enhancement: link clickable span detection

[diegoberaldin]

Hi! I am using your library in my Lemmy client and I really like it (lightweight, easy to use, extensible, multiplatform to the core, perfect MD integration, etc.). Kudos!

My users reported the following bug: if there is a link inside a text node, the whole line is clickable and opens it even if the tapped position exceeds the anchor text length, i. e. provided that the tap event occurs in the same line after the link, the link is opened anyways.

Steps to reproduce: in the example application change the last two lines of the sample text like this

  [https://github.com/mikepenz](https://github.com/mikepenz), but this part of the line is not clickable
  [Mike Penz's Blog](https://blog.mikepenz.dev/) neither this part of the line should be clickable

The solution was a little more difficult than expected, because the LINK_DEFINITION node handler is not called for autolinks or links inside a text node, but just for internal Markdown references, so I had to modify even the text handlers. Another "weird" behaviour which I found is that annotations are added multiple times for the same reference, which was somehow "hidden" by the reversed().firstOrNull() access to the list, but the test on the hit rect against the annotation position had to be improved.

[mikepenz]

Thank you very much for the PR.
Will try to get it reviewed as soon as possible. had a quick peek, and yeah as you wrote it's quite a tricky change :D

[diegoberaldin]

Hi Mike, don't worry take your time to fully review it and making sure it does not break more things than it appears to fix. I hope I managed to explain what the issue was (a tap happening on the same line where a hyperlink is, but outside the text bounds on the "right" side).

PS. I hope I'll be able to open more PRs in the future, "my" users found some other minor issues and some of them can be addressed. Others are unmanageable in my opinion (because of Lemmy's Markdown dialect, e.g. spoilers) so I can work around them client-side in some way.

[mikepenz]

Thank you again for the PR. I identified a simpler solution for the problem, however I am not sure it solves all problems you noted. I will open a PR and get it merged. Would you be so kind to check it out with your usecases please?

[mikepenz]

Please see here: #135