Add Dependencies to migrations guide (#1504)

* Add Dependencies to migrations guide

* Added changelog

CHANGES/1504.doc.rst

@@ -0,0 +1 @@
+Added information about dependency changes to the :code:`2.x --> 3.x` migration guide.

docs/migration_2_to_3.rst

@@ -6,9 +6,9 @@ Migration FAQ (2.x -> 3.0)
 
     This guide is still in progress.
 
-This version introduces numerous breaking changes and architectural improvements. 
+This version introduces numerous breaking changes and architectural improvements.
-It helps reduce the count of global variables in your code, provides useful mechanisms 
+It helps reduce the count of global variables in your code, provides useful mechanisms
-to modularize your code, and enables the creation of shareable modules via packages on PyPI. 
+to modularize your code, and enables the creation of shareable modules via packages on PyPI.
 It also makes middlewares and filters more controllable, among other improvements.
 
 
@@ -16,18 +16,26 @@ On this page, you can read about the changes made in relation to the last stable
 
 .. note::
 
-    This page more closely resembles a detailed changelog than a migration guide, 
+    This page more closely resembles a detailed changelog than a migration guide,
     but it will be updated in the future.
 
     Feel free to contribute to this page, if you find something that is not mentioned here.
 
+Dependencies
+==========
+
+- The dependencies required for :code:`i18n` are no longer part of the default package.
+  If your application uses translation functionality, be sure to add an optional dependency:
+
+  :code:`pip install aiogram[i18n]`
+
 
 Dispatcher
 ==========
 
-- The :class:`Dispatcher` class no longer accepts a `Bot` instance in its initializer. 
+- The :class:`Dispatcher` class no longer accepts a `Bot` instance in its initializer.
-  Instead, the `Bot` instance should be passed to the dispatcher only for starting polling 
+  Instead, the `Bot` instance should be passed to the dispatcher only for starting polling
-  or handling events from webhooks. This approach also allows for the use of multiple bot 
+  or handling events from webhooks. This approach also allows for the use of multiple bot
   instances simultaneously ("multibot").
 - :class:`Dispatcher` now can be extended with another Dispatcher-like
   thing named :class:`Router` (:ref:`Read more » <Nested routers>`).
@@ -35,11 +43,11 @@ Dispatcher
 - Removed the **_handler** suffix from all event handler decorators and registering methods.
   (:ref:`Read more » <Event observers>`)
 - The Executor has been entirely removed; you can now use the Dispatcher directly to start poll the API or handle webhooks from it.
-- The throttling method has been completely removed; you can now use middlewares to control 
+- The throttling method has been completely removed; you can now use middlewares to control
   the execution context and implement any throttling mechanism you desire.
 - Removed global context variables from the API types, Bot and Dispatcher object,
-  From now on, if you want to access the current bot instance within handlers or filters, 
+  From now on, if you want to access the current bot instance within handlers or filters,
-  you should accept the argument :code:`bot: Bot` and use it instead of :code:`Bot.get_current()`. 
+  you should accept the argument :code:`bot: Bot` and use it instead of :code:`Bot.get_current()`.
   In middlewares, it can be accessed via :code:`data["bot"]`.
 - To skip pending updates, you should now call the :class:`aiogram.methods.delete_webhook.DeleteWebhook` method directly, rather than passing :code:`skip_updates=True` to the start polling method.
 
@@ -49,18 +57,18 @@ Filtering events
 ================
 
 - Keyword filters can no longer be used; use filters explicitly. (`Read more » <https://github.com/aiogram/aiogram/issues/942>`_)
-- Due to the removal of keyword filters, all previously enabled-by-default filters 
+- Due to the removal of keyword filters, all previously enabled-by-default filters
-  (such as state and content_type) are now disabled. 
+  (such as state and content_type) are now disabled.
   You must specify them explicitly if you wish to use them.
   For example instead of using :code:`@dp.message_handler(content_types=ContentType.PHOTO)`
   you should use :code:`@router.message(F.photo)`
 - Most common filters have been replaced with the "magic filter." (:ref:`Read more » <magic-filters>`)
-- By default, the message handler now receives any content type. 
+- By default, the message handler now receives any content type.
   If you want a specific one, simply add the appropriate filters (Magic or any other).
-- The state filter is no longer enabled by default. This means that if you used :code:`state="*"` 
+- The state filter is no longer enabled by default. This means that if you used :code:`state="*"`
-  in v2, you should not pass any state filter in v3. 
+  in v2, you should not pass any state filter in v3.
   Conversely, if the state was not specified in v2, you will now need to specify it in v3.
-- Added the possibility to register global filters for each router, which helps to reduce code 
+- Added the possibility to register global filters for each router, which helps to reduce code
   repetition and provides an easier way to control the purpose of each router.
 
 
@@ -68,18 +76,18 @@ Filtering events
 Bot API
 =======
 
-- All API methods are now classes with validation, implemented via 
+- All API methods are now classes with validation, implemented via
-  `pydantic <https://docs.pydantic.dev/>`. 
+  `pydantic <https://docs.pydantic.dev/>`.
   These API calls are also available as methods in the Bot class.
-- More pre-defined Enums have been added and moved to the `aiogram.enums` sub-package. 
+- More pre-defined Enums have been added and moved to the `aiogram.enums` sub-package.
   For example, the chat type enum is now :class:`aiogram.enums.ChatType` instead of :class:`aiogram.types.chat.ChatType`.
-- The HTTP client session has been separated into a container that can be reused 
+- The HTTP client session has been separated into a container that can be reused
   across different Bot instances within the application.
-- API Exceptions are no longer classified by specific messages, 
+- API Exceptions are no longer classified by specific messages,
-  as Telegram has no documented error codes. 
+  as Telegram has no documented error codes.
-  However, all errors are classified by HTTP status codes, and for each method, 
+  However, all errors are classified by HTTP status codes, and for each method,
-  only one type of error can be associated with a given code. 
+  only one type of error can be associated with a given code.
-  Therefore, in most cases, you should check only the error type (by status code) 
+  Therefore, in most cases, you should check only the error type (by status code)
   without inspecting the error message.
 
 
@@ -87,12 +95,12 @@ Bot API
 Middlewares
 ===========
 
-- Middlewares can now control an execution context, e.g., using context managers. 
+- Middlewares can now control an execution context, e.g., using context managers.
   (:ref:`Read more » <middlewares>`)
 - All contextual data is now shared end-to-end between middlewares, filters, and handlers.
   For example now you can easily pass some data into context inside middleware and
   get it in the filters layer as the same way as in the handlers via keyword arguments.
-- Added a mechanism named **flags** that helps customize handler behavior 
+- Added a mechanism named **flags** that helps customize handler behavior
   in conjunction with middlewares. (:ref:`Read more » <flags>`)
 
 
@@ -116,24 +124,24 @@ Callbacks data
 Finite State machine
 ====================
 
-- State filters will no longer be automatically added to all handlers; 
+- State filters will no longer be automatically added to all handlers;
   you will need to specify the state if you want to use it.
-- Added the possibility to change the FSM strategy. For example, 
+- Added the possibility to change the FSM strategy. For example,
-  if you want to control the state for each user based on chat topics rather than 
+  if you want to control the state for each user based on chat topics rather than
   the user in a chat, you can specify this in the Dispatcher.
 - Now :class:`aiogram.fsm.state.State` and :class:`aiogram.fsm.state.StateGroup` don't have helper
   methods like :code:`.set()`, :code:`.next()`, etc.
 
 - Instead, you should set states by passing them directly to
   :class:`aiogram.fsm.context.FSMContext` (:ref:`Read more » <Finite State Machine>`)
-- The state proxy is deprecated; you should update the state data by calling 
+- The state proxy is deprecated; you should update the state data by calling
   :code:`state.set_data(...)` and :code:`state.get_data()` respectively.
 
 
 Sending Files
 =============
 
-- From now on, you should wrap files in an InputFile object before sending them, 
+- From now on, you should wrap files in an InputFile object before sending them,
   instead of passing the IO object directly to the API method. (:ref:`Read more » <sending-files>`)
 
 
@@ -154,7 +162,7 @@ Telegram API Server
 Telegram objects transformation (to dict, to json, from json)
 =============================================================
 
-- Methods :class:`TelegramObject.to_object()`, :class:`TelegramObject.to_json()` and :class:`TelegramObject.to_python()` 
+- Methods :class:`TelegramObject.to_object()`, :class:`TelegramObject.to_json()` and :class:`TelegramObject.to_python()`
   have been removed due to the use of `pydantic <https://docs.pydantic.dev/>`_ models.
 - :class:`TelegramObject.to_object()` should be replaced by :class:`TelegramObject.model_validate()`
   (`Read more <https://docs.pydantic.dev/2.7/api/base_model/#pydantic.BaseModel.model_validate>`_)
@@ -172,19 +180,19 @@ Here are some usage examples:
     # Version 2.x
     message_dict = {"id": 42, ...}
     message_obj = Message.to_object(message_dict)
-    print(message_obj)  
+    print(message_obj)
     # id=42 name='n' ...
     print(type(message_obj))
     # <class 'aiogram.types.message.Message'>
-    
+
     # Version 3.x
     message_dict = {"id": 42, ...}
     message_obj = Message.model_validate(message_dict)
-    print(message_obj)  
+    print(message_obj)
     # id=42 name='n' ...
     print(type(message_obj))
     # <class 'aiogram.types.message.Message'>
-        
+
 - Creating a json representation of an object
 
 .. code-block::
@@ -192,14 +200,14 @@ Here are some usage examples:
     async def handler(message: Message) -> None:
         # Version 2.x
         message_json = message.as_json()
-        print(message_json)  
+        print(message_json)
         # {"id": 42, ...}
         print(type(message_json))
         # <class 'str'>
-        
+
         # Version 3.x
         message_json = message.model_dump_json()
-        print(message_json)  
+        print(message_json)
         # {"id": 42, ...}
         print(type(message_json))
         # <class 'str'>
@@ -211,14 +219,14 @@ Here are some usage examples:
     async def handler(message: Message) -> None:
         # Version 2.x
         message_dict = message.to_python()
-        print(message_dict)  
+        print(message_dict)
         # {"id": 42, ...}
         print(type(message_dict))
         # <class 'dict'>
-        
+
         # Version 3.x
         message_dict = message.model_dump()
-        print(message_dict)  
+        print(message_dict)
         # {"id": 42, ...}
         print(type(message_dict))
         # <class 'dict'>