392359 - Some values of pointer-events not working

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[tor]

Attachment: force hit testing when needed

The addition of containing rects to avoid unnecessary work broke some values of pointer-events in some circumstances, because the rect will not get initialized.

See test cases on bug 347374 and this portion of the spec:

  http://www.w3.org/TR/SVG11/interact.html#PointerEventsProperty

Comment 1

[Robert Longson [:longsonr]]

Comment on attachment 276811 [details] [diff] [review]
force hit testing when needed

> #define HITTEST_MASK_FILL 1
> #define HITTEST_MASK_STROKE 2
>+#define HITTEST_FORCE_TEST 3

Shouldn't that be
#define HITTEST_FORCE_TEST 4

Although I'd rather see

#define HITTEST_MASK_FILL    0x01
#define HITTEST_MASK_STROKE  0x02
#define HITTEST_FORCE_TEST   0x04

Then it would be a little clearer these are supposed to form a set of bits

At the moment HITTEST_MASK_FILL | HITTEST_MASK_STROKE  == HITTEST_FORCE_TEST

+      mask |=
+        HITTEST_MASK_FILL |
+        HITTEST_MASK_STROKE |
+        HITTEST_FORCE_TEST;

So this is mask |= HITTEST_FORCE_TEST for instance although I don't think you meant that.

Comment 2

[Robert Longson [:longsonr]]

In case I wasn't clear I think the only thing wrong is the #define value

Comment 3

[tor]

(In reply to comment #1)
> (From update of attachment 276811 [details] [diff] [review])
> > #define HITTEST_MASK_FILL 1
> > #define HITTEST_MASK_STROKE 2
> >+#define HITTEST_FORCE_TEST 3
> 
> Shouldn't that be
> #define HITTEST_FORCE_TEST 4

Gah, how did I end up with that?  I remember thinking when typing that line that it was a mask and I need a power of two.  Gremlins, probably...

Comment 4

[tor]

Attachment: darn gremlins...

Comment 5

[tor]

Comment on attachment 277009 [details] [diff] [review]
darn gremlins...

Low risk specification conformance fix.

Comment 6

[tor]

Checked in.