Unreliable detection

[cuddeford]

It's relatively easy to pass the profanity detection. Here's an example:

dickface -> profane
dicklord -> not profane

Surely there should be some kind of mechanism to detect the hotword at any position in a string, regardless of prefix and suffix?

[Andrews54757]

@cuddeford #1

This is actually an outdated, broken library. The updated library is more advanced, and incorporates a machine learning algorithm.

However, the new version is very slow. During my test runs, people have purposely inundated the bot with text to crash it.

As a result, I have not published that version yet. In the future, I may release once this issue has been resolved.

For now, try using another library. The current version of this library is experimental at best.

[romzamir]

@cuddeford #1

This is actually an outdated, broken library. The updated library is more advanced, and incorporates a machine learning algorithm.
However, the new version is very slow. During my test runs, people have purposely inundated the bot with text to crash it.
As a result, I have not published that version yet. In the future, I may release once this issue has been resolved.
For now, try using another library. The current version of this library is experimental at best.

any idea when will it will be released?