Awesome

pyfram-telegram-bot

Open-source WolframAlpha Telegram bot

• Telegram API wrapper: telebot
• Deployed on: Heroku
• Static type checking: mypy
  • mypy pyfram_bot.py api_calls.py -> Success: no issues found in 2 source files
• Unit tests: test_bot.py
  • Sometimes you'll have to run tests twice (or even thrice). It's rather a problem with the API and time needed to process the request for step-by-step image.

Current status:

The bot works pretty fine for the end-user, all the functionality I wanted to introduce is already there. Only the code needs some polishing

What can the end-user achieve with bot? How to use it?

• /s or /steps command before actual query — get step-by-step solution (if possible).
  • NOTE This may take quite a long time but it's not a problem with my code, it's WolframAlpha API issue.
• /i or /image command before actual query — force get result as an image.
• /start or /help commands — provide basic help.
• If no commands are provided, the bot has these defaults:
  • Try getting and sending short text answer
  • If the short answer is not available, return image (as in /image)
• Inline mode works only for simple answers:
  • If available, prompt will display your question, the result will be the answer
  • If unavailable, both prompt and result will be "No short answer available"

Video preview:

How the code works?

Well, it's no dark magic. I simply query WolframAlpha API and send text answer or image:

• get user input via telebot
• requests.get()
• then send the result via telebot

The short import lists are also self-explanatory. As you can see in TODO, in free time I'll also introduce type checking. Bot is ready to go to Heroku, code is clear and understandable (at least I hope so).

Why won't we develop a public bot?

Wolfram API for a non-personal use is quite expensive and we cannot afford being Santa Claus. That's why we provide only the code and left the appID empty. Feel free to set up such a bot for personal use with your own API key or to gain funds for a fully public one.

But if you want to see the bot in action: @wolaph_bot

api_calls.py simple documentation

The whole API wrapper is developed in WolframBot class.

• Attributes:

  • __app_ids (list) — list of app_ids to use with the WolframBot
  • __num_of_appids (int) — how many app IDs are provided?

• Methods:

  • __init__
    • Arguments:
      • app_ids (list) — list of app_ids to use with the WolframBot
    • Returns:
      • None
  • __get_random_id
    • Arguments:
      • None
    • Returns:
      • Randomly chosen app ID from self.__app_ids (str)
  • __short_answer
    • Arguments:
      • query (str) — question to ask WolframAlpha
    • Returns:
      • Result of querying WolframAlpha. May return "No short answer available" (str)
  • __get_image
    • Arguments:
      • query (str) — question to ask WolframAlpha
    • Returns:
      • Name of the file where result is stored. May return "Error processing your request." (str)
  • __get_url_of_steps_image
    • Arguments:
      • img_xml (str) — raw query result.
    • Returns:
      • Desired url from XML. May return "No step by step solution available." (str)
  • get_step_by_step
    • Arguments:
      • query (str) — question to ask WolframAlpha
    • Returns:
      • Name of the file where result is stored. May return "Error processing your request." or "No step by step solution available." (str)
  • query_wolfram
    • Arguments:
      • query (str) — question to ask WolframAlpha
      • is_image (bool, optional) — if True, the method will skip trying to get short answer and proceed to fetching image in the first place
      • inline_mode (bool, optional) — if True, the method will skip trying to get image answer and will send any result of fetching simple answer.
    • Returns:
      • Short answer, filename, may also return "Error processing your request." or "No step by step solution available." (str)

Contribution

Feel free to create issues, to make PRs, every kind of help is welcome here

You can reach me and talk about this project on:

• Telegram (preferred): @skelly37
• Discord: skelly#3578

License

The code is and will be distributed under the permissive MIT License