def parse_episodes(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/episodes")
soup = BeautifulSoup(content)
seasons = soup.find('dl', {'class': 'episodesTable'})
episodes_list = []
for element in seasons.children:
if element.name == 'dt':
h3 = element.next_element
try:
season_number = int(h3.text.split(" ")[1])
except:
break
if element.name == 'dd':
li_episodes = element.find_all("li")
for episode in li_episodes:
episode_number = int(
re.match(r'\d+', episode.contents[0].text).group())
date_str = episode.find('div', {
'class': 'countryPremiereDate'
}).text
episode_name = episode.find('div', {'class': 'title'}).text
episode_date = datetime.datetime.strptime(
date_str, '%d.%m.%Y')
episodes_list.append({
'season': season_number,
'number': episode_number,
'date': episode_date,
'name': episode_name
})
return episodes_list
def parse_episodes(self):
grabber = HTMLGrabber()
content = grabber.retrieve( self.obj.url+"/episodes" )
soup = BeautifulSoup(content)
seasons = soup.findAll("table")
episodeList = []
for season in seasons:
try:
seasonNumber = season.find("h3").text.split()
if seasonNumber[0] == 'sezon':
seasonNumber = int(seasonNumber[1])
else:
seasonNumber = 0
except:
seasonNumber = 0
episodes = season.findAll("td")
for i in range(0, len(episodes), 3):
number = episodes[i].text.split()
if number[0] == 'odcinek':
number = int(number[1])
else:
number = 0
episodeDate = get_datetime_or_none(episodes[i+1].find('div'))
episodeName = episodes[i+2].text
episodeList.append({'season':seasonNumber, 'number':number, 'date':episodeDate, 'name': episodeName})
return episodeList
def _search_movie(self, title, results, genre_id, search_type, start_year, end_year):
"""Return list of movies"""
grabber = HTMLGrabber()
li_list = []
img_list = []
params = {"q": title.encode("utf-8"), "page": 1}
if genre_id: params['genreIds'] = genre_id
if start_year: params['startYear'] = start_year
if end_year: params['endYear'] = end_year
search_url = ""
if search_type:
search_url = "/" + search_type
url = filmweb_search_blank + search_url + "?" + urlencode(params)
content = grabber.retrieve(url) # @Make search more pages not only 1
soup = BeautifulSoup(content)
li_list.extend(soup.findAll('div', {'class': 'hitDescWrapper'}))
img_list.extend(soup.findAll('div', {'class': 'hitImage'}))
for i, li in enumerate(li_list):
a = li.find('a', {'class': 'hdr hdr-medium hitTitle'})
title = a.text
url = a['href']
# have to do another check because sometimes url doesnt provide movieID
aimg = img_list[i].find('a')
if aimg is not None:
img = aimg.find("img")
movieID = get_real_id(url, img['src'])
yield movieID, title, url
def _search_filtered_movie(self, title, results, genre_id, search_type):
grabber = HTMLGrabber()
params = {}
params['page'] = 1
if title: params['q'] = title.encode("utf-8")
if genre_id: params['genreIds'] = genre_id
search_url = ""
if search_type:
search_url = "/" + search_type
url = filmweb_search_blank + search_url + "?" + urllib.urlencode(params)
content = grabber.retrieve(url)
soup = BeautifulSoup(content)
hits = soup.findAll('li', {'id': re.compile('hit_([0-9]*)')})
for hit in hits:
h3 = hit.find("h3")
url = h3.find("a")['href']
div_img = hit.find("div", {'class': 'filmPoster-1'})
img = div_img.find("img")
movieID = get_real_id(url, img['src'])
yield movieID, title, url
def parse_posters(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/posters")
soup = BeautifulSoup(content)
photoList = soup.find("ul", 'block-list postersList')
images = []
for photo in photoList("img", {'class': "lbProxy"}):
images.append({'href': photo['src'].replace(".2.jpg", '.3.jpg'), 'thumb': photo['src']})
return images
def parse_photos(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/photos")
soup = BeautifulSoup(content)
photos_list = soup.find("ul", {'class','photosList'})
images = []
for photo in photos_list.findAll("img"):
images.append({'href': photo.parent['href'], 'thumb': photo['src'], 'image': photo.parent['data-photo']})
return images
def parse_cast(self):
personList = []
for url in ["/cast/actors", "/cast/crew"]:
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + url)
soup = BeautifulSoup(content)
for filmCastBox in soup.findAll("div", {'class': 'filmCastBox'}):
personType = filmCastBox.previous
personTypesChange = {
'obsada': 'aktor',
'scenariusz': 'scenarzysta',
'produkcja': 'producent'
} # backward compatibility
if personType in personTypesChange: #
personType = personTypesChange[personType]
for cast in filmCastBox.findAll('tr', id=re.compile("role_")):
url_html = cast.find("a", {'class': 'pImg49'})
url = url_html['href']
img_html = url_html.find("img")
pattern_images = [
"http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*).([0-3]*).jpg",
"http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*)_1.([0-3]*).jpg"
]
pattern_link = "/person/(.+)-(?P<id>[0-9]*)"
id = 0
results = re.search(pattern_link, url_html['href'])
if results:
id = results.group("id")
else:
for pattern in pattern_images:
results = re.search(pattern,
repr(img_html.extract()))
if results:
id = results.group("id")
role_html = cast.find('a', {'rel': 'v:starring'})
role = role_html.parent.nextSibling.nextSibling.text
name = role_html.parent.nextSibling.text
personList.append(
Person(id,
title=name,
roleType=personType,
roleName=role,
url=url))
return personList
def get_list_genres():
grabber = HTMLGrabber()
content = grabber.retrieve(filmweb_search_blank + "/film")
soup = BeautifulSoup(content)
genres = soup.findAll('input', {'name': 'genreIds'})
list_genre = []
for genre in genres:
genre_id = genre.attrs['value']
genre_name = genre.next_element.next_element.text
list_genre.append({'genre_id': genre_id, 'genre_name': genre_name})
return list_genre
def get_list_genres():
grabber = HTMLGrabber()
content = grabber.retrieve(filmweb_search_blank + "/film")
soup = BeautifulSoup(content)
genres = soup.findAll('input', {'name': 'genreIds'})
list_genre = []
for genre in genres:
genre_id = genre.attrs['value']
genre_name = genre.next_element.next_element.text
list_genre.append({'genre_id': genre_id, 'genre_name': genre_name})
return list_genre
def parse_posters(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/posters")
soup = BeautifulSoup(content)
photoList = soup.find("ul", 'block-list postersList')
images = []
for photo in photoList("img", {'class': "lbProxy"}):
images.append({
'href': photo['src'].replace(".2.jpg", '.3.jpg'),
'thumb': photo['src']
})
return images
def parse_photos(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/photos")
soup = BeautifulSoup(content)
photos_list = soup.find("ul", {'class', 'photosList'})
images = []
for photo in photos_list.findAll("img"):
images.append({
'href': photo.parent['href'],
'thumb': photo['src'],
'image': photo.parent['data-photo']
})
return images
def parse_cast(self):
personList = []
for url in ["/cast/actors", "/cast/crew"]:
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + url)
soup = BeautifulSoup(content)
for filmCastBox in soup.findAll("div", {'class': 'filmCastBox'}):
personType = filmCastBox.previous
personTypesChange = {'obsada': 'aktor', 'scenariusz': 'scenarzysta',
'produkcja': 'producent'} # backward compatibility
if personType in personTypesChange: #
personType = personTypesChange[personType]
for cast in filmCastBox.findAll('tr', id=re.compile("role_")):
url_html = cast.find("a", {'class': 'pImg49'})
url = url_html['href']
img_html = url_html.find("img")
pattern_images = [
"http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*).([0-3]*).jpg",
"http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*)_1.([0-3]*).jpg"
]
pattern_link = "/person/(.+)-(?P<id>[0-9]*)"
id = 0
results = re.search(pattern_link, url_html['href'])
if results:
id = results.group("id")
else:
for pattern in pattern_images:
results = re.search(pattern, repr(img_html.extract()))
if results:
id = results.group("id")
role_html = cast.find('a', {'rel': 'v:starring'})
role = role_html.parent.nextSibling.nextSibling.text
name = role_html.parent.nextSibling.text
personList.append(Person(id, title=name, roleType=personType, roleName=role, url=url))
return personList
def _search_movie(self,title,results,):
"""Return list of movies"""
grabber = HTMLGrabber()
p_title = grabber.encode_string(title)
li_list = []
img_list = []
#for type in ['film','serial']:
content = grabber.retrieve(filmweb_search % (p_title,1)) #@Make search more pages not only 1
soup = BeautifulSoup(content)
li_list.extend( soup.findAll('div', {'class':'hitDescWrapper'}) )
img_list.extend( soup.findAll('div', {'class':'hitImage'}) )
for i, li in enumerate(li_list):
a = li.find('a',{'class':re.compile('hdr.*')}) # class="hdr hdr-medium hitTitle" for now
title = a.text
url = a['href']
# have to do another check because sometimes url doesnt provide movieID
aimg = img_list[i].find('a')
if aimg is not None:
img = aimg.find("img")
movieID = get_real_id(url,img['src'])
yield movieID,title,url
def _search_person(self, title, results=20):
# http://www.filmweb.pl/search/person?q=Tom+Cruise
"""Return list of persons"""
grabber = HTMLGrabber()
p_title = grabber.encode_string(title)
li_list = []
img_list = []
content = grabber.retrieve(filmweb_person_search % (p_title, 1)) #@Make search more pages not only 1
soup = BeautifulSoup(content)
li_list.extend(soup.findAll('div', {'class': 'hitDescWrapper'}))
img_list.extend(soup.findAll('div', {'class': 'hitImage'}))
for i, li in enumerate(li_list):
a = li.find('a', {'class': 'hdr hdr-medium hitTitle'})
title = a.text
url = a['href']
# have to do another check because sometimes url doesnt provide movieID
aimg = img_list[i].find('a')
if aimg is not None:
img = aimg.find('img')
personID = get_real_id(url, img['src'])
yield personID, title, url
def parse_episodes(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/episodes")
soup = BeautifulSoup(content)
seasons = soup.find('dl', {'class': 'episodesTable'})
episodes_list = []
for element in seasons.children:
if element.name == 'dt':
h3 = element.next_element
try:
season_number = int(h3.text.split(" ")[1])
except:
break
if element.name == 'dd':
li_episodes = element.find_all("li")
for episode in li_episodes:
episode_number = int(re.match(r'\d+', episode.contents[0].text).group())
date_str = episode.find('div',{'class':'countryPremiereDate'}).text
episode_name = episode.find('div',{'class':'title'}).text
episode_date = datetime.datetime.strptime(date_str,'%d.%m.%Y')
episodes_list.append({'season': season_number, 'number': episode_number, 'date': episode_date, 'name': episode_name})
return episodes_list
def parse_cast(self):
grabber = HTMLGrabber()
content = grabber.retrieve( self.obj.url+"/cast" )
soup = BeautifulSoup(content)
castList = soup.find("div",{'class':'filmSubpageContentWrapper'})
castGroups = castList.findAll(["dd","dt"])
personList = []
for cast in castGroups:
if cast.name == 'dt':
roleType = cast.text.split("/")[0].strip()
roleType
elif cast.name == 'dd':
castList = cast.findAll("li")
for person in castList:
try:
role = person.find("span",{'class':'roleName'}).text
except AttributeError:
role = None
name = person.find("span",{'class':'personName'}).text
patternlink = "/person/(.+)-(?P<id>[0-9]*)"
patternimg = "http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*).([0-3]*).jpg"
href = person.find("span",{'class':'personName'}).find("a")['href']
results = re.search(patternlink,href)
if results:
id = results.group("id")
else:
results = re.search(patternimg,unicode(person.extract()))
id = results.group("id")
personList.append( Person(id,title=name,roleType=roleType,roleName=role,url=href) )
return personList
import re
import urllib
from bs4 import BeautifulSoup
from filmweb.parser.HTMLGrabber import HTMLGrabber
from filmweb.vars import filmweb_search_blank
grabber = HTMLGrabber()
params = {}
params['page'] = 1
title = None
genre_id = 2
search_type = "film"
if title: params['q'] = title.encode("utf-8")
if genre_id: params['genreIds'] = genre_id
search_url = ""
if search_type:
search_url = "/" + search_type
url = filmweb_search_blank + search_url + "?" + urllib.urlencode(params)
content = grabber.retrieve(url)
soup = BeautifulSoup(content)
hits = soup.find(id="resultsCount")
all = hits.get_text()
p = re.compile('z \d+')
all = p.findall(all)
all = all[0]
def parse_cast(self):
grabber = HTMLGrabber()
content = grabber.retrieve(self.obj.url + "/cast/actors")
soup = BeautifulSoup(content)
cast_list_table = soup.find("table", {'class': 'filmCast'})
cast_list = cast_list_table.find("tbody")
personList = []
for cast in cast_list_table.findAll('tr',id=re.compile("role_")):
url_html = cast.find("a",{'class':'pImg49'})
url = url_html['href']
img_html = url_html.find("img")
pattern_img = "http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*).([0-3]*).jpg"
pattern_link = "/person/(.+)-(?P<id>[0-9]*)"
results = re.search(pattern_link, url_html['href'])
if results:
id = results.group("id")
else:
results = re.search(pattern_img, unicode(img_html.extract()))
id = results.group("id")
role_html = cast.find('a',{'rel':'v:starring'})
role = role_html.parent.nextSibling.nextSibling.text
name = role_html.parent.nextSibling.text
personList.append(Person(id, title=name, roleType='aktor', roleName=role, url=url))
if cast.name == 'dt':
roleType = cast.text.split("/")[0].strip()
roleType
elif cast.name == 'dd':
castList = cast.findAll("li")
for person in castList:
try:
role = person.find("span", {'class': 'roleName'}).text
except AttributeError:
role = None
name = person.find("span", {'class': 'personName'}).text
patternlink = "/person/(.+)-(?P<id>[0-9]*)"
patternimg = "http://1.fwcdn.pl/p/([0-9]{2})/([0-9]{2})/(?P<id>[0-9]*)/([0-9]*).([0-3]*).jpg"
href = person.find("span", {'class': 'personName'}).find("a")['href']
results = re.search(patternlink, href)
if results:
id = results.group("id")
else:
results = re.search(patternimg, unicode(person.extract()))
id = results.group("id")
personList.append(Person(id, title=name, roleType=roleType, roleName=role, url=href))
return personList
def _download_content(self, url):
from filmweb.parser.HTMLGrabber import HTMLGrabber
grabber = HTMLGrabber()
self._content = grabber.retrieve(url)
self._soup = BeautifulSoup(self.content)
