Speed up JSON parsing.

It turns out calling StringOutputStream::Next()/BackUp() repeatedly is
very costly in opensource protobuf because it keeps resize() the string
back and forth. The current JSON conversion API suffers this problem and
leads to ridiculously long parsing time:
https://github.com/google/protobuf/issues/2305#issuecomment-257785492

This change fixes the problem but caching the buffer of Next() and avoid
calling BackUp() as much as possible.

2 files changed, 20 insertions, 14 deletions

diff --git a/src/google/protobuf/util/json_util.cc b/src/google/protobuf/util/json_util.cc
index d7ac2dba..5974cc4a 100644
--- a/src/google/protobuf/util/json_util.cc
+++ b/src/google/protobuf/util/json_util.cc
@@ -49,22 +49,25 @@ namespace protobuf {
 namespace util {
 
 namespace internal {
+ZeroCopyStreamByteSink::~ZeroCopyStreamByteSink() {
+  stream_->BackUp(buffer_size_);
+}
+
 void ZeroCopyStreamByteSink::Append(const char* bytes, size_t len) {
-  while (len > 0) {
-    void* buffer;
-    int length;
-    if (!stream_->Next(&buffer, &length)) {
-      // There isn't a way for ByteSink to report errors.
+  while (true) {
+    if (len <= buffer_size_) {
+      memcpy(buffer_, bytes, len);
+      buffer_ = static_cast<char*>(buffer_) + len;
+      buffer_size_ -= len;
       return;
     }
-    if (len < length) {
-      memcpy(buffer, bytes, len);
-      stream_->BackUp(length - len);
-      break;
-    } else {
-      memcpy(buffer, bytes, length);
-      bytes += length;
-      len -= length;
+    memcpy(buffer_, bytes, buffer_size_);
+    bytes += buffer_size_;
+    len -= buffer_size_;
+    if (!stream_->Next(&buffer_, &buffer_size_)) {
+      // There isn't a way for ByteSink to report errors.
+      buffer_size_ = 0;
+      return;
     }
   }
 }
diff --git a/src/google/protobuf/util/json_util.h b/src/google/protobuf/util/json_util.h
index 6d3cee52..0e45ef81 100644
--- a/src/google/protobuf/util/json_util.h
+++ b/src/google/protobuf/util/json_util.h
@@ -172,12 +172,15 @@ namespace internal {
 class LIBPROTOBUF_EXPORT ZeroCopyStreamByteSink : public strings::ByteSink {
  public:
   explicit ZeroCopyStreamByteSink(io::ZeroCopyOutputStream* stream)
-      : stream_(stream) {}
+      : stream_(stream), buffer_size_(0) {}
+  ~ZeroCopyStreamByteSink();
 
   virtual void Append(const char* bytes, size_t len);
 
  private:
   io::ZeroCopyOutputStream* stream_;
+  void* buffer_;
+  int buffer_size_;
 
   GOOGLE_DISALLOW_EVIL_CONSTRUCTORS(ZeroCopyStreamByteSink);
 };