Add Mojo IPC for seeding new Renderer with Browser's cached blob state.

blimp/engine/renderer/blob_channel_sender_proxy.cc

Index: blimp/engine/renderer/blob_channel_sender_proxy.cc
diff --git a/blimp/engine/renderer/blob_channel_sender_proxy.cc b/blimp/engine/renderer/blob_channel_sender_proxy.cc
index efdfeae668bcd02cc3e329944e7a495559337843..6577da68741af76eda2ddf985cb8c4edfeb98d40 100644
--- a/blimp/engine/renderer/blob_channel_sender_proxy.cc
+++ b/blimp/engine/renderer/blob_channel_sender_proxy.cc
@@ -22,7 +22,13 @@ mojom::BlobChannelPtr GetConnectedBlobChannel() {
 }  // namespace
 
 BlobChannelSenderProxy::BlobChannelSenderProxy()
-    : blob_channel_(GetConnectedBlobChannel()) {}
+    : blob_channel_(GetConnectedBlobChannel()), weak_factory_(this) {
+  // TODO(kmarshall): Switch to observer push model (vs. the current pull model)
+  // once cache invalidations are introduced.

[Wez, 2016/07/01 00:30:20]

nit: Either refer to the bug for this TODO, or remove it, I think - no need for
a reminder in the code, as such, I don't think?

[nyquist, 2016/07/15 22:13:46]

Yeah, and if there is no bug for it, just file one. Nowadays we should hopefully
also be able to do this by ourselves instead of waiting for fancy functionality
from Skia, pending the big change in how the compositor protocol works.

[Kevin M, 2016/07/18 16:58:16]

Done.

[Kevin M, 2016/07/18 16:58:16]

Done.

+  blob_channel_->GetCachedBlobIds(
+      base::Bind(&BlobChannelSenderProxy::OnGetCacheStateComplete,
+                 weak_factory_.GetWeakPtr()));
+}
 
 BlobChannelSenderProxy::~BlobChannelSenderProxy() {}
 
@@ -52,5 +58,22 @@ void BlobChannelSenderProxy::DeliverBlob(const std::string& id) {
   blob_channel_->DeliverBlob(id);
 }
 
+std::vector<BlobChannelSender::CacheStateEntry>
+BlobChannelSenderProxy::GetCachedBlobIds() const {
+  NOTREACHED();
+  return std::vector<BlobChannelSender::CacheStateEntry>();
+}
+
+void BlobChannelSenderProxy::OnGetCacheStateComplete(
+  mojo::Map<mojo::String, bool> cache_state
+) {

[Wez, 2016/07/01 00:30:21]

nit: This is very strange looking line-wrapping! Is this what git cl format does
now?

[Kevin M, 2016/07/18 16:58:16]

Re-ran git cl format, formatting is saner. I must've neglected to run it before
uploading the last patch.

+  VLOG(1) << "Received cache state from browser (" << cache_state.size()
+          << " items)";
+  replication_state_.clear();
+  for (const auto& next_item : cache_state) {
+    replication_state_[next_item.first] = next_item.second;
+  }
+}
+
 }  // namespace engine
 }  // namespace blimp