EasyTeleBot

EasyTeleBot is a package for creating an automated bot responder for telegram in python.

The program uses flask and webhook to get user messages from telegram.

Installation

Install the package by entering in terminal

pip install easytelebot

Package Dependencies:

• flask
• python-telegram-bot

Usage

from EasyTeleBot.TelegramBot import EasyTelegramBot

After importing the EasyTeleBot get the config json file in one of two ways:

1. As string path config_file = "config_file.json"
2. As IO readable config_file = open("config_file.json")
3. As a list of the two above config_files = [config_file1, config_file2]

Use CreateEasyTelegramBot to create the bot it will create the whole bot automatically.

bot = EasyTelegramBot(config_files)

Inside this class lays a flask app which you can get by typing - bot.flask_app. Run this flask app to start handling the messages telegram is sending to the bot (user message go through telegram servers).

Tested WSGI Usages:

• gunicorn gunicorn myapp:bot.flask_app
• nginx
• Django

Telegram bot chats start with the word "/start" so you should enter this kind of action

Config File

Create a config file in json format according to these rules :

1. "telegram_token" - Is the telegram api token you get after creating a new bot through BotFather.
2. "webhook_url" - This is the url to tell telegram servers to forward the messages users send to bot chat, typically it will be the server url beginning and an identifier. example: "https://example-server.herokuapp.com/example-identifier". The bot listens to this url.
3. "bot_name" - Your bot name. Will make the flask server with this name.
4. "actions" - A list of actions to configure the bot behavior.

Optional:

• "default_action_id" - If no action triggers this action id will perform.

You can also put those setting when constructing the class, but not actions.

bot = EasyTelegramBot([file, file2], telegram_token="123", webhook_url="hook_url", bot_name="new_bot", default_action_id=3)

Example config_file.json

{
    actions:[
        ...
    ]  
    "telegram_token": "example_token:KG5Gdv6AK05aeGctf25",
    "webhook_url": "https://example-server.herokuapp.com/example-identifier",
    "bot_name": "example_bot",

    "default_action_id": 50621
}

Action

Each action has these attributes:

1. "id" - Action identifier.
2. "triggers" - List of user messages that initiate this action.
3. "type" - The type of action. Each type of action does a different thing.
4. "data" - The data of the function (read Action Types to better understand). If there is a format, it will try to extract saved information, example "${name}" will change to the value saved in "name". "${last_text_received}" will be the last text message the user sent to chat.

Optional attributes:

• "next_action_id" - Do this ids action right after the this action finishes.
• "follow_up_action_id" - The follow up action id, does this after the user sends back a message. This is for question actions.
• "markup_type" - When sending a markup, the users chat will have options to press, like "/start" and "/help". Every ',' will separate between words and every ':' will separate between rows.
  • "remove" - Removes the previous markup the users chat has.
  • "reply" - Sends the markup to the user, when the user presses on one, it minimizes it.
  • "static-reply" - Sends the markup to the user, it will stay until you remove it.
• "markup_data" - If "markup_type" is "reply" or "static-reply", uses this to make a list of reply keyboard options.

Do not use both "next_action_id" and "follow_up_action_id" in the same action. Use only one of them or none.

Example action

...
{
    "id": 1,
    "triggers": ["hi", "hello", "hey"],
    "type": "text",
    "data": "server says hello back",
    "next_action_id": 2
},
{
    "id": 2,
    "triggers": [],
    "type": "text",
    "data": "server got to this action right after action id 1, now you got markup options",
    "markup_type": "reply",
    "markup_data": "left top,right top:left bottom,right bottom"
},
...

Action Types

text

Sends back a message containing the value in "data".

...
    "triggers": ["hi", "hello", "hey"],
    "type": "text",
    "data": "server says hello back",
...

animation

Sends back an animation where "data" value is the url of the animation. Converts all to gifs.

...
    "triggers": ["puppy"],
    "type": "animation",
    "data": "https://media.giphy.com/media/5awVf7q1nwkus/giphy.gif",
...

save

Saves the value in "data" into "save_to_data_name" name space. To output this value, the "data" format is "${data_name}".

...
    "triggers": ["save 'ok' into 'blop' data name"],
    "type": "save",
    "data": "ok",
    "save_to_data_name": "blop"
...

Usually you should use it to save user input like this example:

...
{
    "id": 1,
    "triggers": ["save my name"],
    "type": "text",
    "data": "please enter your name",
    "follow_up_action_id": 2
},
{
    "id": 2,
    "triggers": [],
    "type": "save",
    "data": "${last_text_received}",
    "save_to_data_name": "name",
    "next_action_id": 3
},
{
    "id": 3,
    "triggers": [],
    "type": "text",
    "data": "your name is ${name}",
},

calculate

Calculates the expression in "data" using math and saving it into "calculate_result".

...
{
    "id": 1
    "triggers": ["calculate expression"],
    "type": "calculate",
    "data": "5^5",
    "next_action_id": 2
},
{
    "id": 2,
    "triggers": [],
    "type": "text",
    "data": "5^5 = ${calculate_result}"
}
...

random

"data" is a list of possible values separated by ',', randomly will chose one of them and save into "${random_result}"

{
  "id": 1,
  "triggers": [
    "random"
  ],
  "type": "random",
  "data": "1,2,3",
  "next_action_id": 2
},
{
  "id": 2,
  "triggers": [],
  "type": "text",
  "data": "${random_result}"
}

redirect

"data" refers to the redirected action id, will works similarly to "next_action_id" but it is changeable by data saved.

{
  "id": 1,
  "triggers": ["redirect to some_action_id"],
  "type": "redirect",
  "data": "${some_action_id}"
}

You can use it like the example below but it can be easily be the action it redirects to:

{
  "id": 2,
  "triggers": ["redirect to 3"],
  "type": "redirect",
  "data": "3"
},
{
  "id": 3,
  "triggers": [],
  "type": "text",
  "data": "Hello"
}

{
  "id": 3,
  "triggers": ["redirect to 3"],
  "type": "text",
  "data": "Hello"
}

if

"data" is the evaluted text that checks for true or false, again it works with ${data} strings. "true_action_id" and "false_action_id" are the ids of the next actions if the result is true or false.

{
  "id": 1,
  "triggers": ["check"],
  "type": "if",
  "data": "${some_action_id}"
}

You can use it like the example below:

...
{
  "id": 2,
  "triggers": ["random the next action"],
  "type": "random",
  "data": "1,2",
  "next_action_id": 3
},
{
  "id": 3,
  "triggers": [],
  "type": "if",
  "data": "${random_result}>1",
  "true_action_id": 4,
  "false_action_id": 5
}
...

Or use it as a result to a message, but you need to account that it is a string so surround it with quotes. This example will always go to 4 because its always true unless some other action will redirect to it:

...
{
  "id": 3,
  "triggers": ["something"],
  "type": "if",
  "data": "'${last_text_received}'=='something'",
  "true_action_id": 4,
  "false_action_id": 5
}
...