Open Bug 1302822 Opened 8 years ago Updated 2 years ago

[e10s] Popup Window - innerHeight and innerWidth return 0

Categories

(Core :: DOM: CSS Object Model, defect, P3)

defect

Tracking

()

Tracking Status
e10s + ---

People

(Reporter: beberveiga, Unassigned)

References

Details

(Keywords: nightly-community)

Attachments

(2 files)

Attached file 1.html
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:51.0) Gecko/20100101 Firefox/51.0
Build ID: 20160914030200

Steps to reproduce:

- Open 1.html (attachment) in Firefox Nightly (51.0a1).
- Open your console.
- Click the "Open" button.


Actual results:

popup.innerHeight and popup.innerWidth returns 0. Console output: "Object { innerHeight: 0, innerWidth: 0 }".


Expected results:

popup.innerHeight should be 600 and popup.innerWidth should be 800. Expected console output: "Object { innerHeight: 600, innerWidth: 800 }".
Whiteboard: [nightly-community]
Attached file 2.html
Getting innerHeight and innerWidth with a window.setTimeout call is a possible workaround. Notice that even when you pass 0 or null the problem does not occurs.

This bug is not present in Firefox ESR 45.3.0.
It's an issue with e10s, if you disable e10s in Nightly, it works as expected. The bug is probably here since the first steps of e10s in Firefox.
Blocks: e10s
tracking-e10s: --- → ?
Component: Untriaged → DOM: Core & HTML
Product: Firefox → Core
Summary: Popup Window - innerHeight and innerWidth return 0 → [e10s] Popup Window - innerHeight and innerWidth return 0
The page here is trying to retrieve the width and height immediately after calling window.open. In e10s we don't pass this size to the child process, but the parent has this inner width and height by the end of nsWindowWatcher::OpenWindowWithTabParent so we could perhaps pass this down to the child in some manner.

A workaround is to wait for a load event before getting the size.

Note that Chrome has the same issue.
So ... INVALID?
Flags: needinfo?(enndeakin)
I don't think it is invalid. This could be fixed as I described above. It might be wontfix, but I don't feel confident making that call myself.
Flags: needinfo?(enndeakin)
(In reply to Neil Deakin (away until Oct 3) from comment #5)
> I don't think it is invalid. This could be fixed as I described above. It
> might be wontfix, but I don't feel confident making that call myself.

Is there anything we want to do here?
Flags: needinfo?(bugs)
We have several people who aren't sure what call to make on this. Neil suggested someone from Layout. Perhaps given the feedback above, smaug can help make the call on the approach (or whether we just wontfix).
Not really layout, but whoever deals with window.open.
This does sound like a bug to fix, but given some other browsers have the same bug, perhaps it isn't urgent.

mconley?
Flags: needinfo?(bugs) → needinfo?(mconley)
This is not urgent, though a fix would be nice. I'd put this in the "P3" bucket of the new priority system.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Flags: needinfo?(mconley)
Priority: -- → P3
Whiteboard: [nightly-community]
Component: DOM: Core & HTML → DOM: CSS Object Model
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: