screenshot disabling policy

chrome/browser/ui/views/ash/screenshot_taker.cc

Index: chrome/browser/ui/views/ash/screenshot_taker.cc
diff --git a/chrome/browser/ui/views/ash/screenshot_taker.cc b/chrome/browser/ui/views/ash/screenshot_taker.cc
index 9549e2c964b6939ee9d18296c559c61aa3bc6e39..03d4f461c223fb8e4fcfce539427a2b1201517b0 100644
--- a/chrome/browser/ui/views/ash/screenshot_taker.cc
+++ b/chrome/browser/ui/views/ash/screenshot_taker.cc
@@ -18,6 +18,7 @@
 #include "base/memory/ref_counted_memory.h"
 #include "base/stringprintf.h"
 #include "base/time.h"
+#include "chrome/browser/browser_process.h"
 #include "chrome/browser/download/download_prefs.h"
 #include "chrome/browser/prefs/pref_service.h"
 #include "chrome/browser/profiles/profile.h"
@@ -113,6 +114,12 @@ bool GrabWindowSnapshot(aura::Window* window,
                         const gfx::Rect& snapshot_bounds,
                         std::vector<unsigned char>* png_data) {
 #if defined(OS_LINUX)
+  // browser::GrabWindowSnapshot checks this too, but RootWindow::GrabSnapshot
+  // does not. The statement below is only to support linux-specific XGetImage
+  // optimization.
+  if (g_browser_process->local_state()->GetBoolean(prefs::kDisableScreenshots))
+    return false;

[sky, 2012/07/17 22:05:57]

This is going to result in a LOG(ERROR) on line 180ish, which I don't think
makes sense here. Maybe you should put this check in HandleTakePartialScreenshot
so that we early out there.

[qfel, 2012/07/18 09:24:27]

In any case, I think we should leave GrabWindowSnapshot as is (safer, policy
checked in one place), and eventually do redundant checks elsewhere
(HandleTakePartialScreenshot) for nicer log output.
Added check in HandleTakePartialScreenshot.

+
   // We use XGetImage() for Linux/ChromeOS for performance reasons.
   // See crbug.com/119492
   // TODO(mukai): remove this when the performance issue has been fixed.