[webkit-reviews] review denied: [Bug 20483] WebKit GTK Port needs a smartpointer to handle g_free (GFreePtr?) : [Attachment 23345] Add and use smart pointers for glib memory management

[bugzilla-daemon at webkit.org]

Eric Seidel <eric at webkit.org> has denied Marco Barisione
<marco.barisione at collabora.co.uk>'s request for review:
Bug 20483: WebKit GTK Port needs a smartpointer to handle g_free (GFreePtr?)
https://bugs.webkit.org/show_bug.cgi?id=20483

Attachment 23345: Add and use smart pointers for glib memory management
https://bugs.webkit.org/attachment.cgi?id=23345&action=edit

------- Additional Comments from Eric Seidel <eric at webkit.org>
ap: I'm not sure it would be possible to use OwnPtr<> for this, since g_free
takes a void* and thus it would be difficult to differentiate when things
should have fastFree() called on them, vs. g_free.

This patch would be much easier to review if it was just GOwnPtr or GRefPtr,
not both.  Or if you were adding one and converting *some* callers, but not
all.

I really don't think I can do a good review of this whole thing as-is.	(Even
though I'm *very* excited about this patch landing for Gtk)

Why are you calling the destructor directly here?
+    priv->~WebKitWebFramePrivate();

And using placement new here?
+    new(priv) WebKitWebFramePrivate;

(btw, we would normally put a space between new and ( -- at least that how
every other placement new in WebCore that I have seen has been written)

I look forward to reviewing this patch in smaller pieces.  Maybe you can find
someone who can handle it in this huge hunk, but that's not me. :(