Change GetProfileToUse method from Profile to content::BrowserContext

chrome/browser/profiles/incognito_helpers.h

+// Copyright 2013 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#ifndef CHROME_BROWSER_PROFILES_INCOGNITO_HELPERS_H_
+#define CHROME_BROWSER_PROFILES_INCOGNITO_HELPERS_H_
+
+namespace content {
+class BrowserContext;
+}
+
+namespace chrome {
+
+// Returns the original browser context even for Incognito contexts.
+content::BrowserContext* GetBrowserContextRedirectedInIncognito(
+    content::BrowserContext* context);
+
+// Returns non-NULL even for Incognito contexts so that a separate
+// instance of a service is created for the Incognito context.
+content::BrowserContext* GetBrowserContextOwnInstanceInIncognito(

[Jói, 2013/05/03 17:45:14]

Seeing as this is basically a no-op, is it even needed?

[Paweł Hajdan Jr., 2013/05/03 18:08:19]

Yes - mostly to make code readable and obvious.

It looks like a no-op but actually _does_ a thing (and in a pretty sensitive
area, i.e. Incognito mode). 

Without it the service would not be created in Incognito mode. With this helper
used, it does get created. This is not obvious from just reading the code (as
your comment above confirms), and it's used in many places throughout the
codebase (where it's even easier to miss the context).

Also, making this a function makes it more greppable, which IMO is also very
important.

[Jói, 2013/05/03 18:54:09]

Hmm, it just returns the parameter passed to it.  Are there any side effects to
that?  If I'm missing something please document.

Regardless, I don't mind if you keep the function, it serves as documentation
for what's happening in the individual PKSs.

+    content::BrowserContext* context);
+
+}  // namespace chrome
+
+#endif  // CHROME_BROWSER_PROFILES_INCOGNITO_HELPERS_H_