slackoverflow.py

import os
import time
from slackclient import SlackClient
import requests
import subprocess
from webScraper import StackoverflowWebScraper

# instantiate Slack client
slack_client = SlackClient(os.environ["SLACK_API_TOKEN"])
# starterbot's user ID in Slack: value is assigned after the bot starts up
bot_id = None

# instantiate scraper
web_scraper = StackoverflowWebScraper()

# constants
RTM_READ_DELAY = 1  # 1 second delay between reading from RTM
CHARACTER_LIMIT = 200  # character limit for the response.

def reply_thread(thread_ts, channel, response):
    title, link, text = response
    slack_client.api_call(
        "chat.postMessage",
        channel=channel,
        attachments=[
            {
                "title": title,
                "pretext": link,
                "text": text,
                "mrkdwn_in": ["text"]
            }
        ],
        thread_ts=thread_ts
    )

def post_to_channel(channel, response):
    title, link, text = response
    slack_client.api_call(
        "chat.postMessage",
        channel=channel,
        attachments= [
            {
                "title": title,
                "pretext": link,
                "text": text,
                "mrkdwn_in": ["text"]
            }
        ]
    )

def parse_events(slack_events):
    for event in slack_events:
        # TODO: Make sure that we at least take care of the obvious edge cases.
        if event["type"] == "message":
            if "subtype" in event:
                # Ignore bot messages
                if event["subtype"] == "bot_message":
                    return

            if "text" in event:
                message = event["text"]
                # If DMing bot
                if event["channel"][0] == "D":
                    response = generate_answer(message, True)
                    post_to_channel(event["channel"], response)
                # If mentioning bot
                elif bot_id in message:
                    message = message.replace("<@{0}>".format(bot_id), "")
                    response = generate_answer(message, False)
                    reply_thread(event["ts"], event["channel"], response)


def html_mrkdwn(text):
    text = text.replace('<p>', '').replace('</p>', '')
    text = text.replace('<b>', '*').replace('</b>', '*')
    text = text.replace('<code>', '`').replace('</code>', '`')
    text = text.replace('<pre>', '```').replace('</pre>', '```')
    text = text.replace('<em>', '_').replace('</em>', '_')
    text = text.replace('<strong>', '*').replace('</strong>', '*')
    text = text.replace('````', '```')
    return text


def generate_answer(message, dm=False):
    char_limit = CHARACTER_LIMIT * 3 if dm else CHARACTER_LIMIT
    question = web_scraper.get_top_question(message)
    response = (
        "404", "404", "Couldn't find a matching question on stackoverflow. Try rephrasing your question.")
    if question is not None:
        answer = web_scraper.get_top_answer(question["question_id"])
        response = (question["title"], question["link"],
                    html_mrkdwn(answer["body"]))
    return response


if __name__ == "__main__":
    if slack_client.rtm_connect(with_team_state=False):
        print("Slack is Overflowing")
        # Read bot's user ID by calling Web API method `auth.test`
        bot_id = slack_client.api_call("auth.test")["user_id"]
        while True:
            parse_events(slack_client.rtm_read())
            time.sleep(RTM_READ_DELAY)
    else:
        print("Connection failed. Exception traceback printed above.")