Move client certificates retrieval logic out of the SSL sockets.

content/browser/ssl/ssl_error_handler.h

Index: content/browser/ssl/ssl_error_handler.h
diff --git a/content/browser/ssl/ssl_error_handler.h b/content/browser/ssl/ssl_error_handler.h
index 8fcfeae0129e3bf4a51f2eb9cd556a61562551e5..aa301ae7130ac1b0809672d40d0ba88a08ed849c 100644
--- a/content/browser/ssl/ssl_error_handler.h
+++ b/content/browser/ssl/ssl_error_handler.h
@@ -47,7 +47,7 @@ class SSLErrorHandler : public base::RefCountedThreadSafe<SSLErrorHandler> {
   // SSLManager::OnSSLCertificateError() and represents the request.
   // Finally, CancelSSLRequest() or ContinueSSLRequest() will be called after
   // SSLErrorHandler makes a decision on the SSL error.
-  class Delegate {
+  class CONTENT_EXPORT Delegate {

[Ryan Sleevi, 2013/02/04 23:08:33]

Why is this necessary? Shouldn't this be CONTENT_EXPORT_PRIVATE, if for testing?

[ppi, 2013/02/05 17:05:30]

This is a part of ancestor hierarchy for ResourceLoader. We need to export all
ancestors or suppress the warnings to get the code compiling (at least on
Windows).

We do not have a CONTENT_EXPORT_PRIVATE macro. The assumption seems to be that
everything that is exported, but is not in content/public is exported for
testing. See also the discussion here:
https://chromiumcodereview.appspot.com/10735028/ .

Introducing CONTENT_EXPORT_PRIVATE seems like a good idea to me, but I am not
sure that it should be done in this cl.

    public:
     // Called when SSLErrorHandler decides to cancel the request because of
     // the SSL error.