tinybot

This is a small and simple library for quickly building Telegram Bots with Python.

It is made to be dynamic and independent of the Telegram Bot API. It means that whenever a new Telegram API method or callback is added, there is no need to update this library, all calls are dynamic and you can send whatever you like to Telegram.

Installation

It is published on PyPI so to use it simply install it with pip.

$ pip install tinybot

Example

Actually working little bot (from which this library was originated) can be found here.

Here is a simple example:

import tinybot


class Assistant(tinybot.Bot):
    version = '0.0.1-SNAPSHOT'
    token = '123456789:token-for-debugging' # token is not required when using tinybot.run

    def handle_message(self, data, api):
        if '@all' in data.text:
            api.sendMessage(chat_id=data.chat.id, text='@necauqua @other_people..')
            api.sendMessage(chat_id=data.chat.id, text='[debug] update id ' + str(self.update_id))


if __name__ == '__main__':
    # just start longpoll loop for debugging
    Assistant.launch_longpoll(30)

    # tinybot.run is a ready-to-use CLI for launching longpoll or webhook and providing the token
    # tinybot.run(Assistant, 'necauqua assistant bot')

Usage

Your bot is a class which derives from tinybot.Bot class. You can start it with a ready-to-use CLI using this standard construct

if __name__ == '__main__':
    tinybot.run(bot_cls, bot_description)

All of the bot's logic is defined inside the class deriven from tinybot.Bot.

It have static fields name, version,full_name and token.

• name defaults to the class name and is the name of your bot. Used in full_name
• version is a version of your bot, it defaults to 0.1.0. Used in full_name.
• full_name defaults to name/version. It is used as a User-Session header when making requests, and as Server header when responding to Telegram from a webhook server.
• token is an optional field, because usually it is set from command line when running tinybot.run. However, it might be useful for debugging. It is obviously the token which is used for all requests.

Your class should define methods with signarure handle_xxx(self, data, api) where xxx is one of the update types found here.

For example:

def handle_message(self, message, api): pass
# or
def handle_channel_post(self, post, api): pass

• self parameter can contain some additional data, for now it only contains the update_id field.
• data parameter is a dict/list-like recursive structure made with __getattr__'s and __getitems__'s in which the received Update JSON object is given. When you try to get a nonexistent field, NoSuchElementException is raised and then handled properly by the launch_xxx methods caller.
• api parameter worls similarly, any method call which looks like api.anyMethodCall(**kwargs) would send appropriate request to the Telegram Bot API.
  When you do such a request, Telegram returns JSON in format
  {"ok": true, "update": ..data..}
  or
  {"ok": false, "description": "..error message.."}.
  This JSON is unwrapped and if ok is false or non-existent, RequestException is raised with the description and similarly handled by the launch_xxx methods caller. Else, the "update" object is wrapped in the same structure as the data parameter and is returned from the api method call.

All of the above means that if you want your bot to do multiple get-and-then-send requests, you'll want to arrange your code to first get all the data and only after that send your requests. That way, if any of the data weren't available, any of your mutating requests would not be sent.

License

It is licensed under permissive MIT license which means you can use this code in whatever way possible, as long as you include the LICENSE file (by which you mention my authorship). It is included in the package so you have to do nothing yourself when simply using this library.