="reviewBody">(.+?)</div>', re.U|re.S).search(block)
if r:
text = r.group(1)
for tag1, tag2 in ((u'<=end=>', u'\n'), (u'<b>', u'[B]'), (u'</b>', u'[/B]'), (u'<i>', u'[I]'), (u'</i>', u'[/I]'), (u'<u>', u'[U]'), (u'</u>', u'[/U]')):
text = text.replace(tag1, tag2)
r = self.html.text(text)
if r:
review['review'] = r
user = None
r = re.compile('<p class="profile_name"><s></s><a href="[^>]+>([^<]+)</a></p>').search(block)
if r:
user = self.html.string(r.group(1))
else:
r = re.compile('<p class="profile_name"><s></s>([^<]+)</p>').search(block)
if r:
user = self.html.string(r.group(1))
if user:
review['nick'] = user
r = re.compile('<p class="sub_title"[^>]+>([^<]+)</p>').search(block)
if r:
title = self.html.string(r.group(1))
if title:
review['title'] = title
r = re.compile('<span class="date">([^<]+)</span>', re.U|re.S).search(block)
if r:
review['time'] = r.group(1).replace(u' |', u',')
r = re.compile(u'<a href="[^>]+>рецензии \(([0-9]+)\)</a>', re.U|re.S).search(block)
if r:
review['count'] = int(r.group(1))
if review['nick'] and review['review']:
res[query].append(review)
return 3600, res # one hour
def _movie(self, id):
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/', headers=self.headers)
if response.error:
return False, None
html = response.body.decode('windows-1251')
res = {
'id': int(id),
'thumb': None,
'fanart': None,
'trailers': [],
'info': {}
}
# имя, оригинальное имя, девиз, цензура, год, top250
# runtime - длительность фильма (в отдельную переменную, иначе не видно размер файла)
for tag, reg, t in (
('title', '<title>(.+?)</title>', 'str'),
('originaltitle', 'itemprop="alternativeHeadline">([^<]*)</span>', 'str'),
('tagline', '<td style="color\: #555">«(.+?)»</td></tr>', 'str'),
('mpaa', 'images/mpaa/([^\.]+).gif', 'str'),
('runtime', '<td class="time" id="runtime">[^<]+<span style="color\: #999">/</span>([^<]+)</td>', 'str'),
('year', '<a href="/lists/m_act%5Byear%5D/([0-9]+)/"', 'int'),
('top250', 'Топ250\: <a\shref="/level/20/#([0-9]+)', 'int')
):
r = re.compile(reg, re.U).search(html)
if r:
value = r.group(1).strip()
if value:
res['info'][tag] = value
if t == 'int':
res['info'][tag] = int(res['info'][tag])
else:
res['info'][tag] = self.html.string(res['info'][tag])
# режисеры, сценаристы, жанры
for tag, reg in (
('director', u'<td itemprop="director">(.+?)</td>'),
('writer', u'<td class="type">сценарий</td><td[^>]*>(.+?)</td>'),
('genre', u'<span itemprop="genre">(.+?)</span>')
):
r = re.compile(reg, re.U|re.S).search(html)
if r:
r2 = []
for r in re.compile('<a href="[^"]+">([^<]+)</a>', re.U).findall(r.group(1)):
r = self.html.string(r)
if r and r != '...':
r2.append(r)
if r2:
res['info'][tag] = u', '.join(r2)
# актеры
r = re.compile(u'<h4>В главных ролях:</h4>(.+?)</ul>', re.U|re.S).search(html)
if r:
actors = []
for r in re.compile('<li itemprop="actors"><a [^>]+>([^<]+)</a></li>', re.U).findall(r.group(1)):
r = self.html.string(r)
if r and r != '...':
actors.append(r)
if actors:
res['info']['cast'] = actors[:]
#res['info']['castandrole'] = actors[:]
# описание фильма
r = re.compile('<span class="_reachbanner_"><div class="brand_words" itemprop="description">(.+?)</div></span>', re.U).search(html)
if r:
plot = self.html.text(r.group(1).replace('<=end=>', '\n'))
if plot:
res['info']['plot'] = plot
# IMDB
r = re.compile('IMDb: ([0-9.]+) \(([0-9\s]+)\)</div>', re.U).search(html)
if r:
res['info']['rating'] = float(r.group(1).strip())
res['info']['votes'] = r.group(2).strip()
# премьера
r = re.compile(u'премьера \(мир\)</td>(.+?)</tr>', re.U|re.S).search(html)
if r:
r = re.compile(u'data\-ical\-date="([^"]+)"', re.U|re.S).search(r.group(1))
if r:
data = r.group(1).split(' ')
if len(data) == 3:
i = 0
for mon in (u'января', u'февраля', u'марта', u'апреля', u'мая', u'июня', u'июля', u'августа', u'сентября', u'октября', u'ноября', u'декабря'):
i += 1
if mon == data[1]:
mon = str(i)
if len(mon) == 1:
mon = '0' + mon
day = data[0]
if len(day) == 1:
day = '0' + day
res['info']['premiered'] = '-'.join([data[2], mon, day])
break
# постер
r = re.compile(u'onclick="openImgPopup\(([^\)]+)\)', re.U|re.S).search(html)
if r:
poster = r.group(1).replace("'", '').strip()
if poster:
res['thumb'] = 'http://kinopoisk.ru' + poster
menu = re.compile('<ul id="newMenuSub" class="clearfix(.+?)<!\-\- /menu \-\->', re.U|re.S).search(html)
if menu:
menu = menu.group(1)
# фанарт
if menu.find('/film/' + id + '/wall/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/wall/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
fanart = re.compile('<a href="/picture/([0-9]+)/w_size/([0-9]+)/">', re.U).findall(html)
if fanart:
fanart.sort(cmp=lambda (id1, size1), (id2, size2): cmp(int(size1), int(size2)))
# пробуем взять максимально подходящее
fanart_best = [x for x in fanart if int(x[1]) <= 1280]
if fanart_best:
fanart = fanart_best
response = self.http.fetch('http://www.kinopoisk.ru/picture/' + fanart[-1][0] + '/w_size/' + fanart[-1][1] + '/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile('id="image" src="([^"]+)"', re.U|re.S).search(html)
if r:
res['fanart'] = r.group(1).strip()
# если нет фанарта (обоев), то пробуем получить кадры
if not res['fanart'] and menu.find('/film/' + id + '/stills/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/stills/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
fanart = re.compile('<a href="/picture/([0-9]+)/"><img src="[^<]+</a>[^<]+<b><i>([0-9]+)×([0-9]+)</i>', re.U).findall(html)
if fanart:
fanart.sort(cmp=lambda (id1, size1, t1), (id2, size2, t2): cmp(int(size1), int(size2)))
# пробуем взять максимально подходящее
fanart_best = [x for x in fanart if int(x[1]) <= 1280 and int(x[1]) > int(x[2])]
if fanart_best:
fanart = fanart_best
response = self.http.fetch('http://www.kinopoisk.ru/picture/' + fanart[-1][0] + '/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile('id="image" src="([^"]+)"', re.U|re.S).search(html)
if r:
res['fanart'] = r.group(1).strip()
# студии
if menu.find('/film/' + id + '/studio/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/studio/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
r = re.compile(u'<b>Производство:</b>(.+?)</table>', re.U|re.S).search(html)
if r:
studio = []
for r in re.compile('<a href="/lists/m_act%5Bstudio%5D/[0-9]+/" class="all">(.+?)</a>', re.U).findall(r.group(1)):
r = self.html.string(r)
if r:
studio.append(r)
if studio:
res['info']['studio'] = u', '.join(studio)
# трэйлеры
trailers1 = [] # русские трейлеры
trailers2 = [] # другие русские видео
trailers3 = [] # трейлеры
trailers4 = [] # другие видео
if menu.find('/film/' + id + '/video/') != -1:
response = self.http.fetch('http://www.kinopoisk.ru/film/' + id + '/video/', headers=self.headers)
if not response.error:
html = response.body.decode('windows-1251')
for row in re.compile(u'<!-- ролик -->(.+?)<!-- /ролик -->', re.U|re.S).findall(html):
# отсекаем лишние блоки
if row.find(u'>СМОТРЕТЬ</a>') != -1:
# русский ролик?
if row.find('class="flag flag2"') == -1:
is_ru = False
else:
is_ru = True
# получаем имя трейлера
r = re.compile('<a href="/film/' + id + '/video/[0-9]+/[^>]+ class="all">(.+?)</a>', re.U).search(row)
if r:
name = self.html.string(r.group(1))
if name:
trailer = {
'name': name,
'time': None,
'trailer': False,
'ru': is_ru,
'video': []
}
# трейлер или тизер?
for token in (u'Трейлер', u'трейлер', u'Тизер', u'тизер'):
if name.find(token) != -1:
trailer['trailer'] = True
break
# получаем время трейлера
r = re.compile(u'clock.gif"[^>]+></td>\s*<td style="color\: #777">[^0-9]*([0-9\:]+)</td>', re.U|re.S).search(row)
if r:
trailer['time'] = r.group(1).strip()
print 'F**K'
# делим ролики по качеству
for r in re.compile('trailer/([1-3])a.gif"(.+?)link=([^"]+)" class="continue">.+?<td style="color\:#777">([^<]+)</td>\s*</tr>', re.U|re.S).findall(row):
print str(r)
quality = int(r[0])
if r[1].find('icon-hd') != -1:
quality += 3
trailer['video'].append((quality, r[2].strip(), r[3]))
print str(trailer)
if id == '462754':
#raise
pass
if trailer['video']:
if trailer['ru']:
if trailer['trailer']:
trailers1.append(trailer)
else:
trailers2.append(trailer)
else:
if trailer['trailer']:
trailers3.append(trailer)
else:
trailers4.append(trailer)
# склеиваем трейлеры
res['trailers'].extend(trailers1)
res['trailers'].extend(trailers2)
res['trailers'].extend(trailers3)
res['trailers'].extend(trailers4)
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if 'year' not in res['info'] or int(res['info']['year']) >= time.gmtime(time.time()).tm_year:
timeout = 7*24*60*60 #week
return timeout, res
class TvDb:
"""
API:
scraper - скрапер
search - поиск сериалов
movie - профайл фильма
"""
def __init__(self):
self.api_key = '1D62F2F90030C444'
self.cache = Cache('tvdb.db')
self.http = HTTP()
self.headers = {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.thetvdb.com/'
}
# API
def scraper(self, name, year=None):
try:
tag = 'scraper:' + urllib.quote_plus(name.encode('utf8'))
except:
return None
else:
if year:
tag += ':' + str(year)
id = self.cache.get(tag, self._scraper, name, year)
if not id:
return None
return self.movie(id)
def search(self, name):
return self._search(name)
def movie(self, id):
id = str(id)
return self.cache.get('movie:' + id, self._movie, id)
def _movie(self, id):
dirname = tempfile.mkdtemp()
response = self.http.fetch('http://www.thetvdb.com/api/' +
self.api_key + '/series/' + id +
'/all/ru.zip',
headers=self.headers,
download=os.path.join(dirname, 'movie.zip'))
if response.error:
self._movie_clear(dirname)
return False, None
try:
filezip = zipfile.ZipFile(os.path.join(dirname, 'movie.zip'), 'r')
filezip.extractall(dirname)
filezip.close()
movie = file(os.path.join(dirname, 'ru.xml'),
'rb').read().decode('utf8')
except:
self._movie_clear(dirname)
return False, None
self._movie_clear(dirname)
body = re.compile(r'<Series>(.+?)</Series>', re.U | re.S).search(movie)
if not body:
return False, None
body = body.group(1)
res = {
'id': int(id),
'thumb': None,
'fanart': None,
'trailers': [],
'info': {}
}
# режисеры и сценаристы
for tag in ('Director', 'Writer'):
people = {}
people_list = []
[
people_list.extend(x.split('|'))
for x in re.compile(r'<' + tag + r'>([^<]+)</' + tag +
r'>', re.U | re.S).findall(movie)
]
[
people.update({x: 1})
for x in [x.strip() for x in people_list] if x
]
if people:
res['info'][tag.lower()] = u', '.join(
[x for x in people.keys() if x])
for tag, retag, typeof in (('plot', 'Overview',
None), ('mpaa', 'ContentRating', None),
('premiered', 'FirstAired',
None), ('studio', 'Network', None),
('title', 'SeriesName',
None), ('runtime', 'Runtime', None),
('votes', 'RatingCount',
None), ('rating', 'Rating', float),
('genre', 'Genre', list), ('cast', 'Actors',
list)):
r = re.compile(r'<' + retag + r'>([^<]+)</' + retag + r'>',
re.U | re.S).search(body)
if r:
r = r.group(1).strip()
if typeof == float:
res['info'][tag] = float(r)
elif typeof == list:
res['info'][tag] = [
x for x in [x.strip() for x in r.split(u'|')] if x
]
if tag == 'genre':
res['info'][tag] = u', '.join(res['info'][tag])
else:
res['info'][tag] = r
# год
if 'premiered' in res['info']:
res['info']['year'] = int(res['info']['premiered'].split('-')[0])
# постер
r = re.compile(r'<poster>([^<]+)</poster>', re.U | re.S).search(body)
if r:
res['thumb'] = 'http://thetvdb.com/banners/' + r.group(1).strip()
# фанарт
r = re.compile(r'<fanart>([^<]+)</fanart>', re.U | re.S).search(body)
if r:
res['fanart'] = 'http://thetvdb.com/banners/' + r.group(1).strip()
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if 'year' not in res['info'] or int(
res['info']['year']) >= time.gmtime(time.time()).tm_year:
timeout = 7 * 24 * 60 * 60 #week
return timeout, res
def _movie_clear(self, dirname):
for filename in os.listdir(dirname):
try:
os.unlink(os.path.join(dirname, filename))
except:
raise
try:
os.rmdir(dirname)
except:
raise
def _search(self, name):
response = self.http.fetch(
'http://www.thetvdb.com/api/GetSeries.php?language=ru&seriesname='
+ urllib.quote_plus(name.encode('utf8')),
headers=self.headers)
if response.error:
return None
res = []
rows = re.compile('<Series>(.+?)</Series>',
re.U | re.S).findall(response.body.decode('utf8'))
if rows:
recmd = re.compile('<seriesid>([0-9]+)</seriesid>', re.U | re.S)
for row in [
x for x in rows if x.find(u'<language>ru</language>') != -1
]:
r = recmd.search(row)
if r:
res.append(int(r.group(1)))
return {'pages': (1, 0, 1, 0), 'data': res}
def _scraper(self, name, year):
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if year and year >= time.gmtime(time.time()).tm_year:
timeout = 7 * 24 * 60 * 60 #week
ids = self._search(name)
if ids is None:
return False, None
elif not ids['data']:
# сохраняем пустой результат на 3-е суток
return 259200, None
else:
return timeout, ids['data'][0]
class RuTrackerHTTP:
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'"profile\.php\?mode=sendpassword"')
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.org/forum/index.php'
}
def guest(self, url):
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
return body
def get(self, url):
return self._fetch('GET', url)
def post(self, url, params):
return self._fetch('POST', url, params)
def download(self, id):
id = str(id)
# проверяем авторизацию
html = self.get('http://rutracker.org/forum/viewtopic.php?t=' + id)
if not html:
return html
# хакаем куки
cookies = cookielib.MozillaCookieJar()
cookies.load(self.http.request.cookies)
cookies.set_cookie(cookielib.Cookie(version=0, name='bb_dl', value=id, port=None, port_specified=False, domain='.rutracker.org', domain_specified=False, domain_initial_dot=False, path='/', path_specified=True, secure=False, expires=None, discard=True, comment=None, comment_url=None, rest={'HttpOnly': None}, rfc2109=False))
cookies.save(self.http.request.cookies, ignore_discard=True, ignore_expires=True)
# тянем торрент
response = self.http.fetch('http://dl.rutracker.org/forum/dl.php?t=' + id, cookies='rutracker.moz', headers=self.headers, method='POST')
if response.error:
return None
else:
return response.body
def _fetch(self, method, url, params=None):
while True:
response = self.http.fetch(url, cookies='rutracker.moz', headers=self.headers, method=method, params=params)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return body
else:
xbmc.log('RUTRACKER: Request auth', xbmc.LOGDEBUG)
auth = self._auth()
if not auth:
return auth
def _auth(self):
while True:
login = self.setting['rutracker_login']
password = self.setting['rutracker_password']
if not login or not password:
login, password = self._setting(login, password)
if not login:
return False
response = self.http.fetch('http://login.rutracker.org/forum/login.php', cookies='rutracker.moz', headers=self.headers, method='POST', params={'login_username': login, 'login_password': password, 'login': r'Вход'})
if response.error:
return False
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return True
else:
login, password = self._setting(login, password)
if not login:
return False
def _setting(self, login, password):
self.setting.dialog()
login2 = self.setting['rutracker_login']
password2 = self.setting['rutracker_password']
if login == login2 and password == password2:
return None, None
else:
return login2, password2
class TvDb:
"""
API:
scraper - скрапер
search - поиск сериалов
movie - профайл фильма
"""
def __init__(self):
self.api_key = '1D62F2F90030C444'
self.cache = Cache('tvdb.db')
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; rv:10.0.2) Gecko/20100101 Firefox/10.0.2',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://www.thetvdb.com/'
}
# API
def scraper(self, name, year=None):
try:
tag = 'scraper:' + urllib.quote_plus(name.encode('utf8'))
except:
return None
else:
if year:
tag += ':' + str(year)
id = self.cache.get(tag, self._scraper, name, year)
if not id:
return None
return self.movie(id)
def search(self, name):
return self._search(name)
def movie(self, id):
id = str(id)
return self.cache.get('movie:' + id, self._movie, id)
def _movie(self, id):
dirname = tempfile.mkdtemp()
response = self.http.fetch('http://www.thetvdb.com/api/' + self.api_key + '/series/' + id + '/all/ru.zip', headers=self.headers, download=os.path.join(dirname, 'movie.zip'))
if response.error:
self._movie_clear(dirname)
return False, None
try:
filezip = zipfile.ZipFile(os.path.join(dirname, 'movie.zip'), 'r')
filezip.extractall(dirname)
filezip.close()
movie = file(os.path.join(dirname, 'ru.xml'), 'rb').read().decode('utf8')
except:
self._movie_clear(dirname)
return False, None
self._movie_clear(dirname)
body = re.compile(r'<Series>(.+?)</Series>', re.U|re.S).search(movie)
if not body:
return False, None
body = body.group(1)
res = {
'id': int(id),
'thumb': None,
'fanart': None,
'trailers': [],
'info': {}
}
# режисеры и сценаристы
for tag in ('Director', 'Writer'):
people = {}
people_list = []
[people_list.extend(x.split('|')) for x in re.compile(r'<' + tag + r'>([^<]+)</' + tag + r'>', re.U|re.S).findall(movie)]
[people.update({x: 1}) for x in [x.strip() for x in people_list] if x]
if people:
res['info'][tag.lower()] = u', '.join([x for x in people.keys() if x])
for tag, retag, typeof in (
('plot', 'Overview', None),
('mpaa', 'ContentRating', None),
('premiered', 'FirstAired', None),
('studio', 'Network', None),
('title', 'SeriesName', None),
('runtime', 'Runtime', None),
('votes', 'RatingCount', None),
('rating', 'Rating', float),
('genre', 'Genre', list),
('cast', 'Actors', list)
):
r = re.compile(r'<' + retag + r'>([^<]+)</' + retag + r'>', re.U|re.S).search(body)
if r:
r = r.group(1).strip()
if typeof == float:
res['info'][tag] = float(r)
elif typeof == list:
res['info'][tag] = [x for x in [x.strip() for x in r.split(u'|')] if x]
if tag == 'genre':
res['info'][tag] = u', '.join(res['info'][tag])
else:
res['info'][tag] = r
# год
if 'premiered' in res['info']:
res['info']['year'] = int(res['info']['premiered'].split('-')[0])
# постер
r = re.compile(r'<poster>([^<]+)</poster>', re.U|re.S).search(body)
if r:
res['thumb'] = 'http://thetvdb.com/banners/' + r.group(1).strip()
# фанарт
r = re.compile(r'<fanart>([^<]+)</fanart>', re.U|re.S).search(body)
if r:
res['fanart'] = 'http://thetvdb.com/banners/' + r.group(1).strip()
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if 'year' not in res['info'] or int(res['info']['year']) >= time.gmtime(time.time()).tm_year:
timeout = 7*24*60*60 #week
return timeout, res
def _movie_clear(self, dirname):
for filename in os.listdir(dirname):
try:
os.unlink(os.path.join(dirname, filename))
except:
raise
try:
os.rmdir(dirname)
except:
raise
def _search(self, name):
response = self.http.fetch('http://www.thetvdb.com/api/GetSeries.php?language=ru&seriesname=' + urllib.quote_plus(name.encode('utf8')), headers=self.headers)
if response.error:
return None
res = []
rows = re.compile('<Series>(.+?)</Series>', re.U|re.S).findall(response.body.decode('utf8'))
if rows:
recmd = re.compile('<seriesid>([0-9]+)</seriesid>', re.U|re.S)
for row in [x for x in rows if x.find(u'<language>ru</language>') != -1]:
r = recmd.search(row)
if r:
res.append(int(r.group(1)))
return {'pages': (1, 0, 1, 0), 'data': res}
def _scraper(self, name, year):
timeout = True
# если фильм свежий, то кладем в кэш НЕ на долго (могут быть обновления на сайте)
if year and year >= time.gmtime(time.time()).tm_year:
timeout = 7*24*60*60 #week
ids = self._search(name)
if ids is None:
return False, None
elif not ids['data']:
# сохраняем пустой результат на 3-е суток
return 259200, None
else:
return timeout, ids['data'][0]
class RuTrackerHTTP:
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'profile\.php\?mode=sendpassword"')
self.re_captcha = re.compile(r'<img src="(\/\/[^\/]+/captcha/[^"]+)"')
self.re_captcha_sid = re.compile(
r'<input type="hidden" name="cap_sid" value="([^"]+)">')
self.re_captcha_code = re.compile(
r'<input type="text" name="(cap_code_[^"]+)"')
self.captcha_sid = None
self.captcha_code = None
self.captcha_code_value = None
self.http = HTTP()
self.headers = {
'User-Agent':
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36',
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.lib/forum/index.php'
}
def guest(self, url):
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
return body
def get(self, url):
return self._fetch('GET', url)
def post(self, url, params):
return self._fetch('POST', url, params)
def download(self, id):
id = str(id)
# проверяем авторизацию
html = self.get('http://rutracker.lib/forum/viewtopic.php?t=' + id)
if not html:
return html
# хакаем куки
cookies = cookielib.MozillaCookieJar()
cookies.load(self.http.request.cookies)
cookies.set_cookie(
cookielib.Cookie(version=0,
name='bb_dl',
value=id,
port=None,
port_specified=False,
domain='.rutracker.lib',
domain_specified=False,
domain_initial_dot=False,
path='/',
path_specified=True,
secure=False,
expires=None,
discard=True,
comment=None,
comment_url=None,
rest={'HttpOnly': None},
rfc2109=False))
cookies.save(self.http.request.cookies,
ignore_discard=True,
ignore_expires=True)
# тянем торрент
response = self.http.fetch('http://rutracker.lib/forum/dl.php?t=' + id,
cookies='rutracker.moz',
headers=self.headers,
method='POST')
if response.error:
return None
else:
return response.body
def _fetch(self, method, url, params=None):
while True:
response = self.http.fetch(url,
cookies='rutracker.moz',
headers=self.headers,
method=method,
params=params)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return body
xbmc.log('RUTRACKER: Request auth', xbmc.LOGDEBUG)
auth = self._auth()
if not auth:
return auth
def _auth(self):
self.captcha_sid, self.captcha_code, self.captcha_code_value = None, None, None
while True:
login = self.setting['rutracker_login']
password = self.setting['rutracker_password']
if not login or not password:
self.setting.dialog()
login = self.setting['rutracker_login']
password = self.setting['rutracker_password']
if not login or not password:
return None
params = {
'login_username': login,
'login_password': password,
'login': r'вход'
}
if self.captcha_sid:
params['login'] = r'Вход'
params['cap_sid'] = self.captcha_sid
params[self.captcha_code] = self.captcha_code_value
response = self.http.fetch('http://rutracker.lib/forum/login.php',
cookies='rutracker.moz',
headers=self.headers,
method='POST',
params=params)
self.captcha_sid, self.captcha_code, self.captcha_code_value = None, None, None
if response.error:
return None
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return True
# проверяем капчу
r = self.re_captcha.search(body)
if r:
r_sid = self.re_captcha_sid.search(body)
if not r_sid:
return None
self.captcha_sid = r_sid.group(1)
r_code = self.re_captcha_code.search(body)
if not r_code:
return None
self.captcha_code = r_code.group(1)
self.captcha_code_value = self._captcha('http:' + r.group(1))
if not self.captcha_code_value:
return None
# get login
k = xbmc.Keyboard('', 'Enter login')
k.doModal()
if k.isConfirmed():
login = k.getText()
else:
return None
# get password
k = xbmc.Keyboard('', 'Enter password', True)
k.doModal()
if k.isConfirmed():
password = k.getText()
else:
return None
if not login or not password:
return None
self.setting['rutracker_login'] = login
self.setting['rutracker_password'] = password
def _captcha(self, captcha):
response = self.http.fetch(captcha, headers=self.headers, method='GET')
if response.error:
return
import tempfile
filename = tempfile.gettempdir() + '/captcha'
file(filename, 'wb').write(response.body)
win = xbmcgui.Window(xbmcgui.getCurrentWindowId())
# width = 120px, height = 72px
image = xbmcgui.ControlImage(win.getWidth() / 2 - int(120 / 2), 20,
120, 72, filename)
win.addControl(image)
k = xbmc.Keyboard('', 'Enter captcha code')
k.doModal()
code = k.getText() if k.isConfirmed() else None
win.removeControl(image)
return code if code else None
class RuTrackerHTTP:
def __init__(self):
self.setting = Setting()
self.re_auth = re.compile(r'profile\.php\?mode=sendpassword"')
self.re_captcha = re.compile(r'<img src="(\/\/[^\/]+/captcha/[^"]+)"')
self.re_captcha_sid = re.compile(r'<input type="hidden" name="cap_sid" value="([^"]+)">')
self.re_captcha_code = re.compile(r'<input type="text" name="(cap_code_[^"]+)"')
self.captcha_sid = None
self.captcha_code = None
self.captcha_code_value = None
self.http = HTTP()
self.headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'ru-ru,ru;q=0.8,en-us;q=0.5,en;q=0.3',
'Cache-Control': 'no-cache',
'Referer': 'http://rutracker.nl/forum/index.php'
}
def guest(self, url):
response = self.http.fetch(url, headers=self.headers)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
return body
def get(self, url):
return self._fetch('GET', url)
def post(self, url, params):
return self._fetch('POST', url, params)
def download(self, id):
id = str(id)
# проверяем авторизацию
html = self.get('http://rutracker.nl/forum/viewtopic.php?t=' + id)
if not html:
return html
# хакаем куки
cookies = cookielib.MozillaCookieJar()
cookies.load(self.http.request.cookies)
cookies.set_cookie(cookielib.Cookie(version=0, name='bb_dl', value=id, port=None, port_specified=False, domain='.rutracker.nl', domain_specified=False, domain_initial_dot=False, path='/', path_specified=True, secure=False, expires=None, discard=True, comment=None, comment_url=None, rest={'HttpOnly': None}, rfc2109=False))
cookies.save(self.http.request.cookies, ignore_discard=True, ignore_expires=True)
# тянем торрент
response = self.http.fetch('http://rutracker.nl/forum/dl.php?t=' + id, cookies='rutracker.moz', headers=self.headers, method='POST')
if response.error:
return None
else:
return response.body
def _fetch(self, method, url, params=None):
while True:
response = self.http.fetch(url, cookies='rutracker.moz', headers=self.headers, method=method, params=params)
if response.error:
return None
else:
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return body
xbmc.log('RUTRACKER: Request auth', xbmc.LOGDEBUG)
auth = self._auth()
if not auth:
return auth
def _auth(self):
self.captcha_sid, self.captcha_code, self.captcha_code_value = None, None, None
while True:
login = self.setting['rutracker_login']
password = self.setting['rutracker_password']
if not login or not password:
self.setting.dialog()
login = self.setting['rutracker_login']
password = self.setting['rutracker_password']
if not login or not password:
return None
params = {'login_username': login, 'login_password': password, 'login': r'вход'}
if self.captcha_sid:
params['login'] = r'Вход'
params['cap_sid'] = self.captcha_sid
params[self.captcha_code] = self.captcha_code_value
response = self.http.fetch('http://rutracker.nl/forum/login.php', cookies='rutracker.moz', headers=self.headers, method='POST', params=params)
self.captcha_sid, self.captcha_code, self.captcha_code_value = None, None, None
if response.error:
return None
body = response.body.decode('windows-1251')
if body.find(u'>форум временно отключен</p>') != -1:
return 0
if not self.re_auth.search(body):
return True
# проверяем капчу
r = self.re_captcha.search(body)
if r:
r_sid = self.re_captcha_sid.search(body)
if not r_sid:
return None
self.captcha_sid = r_sid.group(1)
r_code = self.re_captcha_code.search(body)
if not r_code:
return None
self.captcha_code = r_code.group(1)
self.captcha_code_value = self._captcha('http:' + r.group(1))
if not self.captcha_code_value:
return None
# get login
k = xbmc.Keyboard('', 'Enter login')
k.doModal()
if k.isConfirmed():
login = k.getText()
else:
return None
# get password
k = xbmc.Keyboard('', 'Enter password', True)
k.doModal()
if k.isConfirmed():
password = k.getText()
else:
return None
if not login or not password:
return None
self.setting['rutracker_login'] = login
self.setting['rutracker_password'] = password
def _captcha(self, captcha):
response = self.http.fetch(captcha, headers=self.headers, method='GET')
if response.error:
return
import tempfile
filename = tempfile.gettempdir() + '/captcha'
file(filename, 'wb').write(response.body)
win = xbmcgui.Window(xbmcgui.getCurrentWindowId())
# width = 120px, height = 72px
image = xbmcgui.ControlImage(win.getWidth()/2 - int(120/2), 20, 120, 72, filename)
win.addControl(image)
k = xbmc.Keyboard('', 'Enter captcha code')
k.doModal()
code = k.getText() if k.isConfirmed() else None
win.removeControl(image)
return code if code else None
