Bug 106371 – Web Inspector: Allow Page.navigate to open file:// urls

• Component: Web Inspector (Deprecated)
• Status: RESOLVED WONTFIX

[Ken Kania]

We would like to be able to navigate to local files for automation/testing via Inspector. Currently, you'll get this error:

Not allowed to load local resource: file://...

We would need to change InspectorPageAgent::navigate to use FrameLoader::load instead of FrameLoader::changeLocation, which seems to make more sense anyways.

[Pavel Feldman]

Created attachment 182736 [details]
Patch

[WebKit Review Bot]

Please wait for approval from [e-mail], [e-mail], [e-mail], [e-mail] or [e-mail] before submitting, as this patch contains changes to the Chromium public API. See also https://trac.webkit.org/wiki/ChromiumWebKitAPI.

[Yury Semikhatsky]

Comment on attachment 182736 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=182736&action=review

> Source/WebCore/ChangeLog:8
> +        Provided a way for embedder to handle Page.navigate.

Please describe here why we cannot use changeLocation on the main frame loader and need this method.

> Source/WebKit/chromium/src/InspectorClientImpl.h:67
> +    virtual bool canClearBrowserCookies() { return true; }

What's the purpose of moving this out of the .cpp file?

[Yury Semikhatsky]

Comment on attachment 182736 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=182736&action=review

> Source/WebKit/chromium/src/InspectorClientImpl.h:69
> +    virtual bool canNavigate() { return true; }

This should return false while chromium implementation is not ready.

[Pavel Feldman]

(In reply to comment #3)
> (From update of attachment 182736 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=182736&action=review
> 
> > Source/WebCore/ChangeLog:8
> > +        Provided a way for embedder to handle Page.navigate.
> 
> Please describe here why we cannot use changeLocation on the main frame loader and need this method.
> 

Will do.

> > Source/WebKit/chromium/src/InspectorClientImpl.h:67
> > +    virtual bool canClearBrowserCookies() { return true; }
> 
> What's the purpose of moving this out of the .cpp file?

Aesthetic?

(In reply to comment #4)
> (From update of attachment 182736 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=182736&action=review
> 
> > Source/WebKit/chromium/src/InspectorClientImpl.h:69
> > +    virtual bool canNavigate() { return true; }
> 
> This should return false while chromium implementation is not ready.

I'll land this one afterwards.

[Build Bot]

Comment on attachment 182736 [details]
Patch

Attachment 182736 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/15869894

[Tony Gentilcore]

Is this still on track to land? This would be useful for Telemetry too.

[Ken Kania]

(In reply to comment #7)
> Is this still on track to land? This would be useful for Telemetry too.

I don't think so. You might be interested in http://src.chromium.org/viewvc/chrome?view=rev&revision=189102 though.

[Tony Gentilcore]

(In reply to comment #8)
> (In reply to comment #7)
> > Is this still on track to land? This would be useful for Telemetry too.
> 
> I don't think so. You might be interested in http://src.chromium.org/viewvc/chrome?view=rev&revision=189102 though.

Thanks. Pavel just pointed that out to me too. I verified that file:// URLs work with page.Navigate at ToT chromium.