def get_movie_url(movie_data):
global imdb_url_str
params_dict = {'s': 'tt'}
data_list = movie_data.split('*')
params_dict['q'] = data_list[1] if data_list[1] else romanize(data_list[0])
response = get_response('/find?' + urllib.urlencode(params_dict))
# first check whether response is desired movie page
re_movie_url = re.compile(r'/title/[\d\w]+/',re.M|re.U|re.I)
response_url = response.geturl()
if re_movie_url.search(response_url):
return response_url
# then check whether there is single link to movie in the loaded page
response_str = response.read()
links_list = []
for link in re_movie_url.findall(response_str):
if link not in links_list: links_list.append(link)
if len(links_list) == 1: return imdb_url_str + links_list[0]
# then check 1st link to movie in response page
if links_list and check_imdb_movie_year(links_list[0],data_list[2]):
return imdb_url_str + links_list[0]
# finally try to find movie in exact matches table
if response_str.find('Titles (Exact Matches)') != -1:
table_str = get_between(response_str,'<table>','</table>',response_str.find('Titles (Exact Matches)'))
for row_match in re.finditer(r'<tr>.*?\((?P<year>\d{4})\).*?</tr>',table_str,flags = re.I|re.M|re.U|re.S):
if int(data_list[2]) - 2 <= int(row_match.group('year')) <= int(data_list[2]) + 2:
return imdb_url_str + get_between(row_match.group(0),'href="','"')
def get_movie_url(movie_data):
url = 'http://www.deanclatworthy.com/imdb/'
params_dict = {}
data_list = movie_data.split('*')
if len(data_list) < 3:
raise ValueError('invalid movie data')
# first trying to find movie using api
params_dict['q'] = data_list[1] if data_list[1] else romanize(data_list[0])
params_dict['year'] = data_list[2]
try:
# first trying to find movie using api
response = urllib.urlopen(url + '?' + urllib.urlencode(params_dict))
response_dict = json.loads(response.read())
if 'imdburl' in response_dict:
return response_dict['imdburl']
# then trying own function
movie_url = imdb_find_movie2.get_movie_url(movie_data)
if movie_url:
return movie_url
except:
pass
return None