* Ignore clicks outside the start/end of a line
`LinkMovementMethod` has a bug in its calculation of the clickable width of a span on a line. If the span is the last thing on the line the clickable area extends to the end of the view. So the user can tap what appears to be whitespace and open a link.
Previous code tried to fix this by adding a zero-width space after the link so that `LinkMovementMethod` wouldn't consider it empty. However the ZWS was selected by copy/paste operations, resulting in junk results if users tried to copy the link.
Fix this by subclassing `LinkMovementMethod` and fixing the click detection code to ignore clicks that are outside the bounds of the line that was clicked on.
Remove the code that adds the ZWS.
Fixes https://github.com/tuskyapp/Tusky/issues/1567
* Assume arguments are all non-null
* Use `object` for singleton
* getInstance as a one-liner
* Reinstate optional login via custom browser tab
* Clarify the buttons for the different login options
* Add informative labels for the different login options
* Move "Login with Browser" to the options menu
* Fix off-by-one error in HttpHeaderLink
Link headers with multiple URLs with multiple parameters were being parsed
incorrectly.
Detected by adding unit tests ahead of converting to Kotlin.
* Convert util/HttpHeaderLink from Java to Kotlin
* Convert util/ThemeUtils from Java to Kotlin
* Convert util/TimestampUtils from Java to Kotlin
* Add tests for PairedList
* Convert util/PairedList from Java to Kotlin
* Implement feedback from PR
* Relicense as GPL
* Show target domains for non-mention/non-hashtag links where the target domain is not provided or differs from the domain in the text.
Addresses #2694
* Add link signifier to the marked-up domain
* Back down on validating hashtags and mentions, don't markup _any_ urls where the text starts with #/@
* Migrate LinkHelper to kotlin
* Support tags field on statuses
* Use embedded tags list in status instead of text scraping to embed tag click handler.
Fixes#2283
* Make mentions and tags lists nonnullable
* Make LinkHelper.openLink a Context extension method
* Use builtin extension for uri conversion
* More cleanup in LinkHelper
* Add tests for LinkHelper.getDomain
* Unbreak tags in places that don't have a tag list (e.g. profiles)
* Fixup javadoc
Goooler
Nik Clayton
Levi Bard
kyori19
Josh Soref
Conny Duck