Closed slifty closed 5 years ago
OK here's the blunt truth I really shouldn't merge this. You brought up some really good points and some linting issues that ought to be resolved.
However... I'm... gonna just merge because there are SO MANY LINTING ERRORS when I run the linter. I call this maneuver an ostrich merge.
However, I'm going to also make an issue for actually fixing the linting problems, and once that happens I'll lint as part of my commit flow going forward.
We now render a list of identified phrases which appear as the user types. The list does NOT account for the size of the text area, and it renders all phrases flagged in the entire document. It also does not currently render the flag types and instance counts as per the visual spec.
This also adds support for real data (though it doesn't gracefully handle missing data -- that's because eventually we are hopeful to be able to commit a baseline dataset directly)