Bug 237403
| Summary: | Can't transfer an ImageBitmap through a MessageChannel | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Kaiido <tristan.fraipont> |
| Component: | Canvas | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED CONFIGURATION CHANGED | ||
| Severity: | Normal | CC: | ahmad.saleem792, cdumez, dino, heycam, karlcow, kkinnunen, sabouhallawa, simon.fraser, webkit-bug-importer |
| Priority: | P2 | Keywords: | InRadar |
| Version: | Safari Technology Preview | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Kaiido
Transferring an ImageBitmap through the ports of a MessageChannel makes the `data` of the MessageEvent `null`.
This doesn't happen when only cloning the ImageBitmap instead of transferring it. This also doesn't happen when transferring from a Worker's `postMessage` (this is being tested by WPT), nor when transferred from an WindowProxy, i.e this only concerns MessageChannel.
You can see a minimal repro at https://jsfiddle.net/1kzxvad9/
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Radar WebKit Bug Importer
<rdar://problem/90072516>
Ahmad Saleem
*** Safari 26.2 Beta ***
"clone", [object ImageBitmap] {
close: function close() {
[native code]
},
height: 20,
width: 20
}
"transfer", [object ImageBitmap] {
close: function close() {
[native code]
},
height: 20,
width: 20
}
*** Chrome Canary 144 ***
"clone", [object ImageBitmap] {
close: function close() { [native code] },
height: 20,
width: 20
}
"transfer", [object ImageBitmap] {
close: function close() { [native code] },
height: 20,
width: 20
}
*** Firefox Nightly 148 ***
clone", [object ImageBitmap] {
close: function close() {
[native code]
},
height: 20,
width: 20
}
"transfer", [object ImageBitmap] {
close: function close() {
[native code]
},
height: 20,
width: 20
}
_________
All browsers now match, It might have been fixed while it was reported. @Karl - anything I missed, or we can mark as `Config Changed`?
Kaiido
It probably has been fixed, indeed.
Ahmad Saleem
(In reply to Kaiido from comment #3)
> It probably has been fixed, indeed.
Thanks for confirming.