I don't disagree with the statement "you shouldn't let the renderer crash the browser" but I'm trying to understand if a null window ref is an expected value for those functions to handle.
If so, the Windows version should explicitly check rather than relying on the fact that the Win32 APIs happen to handle that case by accident. If not, we need to figure out why those null window refs are coming through. Is there anyone who knows what's going on here and can answer this question? Avi /who just wrote this into code in http://codereview.chromium.org/47002 On Thu, Mar 19, 2009 at 1:46 PM, Craig Schlenter <craig.schlen...@gmail.com>wrote: > Hi Avi > > When I did the original change, that function wasn't being called with > a null window. > Clicking on a link in gmail opened the link in a new window as I > recall. At some later > point that changed possibly when some of the tab_contents stuff was hooked > up. I > think it's good practice to check for null since you don't want the > renderer to be able > to crash the browser but I do tend to think that it shouldn't be > happening to begin with > but I'm largely clueless about the code involved :( > > Perhaps someone with a windows build can put a breakpoint in > OnGetRootWindowRect please and see if clicking on a link in an > email in gmail passes a HWND of null at all? > > Thank you, > > --Craig > > On Thu, Mar 19, 2009 at 7:33 PM, Avi Drissman <a...@google.com> wrote: > > We've been seeing calls to ResourceMessageFilter::OnGet(Root)WindowRect > for > > NULL windows. agl put in a fix for GTK with > > http://codereview.chromium.org/42356 and I'm seeing the same problem on > the > > Mac. > > > > 1. Why isn't Windows seeing this? What happens when you pass a null HWND > > into ::GetAncestor and ::GetWindowRect? > > 2. Is this expected, or is this indicative of a bug? > > > > Avi > > > --~--~---------~--~----~------------~-------~--~----~ Chromium Developers mailing list: chromium-dev@googlegroups.com View archives, change email options, or unsubscribe: http://groups.google.com/group/chromium-dev -~----------~----~----~----~------~----~------~--~---
