Closed Bug 1852864 Opened 2 years ago Closed 1 year ago

Dragging image on Linux debug build triggers assertion

Categories

(Core :: DOM: Copy & Paste and Drag & Drop, defect)

Firefox 118
Desktop
Linux
defect

Tracking

()

RESOLVED DUPLICATE of bug 1850213
Tracking Status
firefox-esr102 --- unaffected
firefox-esr115 --- unaffected
firefox117 --- unaffected
firefox118 --- wontfix
firefox119 --- wontfix
firefox120 --- wontfix
firefox121 --- wontfix
firefox122 --- fixed

People

(Reporter: aosmond, Unassigned)

References

(Regression)

Details

(Keywords: regression)

+++ This bug was initially created as a clone of Bug #1852420 +++

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/118.0

STR:

  1. Open https://codepen.io/tkuri2010/pen/bGOqojL
  2. Open another window with an image
  3. Drag image into the codepen
  4. Assert trips

Specifically I get this, where mEvent->mCurrentTarget is the HTMLCanvasElement:
https://searchfox.org/mozilla-central/rev/09f7e00da9eaef55c80620c17e8575ddd4532451/dom/events/Event.cpp#457

Not sure if this can be reproduced on other platforms, but it happens every time for me. Looks like the assert was added by bug 1848409. Commenting out the assert seems to work, as far as the test case goes (with my patch in bug 1852420).

Flags: needinfo?(sefeng)

Set release status flags based on info from the regressing bug 1848409

Given this is an debugging assertion, and the intention of the assertion was to learn if that can be a case and everything seems to work fine without the assertion, so I don't think this needs to be fixed in 118. Though I am trying to fix it and I think likely I just need to remove the assertion.

Andrew I still can't repro this and I am on Linux. When you say open another window, do you mean open a new browser window of the same debugging instance? Or a new tab but in the same window? They are pretty much the same though..And I guess the new tab just behaves like a image viewer? And if you can generate a pernosco session, that helps too.

As I said, I probably should remove the assertion, though I'd like to learn a bit more about this.

Flags: needinfo?(aosmond)

Definitely agree there is no urgency for a fix, especially if the assertion is wrong :).

Here is a screen recording I did on Ubuntu, in an X11 session:
https://drive.google.com/file/d/1MqpzoayFhAezPaKr4w3DcFKc8XkfTnSL/view?usp=drive_link

I can try to do an rr recording and upload it to pernosco.

Flags: needinfo?(aosmond)

Thanks!

I think the reason of why I can't reproduce is my dragging doesn't really drag the image, it drags the url as the content type is "text/x-moz-url". Not sure why...Maybe you know Andrew?

An rr recording would be extremely helpful!

Set release status flags based on info from the regressing bug 1848409

Too late for Fx118 and Fx119 - in terms of priority, anything expected for Fx120?

My understand is same as bug 1850213. Is this fixed by the latest Nightly?

Flags: needinfo?(aosmond)

Closing this as fixed(dup) by bug 1850213.
aosmond: please feel free to reopen if that is not the case

Status: NEW → RESOLVED
Closed: 1 year ago
Duplicate of bug: 1850213
Resolution: --- → DUPLICATE

Clearing the NIs - the bug was closed several months ago, and I don't see pending actions/questions here.

Flags: needinfo?(sefeng)
Flags: needinfo?(aosmond)
You need to log in before you can comment on or make changes to this bug.