Page MenuHomePhabricator

[native] Fix display of TooltipInlineEngagement
ClosedPublic

Authored by ashoat on Wed, Oct 9, 8:25 PM.
Tags
None
Referenced Files
F3096241: D13677.id45036.diff
Tue, Oct 29, 12:13 PM
Unknown Object (File)
Mon, Oct 28, 4:56 PM
Unknown Object (File)
Sun, Oct 27, 11:02 PM
Unknown Object (File)
Sat, Oct 26, 12:35 PM
Unknown Object (File)
Fri, Oct 25, 12:38 PM
Unknown Object (File)
Wed, Oct 23, 3:30 AM
Unknown Object (File)
Mon, Oct 21, 6:56 PM
Unknown Object (File)
Sun, Oct 20, 7:31 AM
Subscribers
None

Details

Summary

This diff does two things:

  1. We need to pass positioning from TooltipInlineEngagement to InlineEngagement in order to make sure that the reactions and sidebar button appear in the same order. The point of TooltipInlineEngagement is to appear directly over InlineEngagement to make the animation look good; if the order of buttons is different, the animation is ruined.
  2. After passing positioning to InlineEngagement, I was able to clean up some of the styles used in TooltipInlineEngagement.

Depends on D13676

Test Plan

On my local iOS simulator, I tested 6 cases: multimedia, text, and robotext, each sent from the viewer and from somebody else. I made sure each test case had a sidebar and a reaction. In all cases, I made sure that the inline engagement appears exactly in the place it should.

Diff Detail

Repository
rCOMM Comm
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

ashoat requested review of this revision.Wed, Oct 9, 9:09 PM
tomek added inline comments.
native/chat/inline-engagement.react.js
494 ↗(On Diff #45036)

Why this can be removed?

This revision is now accepted and ready to land.Fri, Oct 11, 5:56 AM
native/chat/inline-engagement.react.js
494 ↗(On Diff #45036)

To be honest, I'm not sure. I initially thought that passing positioning to InlineEngagement resulted in these no longer being necessary, but after reading the code, that theory no longer make sense to me.

I now suspect that a regression was introduced here recently, and these changes fixed the regression.

What I know is that I tested a wide variety of cases, and the InlineEngagement appeared in the incorrect place without these changes, and in the correct place with these changes.