MTPyroger/docs/source/start/updates.rst

101 lines
3.1 KiB
ReStructuredText
Raw Normal View History

2019-05-12 17:26:55 +00:00
Handling Updates
================
2019-05-10 14:14:10 +00:00
2019-05-28 14:41:55 +00:00
Calling :doc:`API methods <invoking>` sequentially is cool, but how to react when, for example, a new message arrives?
This page deals with updates and how to handle such events in Pyrogram. Let's have a look at how they work.
2019-05-10 14:14:10 +00:00
2020-04-01 18:08:46 +00:00
.. contents:: Contents
:backlinks: none
:depth: 1
2020-04-01 18:08:46 +00:00
:local:
-----
2019-05-13 16:04:44 +00:00
Defining Updates
----------------
2019-05-12 17:26:55 +00:00
First, let's define what are these updates. As hinted already, updates are simply events that happen in your Telegram
2019-05-16 19:28:34 +00:00
account (incoming messages, new members join, bot button presses, etc...), which are meant to notify you about a new
2019-05-17 23:45:01 +00:00
specific state that has changed. These updates are handled by registering one or more callback functions in your app
2019-05-28 14:41:55 +00:00
using :doc:`Handlers <../api/handlers>`.
2019-05-10 14:14:10 +00:00
Each handler deals with a specific event and once a matching update arrives from Telegram, your registered callback
2019-05-17 23:45:01 +00:00
function will be called back by the framework and its body executed.
2019-05-10 14:14:10 +00:00
Registering a Handler
---------------------
2019-05-10 14:14:10 +00:00
To explain how handlers work let's examine the one which will be in charge for handling :class:`~pyrogram.types.Message`
updates coming from all around your chats. Every other handler shares the same setup logic; you should not have
troubles settings them up once you learn from this section.
2019-05-10 14:14:10 +00:00
Using Decorators
^^^^^^^^^^^^^^^^
2019-05-10 14:14:10 +00:00
The most elegant way to register a message handler is by using the :meth:`~pyrogram.Client.on_message` decorator:
2019-05-10 14:14:10 +00:00
.. code-block:: python
from pyrogram import Client
2019-05-10 14:14:10 +00:00
app = Client("my_account")
@app.on_message()
def my_handler(client, message):
message.forward("me")
2019-06-27 21:16:21 +00:00
2019-05-10 14:14:10 +00:00
app.run()
2019-05-10 14:14:10 +00:00
The defined function ``my_handler``, which accepts the two arguments *(client, message)*, will be the function that gets
executed every time a new message arrives.
2019-05-10 14:14:10 +00:00
Asynchronous handlers
^^^^^^^^^^^^^^^^^^^^^
2019-05-10 14:14:10 +00:00
You can also have asynchronous handlers; you only need to define the callback function using ``async def`` and call API
methods by placing ``await`` in front of them:
2019-05-10 14:14:10 +00:00
.. code-block:: python
2019-05-10 14:14:10 +00:00
@app.on_message()
async def my_handler(client, message):
await message.forward("me")
2019-05-10 14:14:10 +00:00
.. note::
2019-05-10 14:14:10 +00:00
You can mix ``def`` and ``async def`` handlers as much as you need, Pyrogram will still work concurrently and
efficiently regardless of what you choose.
2019-05-10 14:14:10 +00:00
Using add_handler()
^^^^^^^^^^^^^^^^^^^
2019-05-10 14:14:10 +00:00
The :meth:`~pyrogram.Client.add_handler` method takes any handler instance that wraps around your defined callback
function and registers it in your Client. It us useful in case you want to programmatically add handlers (or in case,
for some reason, you don't like to use decorators).
2019-05-10 14:14:10 +00:00
.. code-block:: python
2019-05-10 14:14:10 +00:00
from pyrogram import Client
from pyrogram.handlers import MessageHandler
2019-05-10 14:14:10 +00:00
def my_function(client, message):
message.forward("me")
2019-05-10 14:14:10 +00:00
app = Client("my_account")
my_handler = MessageHandler(my_function)
app.add_handler(my_handler)
2019-05-10 14:14:10 +00:00
app.run()
2019-05-10 14:14:10 +00:00
The same about asynchronous handlers applies for :meth:`~pyrogram.Client.add_handler`:
2019-05-10 14:14:10 +00:00
.. code-block:: python
async def my_function(client, message):
await message.forward("me")