User table 생성

README.md

@@ -88,14 +88,21 @@ source env/bin/activate
 pip3 install -r requirements.txt
 ```
 
-### 2. Slack Token 환경 변수 지정
-- 해당 슬랙 workspace의 API를 접근할 수 있는 토큰을 받아와야 합니다.
-- [이곳](https://api.slack.com/legacy/custom-integrations/legacy-tokens)에서 슬랙 토큰을 생성받습니다.
-- 생성 받은 토큰을 `.bash_profile`, `.zshrc` 등의 파일에 다음과 같이 환경 변수로 저장해줍니다.
+### 2. Slack API 
+1. workspace admin 권한을 받습니다.
+2. 새로운 글또 workspace 에 slack app 을 install 합니다. 
+
+    [Slack App 페이지](https://api.slack.com/apps) -> [Create New App]
+3. Your apps -> 글또 앱 -> OAuth & Permissions 탭에서 생성된 API 토큰을 확인할 수 있습니다.
+4. 해당 토큰을 `.bash_profile`, `.zshrc` 등의 파일에 다음과 같이 환경 변수로 저장해줍니다.
 ```
 export SLACK_TOKEN='xoxo-your-token'
 ```
+* Collaborator 권한이 있어야 app 관리 대시보드로 접근이 가능하기 때문에 Collaborators
+ 탭에서 관련 담당자분들을 미리 추가해주시면 좋습니다.
+
 
+
 
 ### 3. 마감 날짜 데이터, 유저 데이터 저장
 - 코드 실행 시 활용되는 유저 데이터와 마감 날짜 데이터를 다음과 같이 `outputs` 디렉토리 아래에 저장해주세요.

common/extract_data.py

@@ -2,11 +2,13 @@
 import pandas as pd
 from datetime import datetime, timedelta
 from utils import bigquery_config, phase, root_path, cardinal
+from google.oauth2 import service_account
 
 _project_id = bigquery_config[phase]['project']
 _suffix = bigquery_config[phase]['suffix']
 _jwt = os.path.join(root_path, 'config', bigquery_config[phase]['jwt'])
 
+_credentials = service_account.Credentials.from_service_account_file(os.getenv('GOOGLE_APPLICATION_CREDENTIALS'))
 
 
 def get_deadline_data(abs_output_directory):
@@ -122,3 +124,13 @@ def send_data_to_gbq(dataz, phase, project_id, log_table_id, status_table_id, pr
     status_df.to_gbq(status_table_id, project_id=project_id, if_exists='replace')
     if phase == 'production':
         status_df.to_gbq(prod_status_table_id, project_id=project_id, if_exists=if_exists_prod)
+
+
+def write_user_table(table_name, user_table):
+    user_data_frame = pd.DataFrame(data=user_table)
+    user_data_frame.to_gbq(table_name, if_exists='replace', credentials=_credentials)
+
+
+def read_user_table(table_name):
+    query = 'select user_id, user_name, channel_id, channel_name from {}'.format(table_name)
+    print(pd.read_gbq(query=query, credentials=_credentials))

common/slack_export.py

@@ -4,11 +4,13 @@
 from datetime import datetime
 from pick import pick
 from time import sleep
+from collections import defaultdict
 
 
 user_names_by_id = {}
 user_ids_by_name = {}
 dry_run = None
+user_name_key = 'user_name'
 
 
 def get_history(pageable_object, channel_id, page_size=100):
@@ -26,8 +28,8 @@ def get_history(pageable_object, channel_id, page_size=100):
         messages.extend(response['messages'])
 
         if response['has_more']:
-            last_timestamp = messages[-1]['ts'] # -1 means last element in a list
-            sleep(1) # Respect the Slack API rate limit
+            last_timestamp = messages[-1]['ts']  # -1 means last element in a list
+            sleep(1)  # Respect the Slack API rate limit
         else:
             break
     return messages
@@ -47,7 +49,7 @@ def parse_time_stamp(time_stamp):
         if len(t_list) != 2:
             raise ValueError('Invalid time stamp')
         else:
-            return datetime.utcfromtimestamp( float(t_list[0]))
+            return datetime.utcfromtimestamp(float(t_list[0]))
 
 
 def channel_rename(old_room_name, new_room_name):
@@ -59,7 +61,7 @@ def channel_rename(old_room_name, new_room_name):
         return
     mkdir(new_room_name)
     for file_name in os.listdir(old_room_name):
-        shutil.move( os.path.join(old_room_name, file_name), new_room_name)
+        shutil.move(os.path.join(old_room_name, file_name), new_room_name)
     os.rmdir(old_room_name)
 
 
@@ -102,8 +104,8 @@ def parse_messages(room_dir, messages, room_type):
             channel_rename(old_room_path, new_room_path)
 
         current_messages.append(message)
-    out_file_name = '{room}/{file}.json'.format(room=room_dir, file=current_file_date )
-    write_message_file( out_file_name, current_messages)
+    out_file_name = '{room}/{file}.json'.format(room=room_dir, file=current_file_date)
+    write_message_file(out_file_name, current_messages)
 
 
 def filter_conversations_by_name(channels_or_groups, channel_or_group_names):
@@ -166,7 +168,7 @@ def dump_user_file(users):
     write to user file, any existing file needs to be overwritten.
     """
     with open(os.path.join('../', 'users.json'), 'w') as userFile:
-        json.dump( users, userFile, indent=4 )
+        json.dump(users, userFile, indent=4)
 
 
 def bootstrap_key_values(slack_obj, _dry_run):
@@ -216,3 +218,54 @@ def finalize(zip_name, output_directory):
     if zip_name:
         shutil.make_archive(zip_name, 'zip', output_directory, None)
         shutil.rmtree(output_directory)
+
+
+# channel_prefix : prefix of the posting channel
+def get_user_table(slacker, channel_prefix):
+    users = slacker.users.list().body['members']
+    channels = slacker.channels.list().body['channels']
+
+    user_names_with_id = get_user_names_with_id(users)
+    user_names_with_channel = get_user_names_with_channel(channels, channel_prefix)
+
+    if len(user_names_with_id) == len(user_names_with_channel):
+        user_table = defaultdict(dict)
+
+        for merged_list in (user_names_with_id, user_names_with_channel):
+            for element in merged_list:
+                user_table[element[user_name_key]].update(element)
+
+        # user_table : key = '홍길동', value = {'name': '홍길동', 'id': 'UTHXXXXX', 'channel_name': 'prefix_직군_게시글'} 인 dict.
+        return user_table.values()
+    else:
+        print("Fail to get user table. cause, length do not match.")
+
+
+def get_user_names_with_channel(channels, channel_prefix):
+    users_with_channel = []
+
+    for channel in channels:
+        channel_id = channel['id']
+        channel_name = channel['name']
+
+        if channel_name.startswith(channel_prefix):
+            member_names = channel['topic']['value'].split()
+
+            for name in member_names:
+                users_with_channel.append({user_name_key: name, 'channel_id': channel_id, 'channel_name': channel_name})
+
+    return users_with_channel
+
+
+def get_user_names_with_id(users):
+    users_with_id = []
+
+    for user in users:
+        if is_valid_user(user):
+            users_with_id.append({'user_id': user['id'], user_name_key: user['real_name']})
+
+    return users_with_id
+
+
+def is_valid_user(user):
+    return not user['is_bot'] and not user['deleted'] and user['real_name'] != 'Slackbot'