def store_movies(year):
print(f'storing-----------{year}')
with open(f'./data/movie_{year}.json', 'r') as fp:
movies = json.load(fp)
# delete stored movies
movies_year = Movie.objects.filter(release_date__year=year)
movies_year.delete()
# using bulk
to_db_movies = []
for mv in tqdm(movies):
movie = Movie()
movie.pk = mv['id']
movie.title_en = mv['original_title']
movie.title_ko = mv['title']
movie.rate = mv['vote_average']
movie.rate_people_count = mv['vote_count']
movie.poster_path = mv['poster_path']
movie.description = mv['overview']
movie.release_date = datetime.datetime.strptime(
mv['release_date'], '%Y-%m-%d')
to_db_movies.append(movie)
div = 100
to_db_movies_batch = []
for i in range(len(to_db_movies) // div + 1):
to_db_movies_batch.append(to_db_movies[i * div:(i + 1) * div])
for idx, batch in enumerate(to_db_movies_batch):
print(round(idx / len(to_db_movies_batch), 2), end='|')
temp = Movie.objects.bulk_create(batch)
# add genre
mgs = []
for mv in movies:
for gid in mv['genre_ids']:
mg = MG()
mg.movie_id = mv['id']
mg.genre_id = gid
mgs.append(mg)
temp = MG.objects.bulk_create(mgs)
def update_dataset(request):
movietitles = []
# 장르 항목 요청
response_gens = requests.get(GENRE_URL)
response_gens = response_gens.json()
print(GENRE_URL)
# print(response_gens, '####')
for gen in response_gens['genres']:
print(gen)
genre_instance = Genre()
genre_instance.genre_id = gen['id']
genre_instance.name = gen['name']
genre_instance.save()
# 요청 보내기 (1, 25페이지 까지)
for page in range(1, 25):
REQ_URL = MOVIE_URL + f'&page={page}'
print(REQ_URL)
response = requests.get(REQ_URL)
response = response.json()
# response에서 20개의 자료 추출하기
for num in range(0, 20):
# movie 인스턴스 호출하고 DB에 저장하기
movie = Movie()
movie.movie_id = response['results'][num]['id']
movie.popularity = response['results'][num]['popularity']
movie.vote_count = response['results'][num]['vote_count']
movie.poster_path = response['results'][num]['poster_path']
movie.backdrop_path = response['results'][num]['backdrop_path']
movie.original_language = response['results'][num][
'original_language']
movie.original_title = response['results'][num]['original_title']
movie.title = response['results'][num]['title']
movie.vote_average = response['results'][num]['vote_average']
movie.overview = response['results'][num]['overview']
movie.release_date = response['results'][num]['release_date']
# # movie detail api요청 및 저장(오래걸림)
# movie_id = movie.movie_id
# DETAIL_URL = BASE_URL + f'movie/{movie_id}?' + API_KEY + '&language=ko-KR'
# detail_res = requests.get(DETAIL_URL).json()
# # runtime 데이터 저장
# movie.runtime = detail_res['runtime']
movie.save()
# # production country data DB에 저장
# for country_data in detail_res['production_countries']:
# country = Country()
# country.iso = country_data['iso_3166_1']
# country.name = country_data['name']
# country.save()
# M2M 관계 중계테이블에 저장(영화-국가)
# for country_data in detail_res['production_countries']:
# iso = country_data['iso_3166_1']
# country = Country.objects.get(iso=iso)
# country.movies.add(movie)
# M2M 관계 중계테이블에 저장(영화-장르)
for genid in response['results'][num]['genre_ids']:
# print(genid)
genre = Genre.objects.get(genre_id=genid)
genre.movies.add(movie)
# 제대로 받아지는지 확인한다.
print(movie.title)
# movietitles에 담아서 template에 출력해보자.
movietitles.append(movie.title)
# 완료시 success문구 출력
print('success')
context = {
'movietitles': movietitles,
}
return render(request, 'make_dataset/update_dataset.html', context)
