Regular Expressions in Training

I would like to be able to train stories based on regular expressions.

For example, if a feed contained articles like “funny cats (youtube)” and “funny dogs (vimeo)” and I wanted all funny videos from youtube I could specify “funny.*youtube.”


Quite a power user feature. Don’t think I’ll support this anytime soon, as it will complicate the UI quite a bit. Phrase matching isn’t doing it, huh?

I figured it was a stretch :slight_smile:

Phrase matching works for probably 95% of my cases, but sometimes I get a phrase where I want to match the front and end, but not the middle. Wildcards would probably work just as well as regular expressions for me.

A more user-friendly option might be something like the interface for iTunes Smart Albums. “funny” AND “youtube” in my example above. or “funny” AND NOT “vimeo.” I think the current behavior of “green + red = green” is the best default behavior, but it would be nice to be able to tweak it.