travitia-talk 0.0.1

ChatBot API wrapper for https://public-api.travitia.xyz/talk

Travitia Talk

Async ChatBot API wrapper for https://public-api.travitia.xyz/talk

Installation

Travitia Talk is available on PyPi

pip install travitia_talk

Getting API key

• Join Travitia API Discord at https://discord.gg/C98nsXt
• TODO

Usage

Interactive session

python -m travitia_talk

Basic usage

import travitia_talk as tt

API_KEY = "SECRET_KEY"

chatbot = tt.ChatBot(API_KEY)
response = await chatbot.ask("How are you?")
print(response)
# I'm doing okay, you?

await chatbot.close()

Emotions

Travitia Talk supports multiple emotions

import random

emotion = random.choice(list(tt.Emotion))
response = await chatbot.ask("How are you?", emotion=emotion)

Contexts

It is possible to add context to query for more accurate answers (up to 2 answers currently). By default library uses in-memory context, but it is possible to define your own by using travitia_talk.Context.

Context looks up previous queries using an identifier. In case of discord bot this could be user id. Example:

@bot.command
async def cb(self, ctx, *, text: str):
    response = await ctx.bot.chatbot.ask(text, id=ctx.author.id)

    await ctx.send(response.text)

Custom session

It is possible to use custom aiohttp session with configured timeouts and other settings.

from aiohttp import ClientSession, ClientTimeout
import travitia_talk as tt

API_KEY = "SECRET_KEY"

chatbot = tt.ChatBot(API_KEY, session=ClientSession(timeout=ClientTimeout))

Custom API URL

In case URL gets changed for some reason or library adds support for other backends it is possible to overwrite API URL:

import travitia_talk as tt

API_KEY = "SECRET_KEY"

chatbot = tt.ChatBot(API_KEY, api_url="https://paid-api.travitia.xyz/talk")