ragepeanut / checky Goto Github PK

The bot should store in its json_metadata the 40 words surrounding each wrong mention. If a user replies to his comment with something like "!details mention" the bot should reply with something like this :

The mention @mention has been detected in this part of the text :

40 words surrounding @mention in the text

If "mention" is not specified in the command's arguments, the reply will contain all the wrong mentions.

Hello, I am a logo designer and I'm a utopian contributor. I saw your project and I want to make a logo for it. So, what do you think?

If a mention is found in a text like ![...@mention...](source) or <img src="source" alt="...@mention...">, it shouldn't be checked as it often relates to the name of an uploaded image.

When no correction is found the bot should check if the wrong mention is actually a popular tag. A lot of users use the @ character instead of a hashtag when talking about tags. If it indeed is a popular tag, send a message where the correction for @mention is #mention.

Posts containing Vue code or Typescript ignores get often matched. This kind of thing shouldn't happen since it's never meant to be a mention but part of the code.
Example: https://steemit.com/utopian-io/@drsensor/creating-rpc-server-on-web-browser-part-3-create-reverse-rpc-through-websocket

To avoid having a day without any node working, @checky should use this service that automatically updates its list of working nodes every 3 hours https://steemit.com/@fullnodeupdate

It can be hard to see what the difference is between a wrong mention and its suggested correction. The bot should put transposed, inserted and replaced characters in bold in the suggested correction. In order to show the difference for deletes, it could also put affected characters in bold in the wrong username.

A post that references another post shouldn't get a comment from Checky if a mention in the title of the other post is wrong. This would require an extra API call but needs to be done.