[webkit-reviews] review denied: [Bug 61016] [WebWorkers][Chromium] Use v8 Isolates for in-process implementation of WebWorkers : [Attachment 103883] This adds an implementation of in-process dedicated workers to chromium port.

[bugzilla-daemon at webkit.org]

Darin Fisher (:fishd, Google) <fishd at chromium.org> has denied Dmitry Lomov
<dslomov at google.com>'s request for review:
Bug 61016: [WebWorkers][Chromium] Use v8 Isolates for in-process implementation
of WebWorkers
https://bugs.webkit.org/show_bug.cgi?id=61016

Attachment 103883: This adds an implementation of in-process dedicated workers
to chromium port. 
https://bugs.webkit.org/attachment.cgi?id=103883&action=review

------- Additional Comments from Darin Fisher (:fishd, Google)
<fishd at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=103883&action=review


I think we can do better than adding WebCommonFooBase interfaces.  Let's try to
just make use of WebCommonFoo, or switch to WebFooBase.

> Source/WebKit/chromium/public/WebCommonWorkerClient.h:53
> +class WebCommonWorkerClient : public WebCommonWorkerClientBase {

WebCommonWorkerClientBase is a funny name.  Isn't WebCommonWorkerClient already

supposed to be the interface that represents the common functions needed by all

types of worker clients?

I see this comment above WebCommonWorkerClient:

// This interface contains common APIs used by both shared and dedicated
// workers.

> Source/WebKit/chromium/public/WebCommonWorkerClientBase.h:56
> +}

WebKit API headers normally put a comment here to indicate the close of the
namespace.

You can see this in part 3 of the coding style guide:
http://www.webkit.org/coding/coding-style.html

> Source/WebKit/chromium/src/WebCommonWorkerBase.h:45
> +    virtual WebView* webView() const = 0;

we normally drop the "web" prefix on method names that return a WebFoo type.

> Source/WebKit/chromium/src/WebCommonWorkerBase.h:47
> +

nit: only one new line here please