Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.16%. Comparing base (77a595e) to head (7802bda).
Report is 494 commits behind head on dev.

@@           Coverage Diff           @@
##              dev   #10089   +/-   ##
=======================================
  Coverage   91.15%   91.16%           
=======================================
  Files         411      411           
  Lines       12466    12467    +1     
  Branches     2419     2420    +1     
=======================================
+ Hits        11364    11365    +1     
  Misses        557      557           
  Partials      545      545           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

This definitely needs some testing from out side as well

cc @Yomodo if you are interested to test

Yah I've got a bunch in queue I used but didn't push. I'll move them to a page then push them in the morning.

Add appbartest to the components page with 3 relevant example sections and a half dozen examples. In addition to those I looked at every component page checking things to try and be sure. Please do the same and let me know if anything is off.

I think @danielchalmers meant to add them in the MudBlazor.UnitTests.Viewer project under MudBlazor/src/MudBlazor.UnitTests.Viewer/TestComponents/Popover directory, with the Test file name in the end. Then you can startup the MudBlazor.UnitTests.Viewer project instead and use it as playground there, to not pollute the docs. Since I think it would make sense to merge it with all this test files for future.

Done. Almost all of these are straight from the github issues just so you know.

This is an epic PR. You fixed an issue even @Garderoben and the entire team failed to fix for a long time and not for a lack of trying. We just never had the right idea how to do it. I only thought of trying cascading parameters or so, @ScarletKuro wanted to replace the entire popover library with a JS library, etc .... ;)

I do not expect this to be a breaking change as anyone overriding the zindex already are doing so to mimic this behavior (or ignore popover problems) and this shouldn't affect that.

I removed the breaking change label based on this statement. I hope that is correct.

This is an epic PR. You fixed an issue even @Garderoben and the entire team failed to fix for a long time and not for a lack of trying. We just never had the right idea how to do it. I only thought of trying cascading parameters or so, @ScarletKuro wanted to replace the entire popover library with a JS library, etc .... ;)

Thank you. I was 10 hours in when I decided to just start commenting everything and figured out the wrong parent was being used to calculate zindex. I appreciate it.

Okay, the popover content display regression is introduced in this PR @versile2

(Doesn't display the results, however without this PR it's working fine)

We missed the "Direction and Location" problem in the docs as well.
It doesn't react when you change anchor origin and transform origin,

To save some time, i figured that the content display problem is cause by the _popover.scss change, but the placement is because of the js change, tho I haven't figured what exact line caused it.

Ok basically there's two types of issues. Issue
1, the popover is a child of any other item that has a higher zindex. Dialog, Overlay, Appbar, whatever. My original idea was to make exclusions for each of those but now I think I just need to do a 2nd iteration if the parent isn't a popover.
2. the popover is a child of another popover - this one I'm certain is fixed.

From the Select examples on the dev branch:

Opening upwards:

Off to the side:

https://dev.mudblazor.com/components/select#visual-playground

Got both fixed in another branch. I'll submit when tested thoroughly.

Wish my viewer changes were in lol