mirror of
https://github.com/MarshalX/telegram-crawler.git
synced 2024-11-23 07:49:20 +01:00
184 lines
6.3 KiB
Python
184 lines
6.3 KiB
Python
import asyncio
|
|
import logging
|
|
import os
|
|
import re
|
|
from typing import Tuple
|
|
|
|
import aiohttp
|
|
|
|
COMMIT_SHA = os.environ['COMMIT_SHA']
|
|
|
|
# commits for test alert builder
|
|
# COMMIT_SHA = '4015bd9c48b45910727569fff5e770000d85d207' # all clients + server and test server + web
|
|
# COMMIT_SHA = '9cc3f0fb7c390c8cb8b789e9377f10ed5e80a089' # web and web res together
|
|
# COMMIT_SHA = '4efaf918af43054ba3ff76068e83d135a9a2535d' # web
|
|
# COMMIT_SHA = 'e2d725c2b3813d7c170f50b0ab21424a71466f6d' # web res
|
|
|
|
TELEGRAM_BOT_TOKEN = os.environ['TELEGRAM_BOT_TOKEN']
|
|
GITHUB_PAT = os.environ['GITHUB_PAT']
|
|
|
|
REPOSITORY = os.environ.get('REPOSITORY', 'MarshalX/telegram-crawler')
|
|
CHAT_ID = os.environ.get('CHAT_ID', '@tgcrawl')
|
|
ROOT_TREE_DIR = os.environ.get('ROOT_TREE_DIR', 'data')
|
|
|
|
BASE_GITHUB_API = 'https://api.github.com/'
|
|
GITHUB_LAST_COMMITS = 'repos/{repo}/commits/{sha}'
|
|
|
|
BASE_TELEGRAM_API = 'https://api.telegram.org/bot{token}/'
|
|
TELEGRAM_SEND_MESSAGE = 'sendMessage'
|
|
|
|
logger = logging.getLogger(__name__)
|
|
logging.basicConfig(level=logging.INFO)
|
|
|
|
STATUS_TO_EMOJI = {
|
|
'added': '✅',
|
|
'modified': '📝',
|
|
'removed': '❌',
|
|
'renamed': '🔄',
|
|
'copied': '📋',
|
|
'changed': '📝',
|
|
'unchanged': '📝',
|
|
}
|
|
|
|
AVAILABLE_HASHTAGS = {
|
|
'web_tr', 'web_res', 'web', 'server', 'test_server', 'client', 'ios', 'macos', 'android'
|
|
}
|
|
HASHTAGS_PATTERNS = {
|
|
# regex will be more flexible. for example, in issue with double hashtag '#web #web_res' when data/res not changed
|
|
'web_tr': os.path.join(ROOT_TREE_DIR, 'web_tr'),
|
|
'web_res': os.path.join(ROOT_TREE_DIR, 'web_res'),
|
|
'web': os.path.join(ROOT_TREE_DIR, 'web'),
|
|
'server': os.path.join(ROOT_TREE_DIR, 'server'),
|
|
'test_server': os.path.join(ROOT_TREE_DIR, 'server', 'test'),
|
|
'client': os.path.join(ROOT_TREE_DIR, 'client'),
|
|
'ios': os.path.join(ROOT_TREE_DIR, 'client', 'ios-beta'),
|
|
'macos': os.path.join(ROOT_TREE_DIR, 'client', 'macos-beta'),
|
|
'android': os.path.join(ROOT_TREE_DIR, 'client', 'android-beta'),
|
|
}
|
|
# order is important!
|
|
PATHS_TO_REMOVE_FROM_ALERT = [
|
|
os.path.join(ROOT_TREE_DIR, 'web_tr'),
|
|
os.path.join(ROOT_TREE_DIR, 'web_res'),
|
|
os.path.join(ROOT_TREE_DIR, 'web'),
|
|
os.path.join(ROOT_TREE_DIR, 'server'),
|
|
os.path.join(ROOT_TREE_DIR, 'client'),
|
|
]
|
|
|
|
GITHUB_API_LIMIT_PER_HOUR = 5_000
|
|
COUNT_OF_RUNNING_WORKFLOW_AT_SAME_TIME = 5 # just random number ;d
|
|
|
|
ROW_PER_STATUS = 5
|
|
|
|
LAST_PAGE_NUMBER_REGEX = r'page=(\d+)>; rel="last"'
|
|
|
|
|
|
async def send_req_until_success(session: aiohttp.ClientSession, **kwargs) -> Tuple[dict, int]:
|
|
delay = 5 # in sec
|
|
count_of_retries = int(GITHUB_API_LIMIT_PER_HOUR / COUNT_OF_RUNNING_WORKFLOW_AT_SAME_TIME / delay)
|
|
|
|
last_page_number = 1
|
|
retry_number = 1
|
|
while retry_number <= count_of_retries:
|
|
retry_number += 1
|
|
|
|
res = await session.get(**kwargs)
|
|
if res.status != 200:
|
|
await asyncio.sleep(delay)
|
|
continue
|
|
|
|
json = await res.json()
|
|
|
|
pagination_data = res.headers.get('Link', '')
|
|
matches = re.findall(LAST_PAGE_NUMBER_REGEX, pagination_data)
|
|
if matches:
|
|
last_page_number = int(matches[0])
|
|
|
|
return json, last_page_number
|
|
|
|
raise RuntimeError('Surprise. Time is over')
|
|
|
|
|
|
async def main() -> None:
|
|
async with aiohttp.ClientSession() as session:
|
|
commit_data, last_page = await send_req_until_success(
|
|
session=session,
|
|
url=f'{BASE_GITHUB_API}{GITHUB_LAST_COMMITS}'.format(repo=REPOSITORY, sha=COMMIT_SHA),
|
|
headers={
|
|
'Authorization': f'token {GITHUB_PAT}'
|
|
}
|
|
)
|
|
commit_files = commit_data['files']
|
|
|
|
coroutine_list = list()
|
|
for current_page in range(2, last_page + 1):
|
|
coroutine_list.append(send_req_until_success(
|
|
session=session,
|
|
url=f'{BASE_GITHUB_API}{GITHUB_LAST_COMMITS}?page={current_page}'.format(
|
|
repo=REPOSITORY, sha=COMMIT_SHA
|
|
),
|
|
headers={
|
|
'Authorization': f'token {GITHUB_PAT}'
|
|
}
|
|
))
|
|
|
|
paginated_responses = await asyncio.gather(*coroutine_list)
|
|
for json_response, _ in paginated_responses:
|
|
commit_files.extend(json_response['files'])
|
|
|
|
html_url = commit_data['html_url']
|
|
|
|
alert_text = f'<b>New changes of Telegram</b>\n\n'
|
|
alert_hashtags = set()
|
|
|
|
global AVAILABLE_HASHTAGS
|
|
available_hashtags = AVAILABLE_HASHTAGS.copy()
|
|
|
|
changes = {k: [] for k in STATUS_TO_EMOJI.keys()}
|
|
for file in commit_files:
|
|
for available_hashtag in available_hashtags:
|
|
pattern = HASHTAGS_PATTERNS[available_hashtag]
|
|
if pattern in file['filename']:
|
|
alert_hashtags.add(available_hashtag)
|
|
|
|
# optimize substring search
|
|
available_hashtags -= alert_hashtags
|
|
|
|
changed_url = file['filename'].replace('.html', '')
|
|
for path_to_remove in PATHS_TO_REMOVE_FROM_ALERT:
|
|
if changed_url.startswith(path_to_remove):
|
|
changed_url = changed_url[len(path_to_remove) + 1:]
|
|
break # can't occur more than one time
|
|
|
|
status = STATUS_TO_EMOJI[file['status']]
|
|
changes[file['status']].append(f'{status} <code>{changed_url}</code>')
|
|
|
|
for i, [status, text_list] in enumerate(changes.items()):
|
|
if not text_list:
|
|
continue
|
|
|
|
alert_text += '\n'.join(text_list[:ROW_PER_STATUS]) + '\n'
|
|
|
|
if len(text_list) > ROW_PER_STATUS:
|
|
count = len(text_list) - ROW_PER_STATUS
|
|
alert_text += f'And <b>{count}</b> {status} actions more..\n'
|
|
|
|
alert_text += '\n'
|
|
|
|
alert_text += f'<a href="{html_url}">View diff on GitHub...</a>'
|
|
if alert_hashtags:
|
|
alert_text += '\n\n' + ' '.join([f'#{hashtag}' for hashtag in sorted(alert_hashtags)])
|
|
|
|
logger.info(alert_text)
|
|
telegram_response = await session.get(
|
|
url=f'{BASE_TELEGRAM_API}{TELEGRAM_SEND_MESSAGE}'.format(token=TELEGRAM_BOT_TOKEN),
|
|
params={
|
|
'chat_id': CHAT_ID,
|
|
'parse_mode': 'HTML',
|
|
'text': alert_text,
|
|
}
|
|
)
|
|
logger.debug(await telegram_response.read())
|
|
|
|
|
|
if __name__ == '__main__':
|
|
asyncio.get_event_loop().run_until_complete(main())
|