Bug 55237 – Web Inspector: extract content loading logic from scripts panel

• Status: RESOLVED
• Resolution: FIXED
• Component: Web Inspector (Deprecated)

[Pavel Podivilov]

Web Inspector: extract content loading logic from scripts panel.

We need to encapsulate source frame content loading logic in debugger presentation model to support beautification, de-obfuscation and source maps in a pluggable way.

[Pavel Podivilov]

Created attachment 83843 [details]
Patch.

[Pavel Feldman]

Comment on attachment 83843 [details]
Patch.

View in context: https://bugs.webkit.org/attachment.cgi?id=83843&action=review

This change looks scary. Could you split it into number of changes?

> Source/WebCore/ChangeLog:950
> +        to -1. Since WebKit2 doesn�t support non-activating clicks yet (<http://webkit.org/b/55053>

Is this change intentional?

> Source/WebCore/inspector/front-end/DebuggerModel.js:47
>      BreakpointAdded: "breakpoint-added",

I thought you were going to get rid of this one?

> Source/WebCore/inspector/front-end/DebuggerModel.js:50
> +    InspectedURLChanged: "inspected-url-changed"

It is not clear why this event should be fired on debugger model.

[Pavel Podivilov]

Created attachment 85467 [details]
Patch.

[Pavel Podivilov]

(In reply to comment #2)
> (From update of attachment 83843 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=83843&action=review
> 
> This change looks scary. Could you split it into number of changes?
> 
> > Source/WebCore/ChangeLog:950
> > +        to -1. Since WebKit2 doesn�t support non-activating clicks yet (<http://webkit.org/b/55053>
> 
> Is this change intentional?
> 
> > Source/WebCore/inspector/front-end/DebuggerModel.js:47
> >      BreakpointAdded: "breakpoint-added",
> 
> I thought you were going to get rid of this one?
> 
> > Source/WebCore/inspector/front-end/DebuggerModel.js:50
> > +    InspectedURLChanged: "inspected-url-changed"
> 
> It is not clear why this event should be fired on debugger model.

Done.

[Pavel Podivilov]

Created attachment 85681 [details]
Rebase.

[Pavel Podivilov]

Created attachment 85926 [details]
Patch.

[Pavel Feldman]

Comment on attachment 85926 [details]
Patch.

View in context: https://bugs.webkit.org/attachment.cgi?id=85926&action=review

> Source/WebCore/inspector/front-end/DebuggerPresentationModel.js:315
> +WebInspector.SourceFileProvider.prototype = {

So SourceFileProvider has a fairly strange interface:
- addScript
- ensureSourceFileAdded
- sourceFile
- requestSourceFileContent
- scriptLocationToSourceLocation

what is it really responsible for?

[Pavel Podivilov]

(In reply to comment #7)
> (From update of attachment 85926 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=85926&action=review
> 
> > Source/WebCore/inspector/front-end/DebuggerPresentationModel.js:315
> > +WebInspector.SourceFileProvider.prototype = {
> 
> So SourceFileProvider has a fairly strange interface:
> - addScript
> - ensureSourceFileAdded
> - sourceFile
> - requestSourceFileContent
> - scriptLocationToSourceLocation
> 
> what is it really responsible for?

It's a private DPM utility class responsible for mapping VM scripts to UI source files. I agree that interface looks strange, but I need an interface like this  to hide scripts <-> sources mapping strategy from DPM.

[Pavel Podivilov]

Committed r81369: <http://trac.webkit.org/changeset/81369>

[WebKit Review Bot]

http://trac.webkit.org/changeset/81369 might have broken SnowLeopard Intel Release (Tests)
The following tests are not passing:
inspector/debugger/debug-inlined-scripts.html

[Pavel Podivilov]

Committed r81467: <http://trac.webkit.org/changeset/81467>