tgexplore/marie-2.0-english

Marie-2.0-English

A modular telegram Python bot running on python3 with an sqlalchemy database.

Originally a simple group management bot with multiple admin features, it has evolved, becoming extremely modular and simple to use.

Can be found on telegram as Marie.

Marie and I are moderating a support group, where you can ask for help setting up your bot, discover/request new features, report bugs, and stay in the loop whenever a new update is available. Of course I'll also help when a database schema changes, and some table column needs to be modified/added. Note to maintainers that all schema changes will be found in the commit messages, and its their responsibility to read any new commits.

Join the news channel if you just want to stay in the loop about new features or announcements.

Alternatively, find me on telegram! (Keep all support questions in the support chat, where more people can help you.)

You can also tap the Deploy To Heroku button below to deploy straight to Heroku!

Starting the bot.

Once you've setup your database and your configuration (see below) is complete, simply run:

python3 -m tg_bot

Setting up the bot (Read this before trying to use!):

Please make sure to use python3.6, as I cannot guarantee everything will work as expected on older python versions! This is because markdown parsing is done by iterating through a dict, which are ordered by default in 3.6.

Configuration

There are two possible ways of configuring your bot: a config.py file, or ENV variables.

The prefered version is to use a config.py file, as it makes it easier to see all your settings grouped together. This file should be placed in your tg_bot folder, alongside the __main__.py file . This is where your bot token will be loaded from, as well as your database URI (if you're using a database), and most of your other settings.

It is recommended to import sample_config and extend the Config class, as this will ensure your config contains all defaults set in the sample_config, hence making it easier to upgrade.

An example config.py file could be:

from tg_bot.sample_config import Config


class Development(Config):
    OWNER_ID = 570400686  # my telegram ID
    OWNER_USERNAME = "Anandus"  # my telegram username
    API_KEY = "your bot api key"  # my api key, as provided by the botfather
    SQLALCHEMY_DATABASE_URI = 'postgresql://username:password@localhost:5432/database'  # sample db credentials
    MESSAGE_DUMP = '-1234567890' # some group chat that your bot is a member of
    USE_MESSAGE_DUMP = True
    SUDO_USERS = [18673980, 83489514]  # List of id's for users which have sudo access to the bot.
    LOAD = []
    NO_LOAD = ['translation']

If you can't have a config.py file (EG on heroku), it is also possible to use environment variables. The following env variables are supported:

• ENV: Setting this to ANYTHING will enable env variables

• TOKEN: Your bot token, as a string.

• OWNER_ID: An integer of consisting of your owner ID

• OWNER_USERNAME: Your username

• DATABASE_URL: Your database URL

• MESSAGE_DUMP: optional: a chat where your replied saved messages are stored, to stop people deleting their old

• LOAD: Space separated list of modules you would like to load

• NO_LOAD: Space separated list of modules you would like NOT to load

• WEBHOOK: Setting this to ANYTHING will enable webhooks when in env mode messages

• URL: The URL your webhook should connect to (only needed for webhook mode)

• SUDO_USERS: A space separated list of user_ids which should be considered sudo users

• SUPPORT_USERS: A space separated list of user_ids which should be considered support users (can gban/ungban, nothing else)

• WHITELIST_USERS: A space separated list of user_ids which should be considered whitelisted - they can't be banned.

• DONATION_LINK: Optional: link where you would like to receive donations.

• CERT_PATH: Path to your webhook certificate

• PORT: Port to use for your webhooks

• DEL_CMDS: Whether to delete commands from users which don't have rights to use that command

• STRICT_GBAN: Enforce gbans across new groups as well as old groups. When a gbanned user talks, he will be banned.

• WORKERS: Number of threads to use. 8 is the recommended (and default) amount, but your experience may vary. Note that going crazy with more threads wont necessarily speed up your bot, given the large amount of sql data accesses, and the way python asynchronous calls work.

• BAN_STICKER: Which sticker to use when banning people.

• ALLOW_EXCL: Whether to allow using exclamation marks ! for commands as well as /.

Python dependencies

Install the necessary python dependencies by moving to the project directory and running:

pip3 install -r requirements.txt.

This will install all necessary python packages.

Database

If you wish to use a database-dependent module (eg: locks, notes, userinfo, users, filters, welcomes), you'll need to have a database installed on your system. I use postgres, so I recommend using it for optimal compatibility.

In the case of postgres, this is how you would set up a the database on a debian/ubuntu system. Other distributions may vary.

• install postgresql:

sudo apt-get update && sudo apt-get install postgresql

• change to the postgres user:

sudo su - postgres

• create a new database user (change YOUR_USER appropriately):

createuser -P -s -e YOUR_USER

This will be followed by you needing to input your password.

• create a new database table:

createdb -O YOUR_USER YOUR_DB_NAME

Change YOUR_USER and YOUR_DB_NAME appropriately.

• finally:

psql YOUR_DB_NAME -h YOUR_HOST YOUR_USER

This will allow you to connect to your database via your terminal. By default, YOUR_HOST should be 0.0.0.0:5432.

You should now be able to build your database URI. This will be:

sqldbtype://username:pw@hostname:port/db_name

Replace sqldbtype with whichever db youre using (eg postgres, mysql, sqllite, etc) repeat for your username, password, hostname (localhost?), port (5432?), and db name.

Modules

Setting load order.

The module load order can be changed via the LOAD and NO_LOAD configuration settings. These should both represent lists.

If LOAD is an empty list, all modules in modules/ will be selected for loading by default.

If NO_LOAD is not present, or is an empty list, all modules selected for loading will be loaded.

If a module is in both LOAD and NO_LOAD, the module will not be loaded - NO_LOAD takes priority.

Creating your own modules.

Creating a module has been simplified as much as possible - but do not hesitate to suggest further simplification.

All that is needed is that your .py file be in the modules folder.

To add commands, make sure to import the dispatcher via

from tg_bot import dispatcher.

You can then add commands using the usual

dispatcher.add_handler().

Assigning the __help__ variable to a string describing this modules' available commands will allow the bot to load it and add the documentation for your module to the /help command. Setting the __mod_name__ variable will also allow you to use a nicer, user friendly name for a module.

The __migrate__() function is used for migrating chats - when a chat is upgraded to a supergroup, the ID changes, so it is necessary to migrate it in the db.

The __stats__() function is for retrieving module statistics, eg number of users, number of chats. This is accessed through the /stats command, which is only available to the bot owner.

marie-2.0-english's Issues

Problem with connect

There is a problem with connect....can't connect group to chat

TRANSLATOR

Hello, @TGExplore
As i noticed there is a issue on your repo.
As it holds CMD - /tr for correcting grammer
I think that's wrong as i noticed most of the people were reporting about translator
As i have suggestion... You Must have CMD - /gr for grammer correction &
you must leave /tr for translator when you get time.
Yeah, do not ment about hurting the work of Mr Paul & Marie it's a req of users to have it.

Thanks
MrHonekawa

Translate

import asyncioimport itertoolsimport loggingimport randomfrom collections import defaultdictfrom datetime import datetimefrom urllib.parse import quote, urljoin import aiohttpfrom aiocache import Cachefrom async_lru import alru_cache from i17obot import configfrom i17obot.models import Stringfrom i17obot.utils import seconds_until_tomorrow TRANSIFEX_API = { "python": "https://www.transifex.com/api/2/project/python-newest/", "jupyter": "https://www.transifex.com/api/2/project/jupyter-meta-documentation/",} PROJECT_URL = { "python": ( "https://www.transifex.com/" "python-doc/python-newest/translate/#{language}/{resource}/1" "?q={query_string}" ), "jupyter": ( "https://www.transifex.com/" "project-jupyter/jupyter-meta-documentation/translate/#{language}/{resource}/1" "?q={query_string}" ),} FILTER_RESOURCES_TO_BE_TRANSLATED = { "python": lambda r: r.split("--")[0] in ["bugs", "howto", "library"], "jupyter": None,} WEEK_IN_SECONDS = 604_800 logging.basicConfig(level=logging.INFO)logger = logging.getLogger() logger.info(config.CACHE_URL)cache = Cache.from_url(config.CACHE_URL) STRINGS_CACHE = defaultdict(dict) async def transifex_api(url, project, data=None, retrying=False, ttl=3600): url = urljoin(TRANSIFEX_API[project], url) if not data and (in_cache := await cache.get(url)): return in_cache if retrying: logger.debug("retrying url=%s", url) auth = aiohttp.BasicAuth(login="api", password=config.TRANSIFEX_TOKEN) async with aiohttp.ClientSession(auth=auth) as session: http_method = session.put if data else session.get kwargs = {"json": data} if data else {} try: async with http_method(url, **kwargs) as response: logger.debug("url=%s, status_code=%s", url, response.status) try: response = await response.json() if not data: await cache.set(url, response, ttl=3600) return response except aiohttp.ContentTypeError: response = await response.text() logger.warn("response=%r", response) return response except aiohttp.client_exceptions.ClientConnectorSSLError as e: logger.error("url=%s, error=%s", url, e) if not retrying: await asyncio.sleep(2) return await transifex_api(url, project, retrying=True) raise async def review_string(project, resource, language, translation, string_hash): return await transifex_api( f"resource/{resource}/translation/{language}/string/{string_hash}", project, data={"translation": translation, "reviewed": True}, ) async def random_resource(project): resources = await transifex_api(f"resources/", project, ttl=WEEK_IN_SECONDS) resources = [resource["slug"] for resource in resources] if FILTER_RESOURCES_TO_BE_TRANSLATED[project]: resources = filter(FILTER_RESOURCES_TO_BE_TRANSLATED[project], resources) resource = random.choice(list(resources)) logger.info("random_resource, resource=%s", resource) return resource async def strings_from_resource(resource, language, project): strings = await transifex_api( f"resource/{resource}/translation/{language}/strings/?details", project, ) logger.info( "getting strings from resource, resource=%s, strings_found=%s", resource, len(strings), ) return [ String.from_transifex( resource=resource, language=language, project=project, **string ) for string in strings ] async def random_string( language, project, resource=None, translated=None, reviewed=None, max_size=None): if not resource: resource = await random_resource(project) strings = await strings_from_resource(resource, language, project) if translated is not None: strings = filter(lambda s: bool(s.translation) == translated, strings) if reviewed is not None: strings = filter(lambda s: s.reviewed == reviewed, strings) if max_size is not None: strings = filter(lambda s: len(s.source) <= max_size, strings) strings = list(strings) if not strings: if max_size: max_size += 300 resource = None return await random_string( language, project, resource, translated, reviewed, max_size ) string = random.choice(strings) return string def transifex_string_url(resource, key, language, project): return PROJECT_URL[project].format( resource=resource, language=language, query_string=quote(f"text:'{key[:20]}'"), ) async def translate_string(user, string): await transifex_api( f"resource/{string.resource}/translation/{string.language}/string/{string.hash}/", string.project, data={"translation": string.translation, "user": user.transifex_username}, ) async def stats_from_resource(resource, language, project="python"): return await transifex_api(f"resource/{resource}/stats/{language}/", project,) async def download_all_strings(language): """ Download all strings in Transifex to JSON file """ today = datetime.today().strftime("%Y-%m-%d") resources = await transifex_api(f"resources/", "python") resources = [resource["slug"] for resource in resources] print("Resources", len(resources)) sema = asyncio.Semaphore(10) async with sema: strings = await asyncio.gather( *[strings_from_resource(resource, language) for resource in resources] ) strings = list(itertools.chain.from_iterable(strings)) print("Strings", len(resources)) return strings async def translation_stats(language): today = datetime.today() key = f"stats-{today.strftime('%Y-%m-%d')}" if (stats := await cache.get(key)) : return stats print(f"Statistics not cached for {language}") resources = await transifex_api(f"resources/", "python") resources = [resource["slug"] for resource in resources] print("Resources", len(resources)) sema = asyncio.Semaphore(1) async with sema: stats = await asyncio.gather( *[stats_from_resource(resource, language) for resource in resources] ) ttl = seconds_until_tomorrow(today) cache.set(key, stat, ttl=ttl) return stats

python update

hello! i have a issue like this python version is unavabile a new python version is avabile, sooo can you fix it please?

Myapkpure_Bot

problem with bans and kicks and mutes

Hello, there is a problem with the bot that every time a manager does not have permission to remove people from the group, it is enough that he has minimal management so the bot lets him do it as if he has the option to remove the people from the group, can you fix the glitch?