From d74b9e518b7448910ac92fa45b53ba8180af7b8c Mon Sep 17 00:00:00 2001
From: Ben Pope <ben@vectorized.io>
Date: Fri, 28 Jan 2022 21:07:10 +0000
Subject: [PATCH] schema_registry/proto: Support encoded proto

Support decoding a protobuf FileDescriptor as encoded protobuf.

Fix #3633

Signed-off-by: Ben Pope <ben@vectorized.io>
---
 src/v/pandaproxy/schema_registry/protobuf.cc | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/src/v/pandaproxy/schema_registry/protobuf.cc b/src/v/pandaproxy/schema_registry/protobuf.cc
index bc41c83d4d66..e5f381abce5b 100644
--- a/src/v/pandaproxy/schema_registry/protobuf.cc
+++ b/src/v/pandaproxy/schema_registry/protobuf.cc
@@ -14,6 +14,7 @@
 #include "pandaproxy/logger.h"
 #include "pandaproxy/schema_registry/errors.h"
 #include "pandaproxy/schema_registry/sharded_store.h"
+#include "utils/base64.h"
 #include "vlog.h"
 
 #include <seastar/core/coroutine.hh>
@@ -21,6 +22,7 @@
 #include <fmt/ostream.h>
 #include <google/protobuf/compiler/parser.h>
 #include <google/protobuf/descriptor.h>
+#include <google/protobuf/descriptor.pb.h>
 #include <google/protobuf/io/tokenizer.h>
 #include <google/protobuf/io/zero_copy_stream.h>
 
@@ -136,10 +138,19 @@ class parser {
         pb::io::Tokenizer t{&is, &error_collector};
         _parser.RecordErrorsTo(&error_collector);
 
+        // Attempt parse a .proto file
         if (!_parser.Parse(&t, &_fdp)) {
-            throw as_exception(error_collector.error());
+            // base64 decode the schema
+            std::string_view b64_def{
+              schema.def().raw()().data(), schema.def().raw()().size()};
+            auto bytes_def = base64_to_bytes(b64_def);
+
+            // Attempt parse as an encoded FileDescriptorProto.pb
+            if (!_fdp.ParseFromArray(
+                  bytes_def.data(), static_cast<int>(bytes_def.size()))) {
+                throw as_exception(error_collector.error());
+            }
         }
-
         _fdp.set_name(schema.sub()());
         return _fdp;
     }