Bug 277856
| Summary: | [WPE] REGRESSION(281870@main) Float16Array undefined on HTTP inspector | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Miguel Gomez <magomez> |
| Component: | WebKitGTK | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED INVALID | ||
| Severity: | Normal | CC: | bugs-noreply, cgarcia, mcatanzaro, yijia_huang, ysuzuki |
| Priority: | P2 | ||
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Miguel Gomez
Inspector seems to be broken both on GTK and WPE.
On GTK, opening a page and clicking inspect doesn't show anything, and this error is shown:
RemoteInspector failed to connect to inspector server at: 0.0.0.0:34575: GDBus.Error:org.freedesktop.portal.Error.NotAllowed: This call is not available inside the sandbox
Disabling the sandbox removes that error, but nothing is shown anyway.
The same happens when trying to use the non HTTP inspector on WPE.
When trying to use the HTTP inspector on WPE, it shows the error:
Uncaught ReferenceError: Float16Array is not defined (at Utilities.js:636:32)
which initially points to https://commits.webkit.org/281870@main, but I haven't been able to verify if the problem comes from there or some subsequent patch related to that.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Miguel Gomez
After some investigation, I found there are 2 different errors with the inspector.
The first one is related to the normal inspector on GTK, whose window doesn't show at all. Clicking on the inspect option doesn't do anything.
I bisected this problem to https://commits.webkit.org/281966@main.
The second one is related with the HTTP inspector that exists only on WPE. By using this version, the previous error doesn't show, but we get instead the one related to the undefined Float16Array.
It seems that the Float16Array definition is missing from the WebInspectorUI for some reason, and I've confirmed that this was introduced by https://commits.webkit.org/281870@main.
I'll use this bug to track the second issue, and open a new one to track the first one.
Michael Catanzaro
(In reply to Miguel Gomez from comment #1)
> I'll use this bug to track the second issue, and open a new one to track the
> first one.
Bug #278015
Miguel Gomez
I've just tested that this is not really a WebKit bug, but a consequence of the browser running the inspector not supporting the Float16Array type. A minibrowser from a recent main that has the Float16Array commit can perfectly use the HTTP inspector. Closing this as invalid, as it wasn't really a bug. Sorry for the noise.