def verify_friends_dictionary(): friends_dictionary = load_friends_dictionary() crawled_set = set(friends_dictionary.keys()) users_set = set(unique_users.user_id) need_to_crawl = users_set - crawled_set dump_needcrawl_set(need_to_crawl) print('Number of users still need to crawl: {}'.format(len(need_to_crawl))) unwanted = set(crawled_set) - set(users_set) for unwanted_key in unwanted: del friends_dictionary[unwanted_key] dump_friends_dictionary(friends_dictionary)
def merge_new_friends_dictionary(): friends_dictionary = load_friends_dictionary() newcrawl_dictionary = load_newcrawl_dictionary() dump_friends_dictionary({**friends_dictionary, **newcrawl_dictionary})
# # Load Friends, and add new crawl relationships if necessary # In[6]: def merge_new_friends_dictionary(): friends_dictionary = load_friends_dictionary() newcrawl_dictionary = load_newcrawl_dictionary() dump_friends_dictionary({**friends_dictionary, **newcrawl_dictionary}) if calculate_friends: merge_new_friends_dictionary() friends_dictionary = load_friends_dictionary() # # Unique Users # In[7]: def find_by_user_name(df, user_name): user = df[df.user == user_name] return user.iloc[0] def find_by_user_id(df, user_id): user = df[df.user_id == user_id] return user.iloc[0]