behaviorbot / sentiment-bot Goto Github PK

Thanks so much for such an awesome tool. Just set it up for my mental health open source project https://github.com/ifmeorg/ifme.

I notice it works for issue comments, but does it also work for pull request comments? I tried testing it out, but no bot.

proposed solutions:

❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️ ❤️
🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉 🎉
💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙 💙
💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯 💯
💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃 💃

You can see from the comment posted in probot/probot#258 (comment) that the bot currently can't mention teams. As discussed in probot/probot#337, the app needs Read permission on Organization members.

What do you think about adding that to the app's permissions? I'm not thrilled that it's necessary, but I imagine this bot will often be used to mention teams.

For a repo or org, have a Slack team and channel to report toxic comments in
Perhaps even a way for a moderator to mark that the comment has been handled so that behavior-bot can remind the channel if something isn't handled within X amount of time

h/t @lee-dohm

So basically this bot depends on these changes: octokit/octokit.js#555. Right now I've simply replace it in my node_modules, but that won't work when this gets deployed, so I'm not sure what's the best option.

The only line I truly need from my PR is the accept header here. According to some of their docs I can add an accept header when I instantiate the GitHub API; unfortunately, probot does the work on that end.

Options:

1. Update probot's node-github handling to allow additional accept headers
2. Don't use context.github and authenticate through node-github directly in my bot and pin my branch in the package.json
3. Beg @zeke to merge my PR 😆

Thoughts @bkeepers?

The bot checks the first span score returned by the Perspective API here:

Why only use the first span score? Wouldn't it be more accurate to use the summary score?

See the Perspective API docs.

1. Summary

It would be nice, if user can customize this part of Sentiment Bot message:

Keep in mind, this repository uses the Contributor Covenant.

Now user can't change this text.

2. Argumentation

More customization — always good.

User can want to:

1. use another message instead of Keep in mind, this repository uses the Contributor Covenant,
2. using [**bold links**](http://example.link) instead of [**link**](http://example.link).

3. Example of expected behavior

If in config.yml:

sentimentBotReplyComment: |
  Please be sure to review the code of conduct and be respectful of other users. cc/ @Kristinita

  [**CODE_OF_CONDUCT.md**](https://<link to CODE_OF_CONDUCT>) for this repository.

sentiment-bot will use [**CODE_OF_CONDUCT.md**](https://<link to CODE_OF_CONDUCT>) for this repository. instead of Keep in mind, this repository uses the Contributor Covenant.

Thanks.

Examples from Perspective:

They are liberal idiots who are uneducated.

Climate change is happening and it's not changing in our favor. If you think differently you're an idiot.

They're allowed to do that. But if they act like assholes about, I will block them.

I think its a farce and stinks like a bathroom after 26 beers

My thoughts are that people should stop being stupid and ignorant. Climate change is scientifically proven. It isn't a debate.

I thought

This is something I'll be fixing later this week for my team 😆 crazy how everything is connected.

cc/ @benbalter @bkeepers for interesting things :)

1. Summary

It would be nice, if sentimentBotToxicityThreshold key description will be more understandable.

2. Details

Sentiment Bot description:

*Required* toxicity threshold between 0 and .99 with the higher numbers being the most toxic
Anything higher than this threshold will be marked as toxic and commented on

What is “being the most toxic”?

Did I understand correctly that:

• 0 — maximum words and phrases will be marked as toxic,
• .99 — minimum words and phrases will be marked as toxic.

?

Or vice versa?

Thanks.

I went to set up the bot for my own repos, but there was no documentation. :sad:

Would it be possible to add a README like the other behaviorbot plugins? 😄

Re: behaviorbot/welcome#8 updating to standard linting, which means: in the package.json changing:

You will probably also need to add to the package.json this:

"standard": {
    "env": [
      "mocha"
    ]
  },

so that our tests can operate without linting errors.

Then run npm test to make sure everything works! Let me know if you have any questions.