[Webkit-unassigned] [Bug 191345] New: Web Inspector: Start moving toward better multi-target support

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Nov 6 20:22:30 PST 2018


            Bug ID: 191345
           Summary: Web Inspector: Start moving toward better multi-target
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: All
                OS: All
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Web Inspector
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: joepeck at webkit.org
                CC: inspector-bugzilla-changes at group.apple.com

Start moving toward better multi-target support

A world where we do:

    if (PageAgent.foo)

Is ambiguous in a multi-target world:

  • Which target is the `PageAgent.foo` command going to go to?
  • Which target is the `PageAgent.foo` feature check happening on? Is that important?

Ultimately, the behavior that we would like to have is to always do:

  • target.PageAgent

Then it is clear exactly which target we are operating on.

Another possibility is that we don't care or have a Target, we just want to "feature check" (common for enums as well)

    if (FooAgent.hasEvent("someEvent"))
       ... support something else ...

In these cases, we should have an agnostic way to just get the complete list of whatever the backend supports:


The key part being:



For now we have hundreds of places in our code that are using unprefixed or window prefixed agents. As we bring up multi-target support we should start eliminating these. Currently the behavior is they use the "Main Target Connection". That ends up being fine right now as the main target is limited to being a Page or ServiceWorker debuggable. But as we move toward supporting multiple such targets each unprefixed access to "PageAgent", "DOMAgent", "NetworkAgent", etc will need to be explicit per-target.

You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20181107/e6035470/attachment.html>

More information about the webkit-unassigned mailing list