Add RapidJSON and nlohmann_json SAX to find_tweet benchmark (#1598)

* Add rapidjson_sax.h .

* Add nlohmann_json_sax.h . Fix typos distinct_user_id/nlohmann_json_sax.h, find_tweet/rapidjson.h and find_tweet/rapidjson_sax.h .

* Add extra check for id key when looking for find_id.

benchmark/bench_ondemand.cpp

@@ -64,7 +64,9 @@ SIMDJSON_POP_DISABLE_WARNINGS
 #include "find_tweet/yyjson.h"
 #include "find_tweet/sajson.h"
 #include "find_tweet/rapidjson.h"
+#include "find_tweet/rapidjson_sax.h"
 #include "find_tweet/nlohmann_json.h"
+#include "find_tweet/nlohmann_json_sax.h"
 
 #include "top_tweet/simdjson_dom.h"
 #include "top_tweet/simdjson_ondemand.h"

benchmark/distinct_user_id/nlohmann_json_sax.h

@@ -3,7 +3,7 @@
 #ifdef SIMDJSON_COMPETITION_NLOHMANN_JSON
 
 #include "distinct_user_id.h"
-#include <string.h>
+
 namespace distinct_user_id {
 
 using json = nlohmann::json;

benchmark/find_tweet/nlohmann_json_sax.h

@@ -0,0 +1,68 @@
+#pragma once
+
+#ifdef SIMDJSON_COMPETITION_NLOHMANN_JSON
+
+#include "find_tweet.h"
+
+namespace find_tweet {
+
+using json = nlohmann::json;
+
+struct nlohmann_json_sax {
+    using StringType=std::string;
+
+    struct Handler : json::json_sax_t
+    {
+        bool text_key = false;
+        bool id_key = false;
+        bool found_id = false;
+        uint64_t find_id;
+        std::string &result;
+
+        Handler(std::string &r,uint64_t id): result(r), find_id(id) { }
+
+        // We assume id is found before text
+        bool key(string_t& val) override {
+            if (found_id) { // If have found id, find text key
+                if (val.compare("text") == 0) { text_key = true; }
+            }
+            else if (val.compare("id") == 0) { id_key = true; } // Otherwise, find id key
+            return true;
+        }
+        bool number_unsigned(number_unsigned_t val) override {
+            if (id_key && (val == find_id)) {  // If id key, check if id value matches find_id
+                found_id = true;
+            }
+            return true;
+        }
+        bool string(string_t& val) override {
+            if (text_key) {
+                result = val;
+                return false;   // End parsing
+            }
+            return true;
+        }
+        // Irrelevant events
+        bool null() override { return true; }
+        bool boolean(bool val) override { return true; }
+        bool number_float(number_float_t val, const string_t& s) override { return true; }
+        bool number_integer(number_integer_t val) override { return true; }
+        bool start_object(std::size_t elements) override { return true; }
+        bool end_object() override { return true; }
+        bool start_array(std::size_t elements) override { return true; }
+        bool end_array() override { return true; }
+        bool binary(json::binary_t& val) override { return true; }
+        bool parse_error(std::size_t position, const std::string& last_token, const json::exception& ex) override { return false; }
+    }; // Handler
+
+    bool run(simdjson::padded_string &json, uint64_t find_id, std::string &result) {
+        Handler handler(result,find_id);
+        json::sax_parse(json.data(), &handler);
+
+        return true;
+    }
+}; // nlohmann_json_sax
+BENCHMARK_TEMPLATE(find_tweet, nlohmann_json_sax)->UseManualTime();
+} // namespace find_tweet
+
+#endif // SIMDJSON_COMPETITION_NLOHMANN_JSON

benchmark/find_tweet/rapidjson.h

@@ -47,6 +47,6 @@ struct rapidjson_insitu : rapidjson_base {
 };
 BENCHMARK_TEMPLATE(find_tweet, rapidjson_insitu)->UseManualTime();
 
-} // namespace partial_tweets
+} // namespace find_tweet
 
 #endif // SIMDJSON_COMPETITION_RAPIDJSON

benchmark/find_tweet/rapidjson_sax.h

@@ -0,0 +1,70 @@
+#pragma once
+
+#ifdef SIMDJSON_COMPETITION_RAPIDJSON
+
+#include "find_tweet.h"
+#include <string.h>
+
+namespace find_tweet {
+
+using namespace rapidjson;
+
+struct rapidjson_sax {
+using StringType=std::string_view;
+
+struct Handler {
+        bool text_key = false;
+        bool id_key = false;
+        bool found_id = false;
+        uint64_t find_id;
+        std::string_view &result;
+
+        Handler(std::string_view &r,uint64_t id): result(r), find_id(id) { }
+
+        // We assume id is found before text
+        bool Key(const char* key, SizeType length, bool copy) {
+            if (found_id) { // If have found id, find text key
+                if ((length == 4) && (memcmp(key,"text",4) == 0)) { text_key = true; }
+            }
+            else if ((length == 2) && (memcmp(key,"id",2) == 0)) { id_key = true; } // Otherwise, find id key
+            return true;
+        }
+        bool Uint64(uint64_t i) {
+            if (id_key && (i == find_id)) {  // If id key, check if id value matches find_id
+                found_id = true;
+            }
+            return true;
+        }
+        bool String(const char* str, SizeType length, bool copy) {
+            if (text_key) {
+                result = {str,length};
+                return false;   // End parsing
+            }
+            return true;
+        }
+        // Irrelevant events
+        bool Null() { return true; }
+        bool Bool(bool b) { return true; }
+        bool Double(double d) { return true; }
+        bool Int(int i) { return true; }
+        bool Int64(int64_t i) { return true; }
+        bool Uint(unsigned i) { return true; }
+        bool RawNumber(const char* str, SizeType length, bool copy) { return true; }
+        bool StartObject() { return true; }
+        bool EndObject(SizeType memberCount) { return true; }
+        bool StartArray() { return true; }
+        bool EndArray(SizeType elementCount) { return true; }
+    }; // handler
+
+  bool run(simdjson::padded_string &json, uint64_t find_id, std::string_view &result) {
+        Reader reader;
+        Handler handler(result,find_id);
+        InsituStringStream ss(json.data());
+        reader.Parse<kParseInsituFlag | kParseValidateEncodingFlag | kParseFullPrecisionFlag>(ss,handler);
+        return true;
+    }
+}; // rapidjson_sax
+BENCHMARK_TEMPLATE(find_tweet, rapidjson_sax)->UseManualTime();
+} // namespace find_tweet
+
+#endif // SIMDJSON_COMPETITION_RAPIDJSON