def insert_actors_info():
""" Check if actor has username.
If so saves id and username on youtube.csv.
Returns:
YoutubeAPI object if successful, None otherwise.
"""
yt_api = YoutubeAPI()
yt_api.generate_folder()
check = yt_api.generate_csv(clean=True)
if check:
with open('config/actors.json') as json_file:
data = json.load(json_file)
for item in data['channels']:
if 'username' in item:
if item['username']:
result = yt_api.get_channel_info_by_username(
item['username'])
item['id'] = result['items'][0]['id']
yt_api.insert_value(column='channel_id',
value=item['id'],
search_cell='actor',
search_value=item['actor'].replace(
'\n', ''))
yt_api.insert_value(column='username',
value=item['username'],
search_cell='actor',
search_value=item['actor'].replace(
'\n', ''))
return yt_api
return None
def test_insert_multiple_values_of_an_actor(self):
yt_api = YoutubeAPI()
insert_actors_info()
result = yt_api.insert_multiple_values(column=['channel_id'],
search_cell='actor',
search_value='Frente Brasil' +
' Popular')
result = yt_api.get_row(column='actor', value='Frente Brasil Popular')
self.assertEqual(result, {'username': '', 'channel_id': 'null'})
def test_insert_value_of_an_actor(self):
yt_api = YoutubeAPI()
insert_actors_info()
result = yt_api.insert_value(column='channel_id',
value='UCX2Aanu4fGewmhP4rf5GQ3Q',
search_cell='actor',
search_value='Frente Brasil Popular')
result = yt_api.get_row(column='actor', value='Frente Brasil Popular')
self.assertEqual(result, {
'username': '',
'channel_id': 'UCX2Aanu4fGewmhP4rf5GQ3Q'
})
self.clean_csv(yt_api)
self.clean_folder(yt_api)
def __init__(self):
with open('config/parameters.json') as data_file:
parameters = json.load(data_file)['parameters']
self.user = YoutubeAPI()
self._activities = {
'part': 'snippet,contentDetails',
'channelId': '',
'maxResults': '',
'publishedAfter': parameters['initial_date'],
'key': self.user._youtube_key
}
self._activities_extent = {
'part': 'snippet,contentDetails',
'channelId': '',
'maxResults': '',
'publishedAfter': '2018-01-01T00:00:01.45-03:00',
'publishedBefore': '',
'key': self.user._youtube_key
}
self._videos = {
'part': 'snippet,statistics,status,contentDetails',
'id': '',
'maxResults': '',
'key': self.user._youtube_key
}
self._search = {
'part': 'id,snippet',
'maxResults': '',
'type': '',
'relatedToVideoId': '',
'key': self.user._youtube_key
}
self._category = {
'part': 'id,snippet',
'hl': 'pt_BR',
'id': '',
'key': self.user._youtube_key
}
def test_clean_csv(self):
yt_api = YoutubeAPI()
self.assertTrue(yt_api.generate_csv(clean=True))
self.insert_value(yt_api)
result = yt_api.get_row(column='actor', value='Frente Brasil Popular')
self.assertEqual(result, {
'username': '',
'channel_id': 'UCX2Aanu4fGewmhP4rf5GQ3Q'
})
self.assertTrue(yt_api.generate_csv(clean=True))
result = yt_api.get_row(column='actor', value='Frente Brasil Popular')
self.assertEqual(result, {'username': '', 'channel_id': ''})
def setUp(self):
self._video = Videos()
self._user = YoutubeAPI()
class TestVideos(unittest.TestCase):
def setUp(self):
self._video = Videos()
self._user = YoutubeAPI()
def test_has_activities(self):
channel_id = 'UCj34AOIMl_k1fF7hcBkD_dw'
max_results = '5'
result = self._video.get_activity_info(channel_id, max_results)
self.assertTrue(result['items'])
def test_category_info(self):
id = '28'
result = self._video.get_category_info(id)
self.assertEqual(result['items'][0]['snippet']['title'],
'Ciência e tecnologia')
def test_has_search(self):
related_to_video_id = 'Xx7bjU7gabM'
max_results = '5'
type = 'video'
result = self._video.get_search_info(max_results, related_to_video_id,
type)
self.assertTrue(result['items'])
def test_has_video(self):
channel_id = 'V6OvM-0SGUU'
max_results = '5'
result = self._video.get_videos_info(channel_id, max_results)
self.assertTrue(result['items'])
def test_all_video_ids(self):
user_id = 'UC9uefWa6TXIPDRWGZYMcTuA'
max_results = '7'
result = self._user.get_channel_info(user_id)
channel_id = self._user.get_channel_id(result)
result_activities = self._video.get_activity_info(
channel_id, max_results)
video_ids = self._video.get_all_video_ids(result_activities)
assert_list = ['FvNsjk62Xi8']
self.assertEqual(video_ids[-1:], assert_list)
def test_all_videos_count_user_id(self):
user_id = 'UCsCI7wlAwbzTPK55yVaX2Ig'
result = self._user.get_channel_info(user_id)
max_results = 50
video_views = self._video.get_all_video_views_user_id(
result, max_results, 0)
video_title = video_views[-1]['title']
self.assertEqual(video_title, 'Encontro Presencial - IAEaD 2018')
video_view = int(video_views[-1]['views'])
video_url = str(video_views[-1]['url'])
video_likes = str(video_views[-1]['likes'])
video_dislikes = str(video_views[-1]['dislikes'])
video_comments = str(video_views[-1]['comments'])
video_favorites = int(video_views[-1]['favorites'])
self.assertGreater(video_view, 16)
self.assertLess(video_view, 100)
self.assertEqual(video_url,
'https://www.youtube.com/watch?v=B-5VHSKIuHw')
self.assertEqual(video_likes, '1')
self.assertEqual(video_dislikes, '0')
self.assertEqual(video_comments, '0')
self.assertGreaterEqual(video_favorites, 0)
class Videos:
"""Gets all video's informations.
Following certain date gets video's id, views and title.
"""
def __init__(self):
with open('config/parameters.json') as data_file:
parameters = json.load(data_file)['parameters']
self.user = YoutubeAPI()
self._activities = {
'part': 'snippet,contentDetails',
'channelId': '',
'maxResults': '',
'publishedAfter': parameters['initial_date'],
'key': self.user._youtube_key
}
self._activities_extent = {
'part': 'snippet,contentDetails',
'channelId': '',
'maxResults': '',
'publishedAfter': '2018-01-01T00:00:01.45-03:00',
'publishedBefore': '',
'key': self.user._youtube_key
}
self._videos = {
'part': 'snippet,statistics,status,contentDetails',
'id': '',
'maxResults': '',
'key': self.user._youtube_key
}
self._search = {
'part': 'id,snippet',
'maxResults': '',
'type': '',
'relatedToVideoId': '',
'key': self.user._youtube_key
}
self._category = {
'part': 'id,snippet',
'hl': 'pt_BR',
'id': '',
'key': self.user._youtube_key
}
def get_category_info(self, id):
self._category['id'] = id
return requests.get(CATEGORY_URL, params=self._category).json()
def get_search_info(self, max_results, related_to_video_id, type):
self._search['maxResults'] = max_results
self._search['relatedToVideoId'] = related_to_video_id
self._search['type'] = type
return requests.get(SEARCH_URL, params=self._search).json()
def get_activity_info(self, channel_id, max_results):
self._activities['maxResults'] = max_results
self._activities['channelId'] = channel_id
self._activities_extent['maxResults'] = max_results
self._activities_extent['channelId'] = channel_id
request = requests.get(ACTIVITIES_URL, params=self._activities).json()
if ('items' in request):
length = len(request['items'])
while (length >= 50):
last_date = request['items'][-1]['snippet']['publishedAt']
self._activities_extent['publishedBefore'] = last_date
request_extent = requests.get(ACTIVITIES_URL,
params=self._activities_extent)
request_extent = request_extent.json()
if ('items' in request_extent):
length = len(request_extent['items'])
request['items'].extend(request_extent['items'])
else:
length = 0
self._activities_extent['publishedBefore'] = ''
return request
def get_videos_info(self, video_id, max_results):
self._videos['maxResults'] = max_results
self._videos['id'] = video_id
return requests.get(VIDEOS_URL, params=self._videos).json()
def get_all_video_ids(self, response):
video_ids = []
for video in response['items']:
try:
video_ids.append(video['contentDetails']['upload']['videoId'])
except KeyError:
pass
return video_ids
def get_all_video_items(self, response, max_results,
related_videos_max_results, get_related_videos):
videos_dic = []
for item in response:
views = self.get_videos_info(item, max_results)
if get_related_videos:
search = self.get_search_info(related_videos_max_results, item,
'video')
id = views['items'][0]['id']
channel_id = views['items'][0]['snippet']['channelId']
if 'viewCount' in views['items'][0]['statistics']:
video_views = views['items'][0]['statistics']['viewCount']
else:
video_views = 'disabled'
if 'likeCount' in views['items'][0]['statistics']:
video_likes = views['items'][0]['statistics']['likeCount']
else:
video_likes = 'disabled'
if 'dislikeCount' in views['items'][0]['statistics']:
video_dislikes = \
views['items'][0]['statistics']['dislikeCount']
else:
video_dislikes = 'disabled'
if 'commentCount' in views['items'][0]['statistics']:
video_comments = \
views['items'][0]['statistics']['commentCount']
else:
video_comments = 'disabled'
if 'favoriteCount' in views['items'][0]['statistics']:
video_favorites = \
views['items'][0]['statistics']['favoriteCount']
else:
video_favorites = 'disabled'
if 'publishedAt' in views['items'][0]['snippet']:
video_published_date = \
views['items'][0]['snippet']['publishedAt']
else:
video_published_date = 'disabled'
if 'description' in views['items'][0]['snippet']:
video_description = \
views['items'][0]['snippet']['description']
else:
video_description = 'disabled'
if 'tags' in views['items'][0]['snippet']:
video_tags = \
views['items'][0]['snippet']['tags']
else:
video_tags = 'disabled'
if 'embeddable' in views['items'][0]['status']:
video_embeddable = \
views['items'][0]['status']['embeddable']
else:
video_embeddable = 'disabled'
if 'duration' in views['items'][0]['contentDetails']:
video_duration = \
views['items'][0]['contentDetails']['duration']
else:
video_duration = 'disabled'
if 'thumbnails' in views['items'][0]['snippet']:
video_thumbnail = \
views['items'][0]['snippet']['thumbnails']['high']['url']
else:
video_thumbnail = 'disabled'
if 'categoryId' in views['items'][0]['snippet']:
category_id = views['items'][0]['snippet']['categoryId']
category_info = self.get_category_info(category_id)
video_category = category_info['items'][0]['snippet']['title']
else:
video_category = 'disabled'
video_titles = views['items'][0]['snippet']['title']
video_url = VIDEOS_BASE_URL + views['items'][0]['id']
if get_related_videos:
if search['items']:
related_to_video = []
for video in search['items']:
related_to_video.append(
self.get_all_video_items([video['id']['videoId']],
1, 0, False)[0])
related_to_video[-1]['original_id'] = item
else:
related_to_video = 'disabled'
videos_dic.append({
'video_id': id,
'channel_id': channel_id,
'title': video_titles,
'views': video_views,
'likes': video_likes,
'dislikes': video_dislikes,
'comments': video_comments,
'favorites': video_favorites,
'url': video_url,
'publishedAt': video_published_date,
'description': video_description,
'tags': video_tags,
'embeddable': video_embeddable,
'duration': video_duration,
'thumbnail': video_thumbnail,
'category': video_category
})
if get_related_videos:
videos_dic[-1]['related_to_video'] = related_to_video
return videos_dic
def get_all_video_views_user_id(self, response, max_results,
related_videos_max_results):
channel_id = self.user.get_channel_id(response)
result_activities = self.get_activity_info(channel_id, max_results)
videos_id = self.get_all_video_ids(result_activities)
video_views = self.get_all_video_items(videos_id, max_results,
related_videos_max_results,
True)
return video_views
def test_check_data_of_an_actor(self):
yt_api = insert_actors_info()
result = YoutubeAPI().get_row(column='actor', value='Bancada Ativista')
self.assertEqual(result, {'username': '', 'channel_id': ''})
def test_generate_youtube_folder(self):
YoutubeAPI().generate_folder()
self.assertTrue(os.path.exists(self._foldername))
def test_generate_youtube_csv(self):
csv = YoutubeAPI().generate_csv()
self.assertTrue(os.path.isfile(self._filename))