905122 - SVG getScreenCTM() seems to be broken or at least changed in Firefox 23 on nested SVG.

Status: RESOLVED WORKSFORME

(Core :: SVG, defect)

Description

[Sascha]

User Agent: Mozilla/5.0 (Windows NT 6.2; WOW64; rv:24.0) Gecko/20100101 Firefox/24.0 (Beta/Release)
Build ID: 20130806170643

Steps to reproduce:

try to paint a rectangle:
http://svg-edit.googlecode.com/svn/branches/2.6/editor/svg-editor.html



Actual results:

getScreenCTM() is called on the nested svg#svgcontent to calculate the coords and this returns other values than in Firefox 22 and Chrome.



Expected results:

The rectangle should be painted at the position of the mouse cursor.

Comment 1

[Alice0775 White]

Regression window(m-c)
Good:
http://hg.mozilla.org/mozilla-central/rev/c8e47b184aba
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20130505 Firefox/23.0 ID:20130505000859
Bad:
http://hg.mozilla.org/mozilla-central/rev/b109e2dbf03b
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20130505 Firefox/23.0 ID:20130505171547
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=c8e47b184aba&tochange=b109e2dbf03b

Regression window(m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/87ee7b242241
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20130505 Firefox/23.0 ID:20130505000858
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/745a6846cf43
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20130505 Firefox/23.0 ID:20130505002546
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=87ee7b242241&tochange=745a6846cf43
Regressed by:
b8b048e24dcd	Robert Longson — Bug 863994 - Support transform attribute on <svg> elements. r=dholbert

Comment 2

[Robert Longson [:longsonr]]

Attachment: testcase

It changed, and as far as I can see it's now correct. Try the testcase in Opera and it gives 1 0 0 1 **1** 0 for the first case but the markup is the same for all elements. The elements co-ordinates should not be included in getScreenCTM.

Comment 3

[Robert Longson [:longsonr]]

If you want the x and y co-ordinates in the CTM then create a child element and get the screenCTM of that.

Comment 4

[Jonathan Watt [:jwatt]]

While I agree that the new behavior is better in terms of consistency with the "from user space" concept, the old, unpleasant inclusion of the x/y values is at least interoperable across browsers. We should consider what we can do to go back to the old behavior I suspect.

huebner, in the meantime you probably want to change svg-edit to use Robert's suggestion from comment 3. Calling getScreenCTM on a child <g> should give the same result in all browsers as the old behavior of calling it on the <svg>.

Comment 5

[Jonathan Watt [:jwatt]]

I'll hold off requesting tracking for now to see if we get any more reports.

Comment 6

[Robert Longson [:longsonr]]

Looks like svg-edit will be fixing things on their end, perhaps we can just leave things as they are in that case.

Comment 8

[Robert Longson [:longsonr]]

svg-edit is fixed now so perhaps we can mark this invalid (again).

Comment 9

[Alice0775 White]

(In reply to Robert Longson from comment #8)
> svg-edit is fixed now so perhaps we can mark this invalid (again).

confirmed.
It works.
http://hg.mozilla.org/releases/mozilla-esr24/rev/d06a17a96fa2
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:24.0) Gecko/20100101 Firefox/24.0 ID:20140131092626
https://hg.mozilla.org/mozilla-central/rev/31113754db3b
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:30.0) Gecko/20100101 Firefox/30.0 ID:20140224030203