[Webkit-unassigned] [Bug 19392] [Gtk] Enable WebInspector in the Gtk port

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=19392





------- Comment #9 from jmalonzo at gmail.com  2008-07-15 04:29 PDT -------
Hi Christian,

(In reply to comment #8)
> >+    /**
> >+    * WebKitWebSettings:enable-developer-extras:
> >+    *
> >+    * Whether developer extensions should be enabled.
> >+    *
> >+    * Since: 1.0.2
> >+    */
> >+    g_object_class_install_property(gobject_class,
> >+                                    PROP_ENABLE_DEVELOPER_EXTRAS,
> >+                                    g_param_spec_boolean(
> >+                                    "enable-developer-extras",
> >+                                    "Enable Developer Extras",
> >+                                    "Enables special extensions that help developers",
> >+                                    FALSE,
> >+                                    flags));
> 
> That property name feels a little awkward. Would enable-debug do it as well?

Mac and Win call this "developer extras". Calling this enable-debug or
diverging for that matter would be confusing especially if we are going to use
the Mac and Win ports as our reference implementation.

> >+    /**
> >+     * WebKitWebView::create-inspector-web-view:
> >+     * @web_view: the object on which the signal is emitted
> >+     * @return: a newly allocated #WebKitWebView or %NULL
> >+     *
> >+     * Emitted when the creation of a new window for the Web Inspector
> >+     * is requested.  If this signal is handled the signal handler
> >+     * should return the newly created #WebKitWebView.
> >+     *
> >+     * The signal handlers should not try to deal with the reference
> >+     * count for the new #WebKitWebView. The widget to which the
> >+     * widget is added will handle that.
> >+     */
> 
> > * create-inspector-web-view doesn't seem right. How about inspect-web-view
> > because isn't that what we're doing here?
> 
> I tend to disagree. There is a web view being created, and the name should
> reflect that. 'inspect-web-view' looks like something that operates on an
> existing inspector. However I wonder if the name can be a little shorter, for
> instance would create-inspector be a good name? On the other hand, it might be
> less clear, not sure about that.

To me "inspect-web-view" means we are inspecting a/the web-view. I agree that
there is a web view being created, but that web view is "special" because
rather than displaying normal content like most web views, it displays
information about a web view's content. 

Gtk events, for example, means something happened (a widget was clicked,
toggled, about to be destroyed, etc..). We should do the same thing here (e.g.
a web view is about to be inspected, whatever). Creating an inspector is less
clear as by itself, an inspector is really not stand-alone but is associated
with/inspecting a web view.

Conceptually, there should be difference between a WebView and a WebInspector
(which is a web view + other things that developers care about).

> > Apart from that looks good. (Not sure if we should have a WebKitWebInspector
> > class though. Mac and win ports have it so maybe we should mimic one as well) 
> 
> What is the purpose of a WebKitWebInspector class here? Right now you can load
> the inspector into an arbitrary web view, including a sub class. But that won't
> work if the inspector is a different class.
> I'd like to see an explanation of why we need another class.
> 

I'm not sure about this either but one thing that comes to mind is testability
of the inspector. Perhaps when there's a use case we can add it or when we have
actual tests.

Cheers.


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.