Two different full-screen behaviors when privacy.resistFingerprinting.letterboxing = true
Categories
(Core :: DOM: Security, defect, P3)
Tracking
()
Tracking | Status | |
---|---|---|
firefox75 | --- | affected |
People
(Reporter: geeknik, Unassigned)
References
Details
(Whiteboard: [fingerprinting][domsecurity-backlog])
Hi. I noticed there are 2 different fullscreen behaviors in Firefox Nightly (ASAN Build ID 20200305095541) on Fedora 31 with xfce4, Nvidia RTX 2070 and v440.64 drivers, but only when you have enabled privacy.resistFingerprinting
and privacy.resistFingerprinting.letterboxing
.
STR:
- Set privacy.resistFingerprinting and privacy.resistFingerprinting.letterboxing to true. Restart as necessary.
- Visit YouTube and start playing any video.
- Tap F11 on the keyboard.
- Notice the letterboxing around the fullscreen video.
- Tap F11 to leave full screen, then tap the full screen icon in the YouTube player.
- Notice the lack of letterboxing.
YouTube's Stats for nerds
reports a viewport of 1477x831 for Theater mode, 1655x931 for F11, and 1920x1080 for YouTube Fullscreen. I'm on a 24" monitor with a native resolution of 1920x1200.
Comment 1•5 years ago
|
||
(In reply to Brian Carpenter [:geeknik] from comment #0)
- Tap F11 to leave full screen, then tap the full screen icon in the YouTube player.
- Notice the lack of letterboxing.
Letterboxing does not apply to elements
- https://ghacksuserjs.github.io/TorZillaPrint/TorZillaPrint.html#screen <--
click here
on the full screen test and it will return two measurements: screen and an element that uses all available space
Tor: https://trac.torproject.org/projects/tor/ticket/32713
tom: duplicate of bug 1450401 ?
I didn't bother to look at the stats for nerds
or how they calculate those values, but 1920x1080 is 16/9 exactly and the other two are as close as possible to 16/9 (whereas your screen is actually 16/10) - so I suspect they're reporting possible video dimensions, not actual viewport/inner/screen measurements
Updated•5 years ago
|
Updated•5 years ago
|
Updated•5 years ago
|
Comment 2•4 years ago
|
||
Seems pretty close to a dupe, but I'm going to leave it open as something to double check if we get into a position to double check it.
Updated•2 years ago
|
Description
•