Refactor the hint positioning to work with more cases
Closed this issue · 2 comments
KennethSundqvist commented
Collect a lot of wonky edge cases before doing this.
Cases (so far):
- Visible element with a single non-wrapping text node. (Single client rect.)
- Position: The element or the client rect. Don't know if there is any practical difference.
- Visible element with a single text node that wraps to multiple lines. (One client rect for each line.)
- Position: First visible client rect. (Some lines could be scrolled out of view and the lines can have different indentation or wrap around a shape.)
- Collapsed element with only non-text child nodes.
- Example: User picture besides a comment on a Github issue. #52
nskmda commented
Not sure if this is in the works or something.
I noticed that if there are more than 10 elements on the page, it's impossible to select the first hint.
E.g, if there are 55 elements, I can't jump to 2, because there are 20, 21, 22 etc.
Or is there a way to activate the first hint matching?
KennethSundqvist commented
Hi @nskmda!
You can press Enter
to activate the hint for the number you've typed. So you would press 2
> Enter
to do that.
Thanks for letting me know that this behavior was unclear. I've updated the README to explain how this works.