Metadata
- Source
- FLUID-5481
- Type
- Bug
- Priority
- Critical
- Status
- Closed
- Resolution
- Cannot Reproduce
- Assignee
- Cindy Li
- Reporter
- Antranig Basman
- Created
2014-07-23T18:27:33.792-0400 - Updated
2017-09-26T10:55:55.207-0400 - Versions
-
- 1.5
- Fixed Versions
-
- 3.0
- Component
-
- Prefs Framework
Description
This problem does not occur in Firefox.
If the browser is zoomed in Chrome, 3 tests will fail from Enactors-test.html. The following details come from a pane which was zoomed to 125%.
6.
1. getLineHeight in px will report "8.97599983215332px" rather than "12px"
8.
2. line spacing in pixels will report the same.
3. The size should be doubled will report "17.95199966430664px" rather than "24px"
The ratio error is close to 4/3 but not quite. Rather than a factor of 1.3333333 it is 1.33689842072. This is likely going to be a tough behaviour to compensate for - especially since any reasonable person would have expected the factor to be related to 5/4 rather than 4/3.
This may be the reason for the report of FLUID-5384
Environments
Windows 7 x64, Chrome 36.0.1985.125, Safari 8.0.6
Comments
-
Antranig Basman commented
2015-06-03T11:00:03.692-0400 This bug also causes "jank" on startup of UIO embedded on a page which has been zoomed in Chrome. On startup, the page will snap to one size, and then to another, even if the user has not selected any adaptation.
-
Cindy Li commented
2015-06-08T12:16:32.946-0400 Seems the issue results from how Chrome calculates the line height based on the element's font size and line height em. Have reported to the Chrome bug tracker:
-
Antranig Basman commented
2015-08-18T22:20:18.949-0400 I am fairly confident that this bug will never be resolved. I suggest we try to hack together an "ad hoc" solution, using input from this now fairly-well supported browser measurement:
http://www.quirksmode.org/blog/archives/2012/06/devicepixelrati.html - my tests today indicate that window.devicePixelRatio is supported on Firefox and Chrome which should be enough coverage. We may run into some problems on retina-resolution devices but I don't think we've done any testing on those anyway. -
Justin Obara commented
2017-09-26T10:55:55.198-0400 Tried using Chrome 61 and Safari 11 on macOS 10.12.6 and couldn't reproduce the issue.