[Webkit-unassigned] [Bug 113331] New: Web Inspector: Encapsulate SetEmbedderData/GetEmbedderData

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=113331

           Summary: Web Inspector: Encapsulate
                    SetEmbedderData/GetEmbedderData
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: All
        OS/Version: All
            Status: UNCONFIRMED
          Severity: Normal
          Priority: P2
         Component: Web Inspector
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: johnjbarton at chromium.org
                CC: keishi at webkit.org, pmuellr at yahoo.com,
                    pfeldman at chromium.org, yurys at chromium.org,
                    apavlov at chromium.org, loislo at chromium.org,
                    jochen at chromium.org, vsevik at chromium.org,
                    haraken at chromium.org,
                    web-inspector-bugs at googlegroups.com


We currently have two different sets of V8 embedder data.

1) V8PerContextData. This data is stored as a pointer on the context via SetAlignedPointerInEmbedderData.

2) A string like "page,232", stored on the context via SetEmbedderData, and used by the Inspector support binding code.

Confusingly both systems index into the *same array* on the Context. Thus if you search the code base for SetEmbedderData it will appear that only index 0 is used.  Using index 1 or 2 in SetEmbedder data leads to mysterious crashes, because in fact these indexes are used by the V8PerContextData and friends.

The overall system would be less confusing and fragile of all uses this underlying array where together. 

(Using the V8PerContextData to store the Inspector value could be an extra step, but mixing the two kinds of data would be more confusing in my opinion).

Patch ready.

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