Use simdjson for json_parse (#7658)

Summary:
Pull Request resolved: #7658

In `json_parse` when the input is invalid, we throw exception and it's slow (both the creation and throwing).  To avoid creating or throwing the exception, we switch the implementation to `simdjson` and set a pre-canned exception when the input is invalid.  This reduces the CPU time in some queries (with JSON validity check in filter) by more than 20 times (from 2.34 days to 2.68 hours).

Reviewed By: mbasmanova

Differential Revision: D51469435

fbshipit-source-id: de126a2b35a51c210a6d2ddadc31f6fb5df22d75

velox/functions/prestosql/JsonFunctions.cpp

@@ -14,6 +14,7 @@
  * limitations under the License.
  */
 #include "velox/expression/VectorFunction.h"
+#include "velox/functions/prestosql/json/SIMDJsonWrapper.h"
 #include "velox/functions/prestosql/types/JsonType.h"
 
 namespace facebook::velox::functions {
@@ -70,19 +71,27 @@ class JsonParseFunction : public exec::VectorFunction {
       const TypePtr& /* outputType */,
       exec::EvalCtx& context,
       VectorPtr& result) const override {
+    folly::call_once(initializeErrors_, [this] {
+      // Initilize errors here so that we get the proper exception context.
+      for (int i = 1; i < simdjson::NUM_ERROR_CODES; ++i) {
+        simdjson::simdjson_error e(static_cast<simdjson::error_code>(i));
+        errors_[i] = toVeloxException(std::make_exception_ptr(e));
+      }
+    });
+
     VectorPtr localResult;
 
     // Input can be constant or flat.
     // TODO(arpitporwal2293) Replace folly::parseJson with a lightweight
     // validation of JSON syntax that doesn't allocate memory or copy data.
     assert(args.size() > 0);
     const auto& arg = args[0];
+    static_assert(simdjson::SIMDJSON_PADDING <= AlignedBuffer::kPaddedSize);
     if (arg->isConstantEncoding()) {
       auto value = arg->as<ConstantVector<StringView>>()->valueAt(0);
-      try {
-        folly::parseJson(value);
-      } catch (const std::exception& e) {
-        context.setErrors(rows, std::current_exception());
+      auto parsed = parser_.parse(value.data(), value.size(), false);
+      if (parsed.error() != simdjson::SUCCESS) {
+        context.setErrors(rows, errors_[parsed.error()]);
         return;
       }
       localResult = std::make_shared<ConstantVector<StringView>>(
@@ -93,8 +102,13 @@ class JsonParseFunction : public exec::VectorFunction {
       auto stringBuffers = flatInput->stringBuffers();
       VELOX_CHECK_LE(rows.end(), flatInput->size());
 
-      context.applyToSelectedNoThrow(
-          rows, [&](auto row) { folly::parseJson(flatInput->valueAt(row)); });
+      rows.applyToSelected([&](auto row) {
+        auto value = flatInput->valueAt(row);
+        auto parsed = parser_.parse(value.data(), value.size(), false);
+        if (parsed.error() != simdjson::SUCCESS) {
+          context.setVeloxExceptionError(row, errors_[parsed.error()]);
+        }
+      });
       localResult = std::make_shared<FlatVector<StringView>>(
           context.pool(),
           JSON(),
@@ -114,6 +128,11 @@ class JsonParseFunction : public exec::VectorFunction {
                 .argumentType("varchar")
                 .build()};
   }
+
+ private:
+  mutable folly::once_flag initializeErrors_;
+  mutable std::exception_ptr errors_[simdjson::NUM_ERROR_CODES];
+  mutable simdjson::dom::parser parser_;
 };
 
 } // namespace
@@ -123,8 +142,13 @@ VELOX_DECLARE_VECTOR_FUNCTION(
     JsonFormatFunction::signatures(),
     std::make_unique<JsonFormatFunction>());
 
-VELOX_DECLARE_VECTOR_FUNCTION(
+VELOX_DECLARE_STATEFUL_VECTOR_FUNCTION(
     udf_json_parse,
     JsonParseFunction::signatures(),
-    std::make_unique<JsonParseFunction>());
+    [](const std::string& /*name*/,
+       const std::vector<exec::VectorFunctionArg>&,
+       const velox::core::QueryConfig&) {
+      return std::make_shared<JsonParseFunction>();
+    });
+
 } // namespace facebook::velox::functions

velox/functions/prestosql/tests/JsonFunctionsTest.cpp

@@ -197,10 +197,11 @@ TEST_F(JsonFunctionsTest, jsonParse) {
   EXPECT_EQ(jsonParse(R"({"k1":"v1"})"), R"({"k1":"v1"})");
   EXPECT_EQ(jsonParse(R"(["k1", "v1"])"), R"(["k1", "v1"])");
 
-  VELOX_ASSERT_THROW(jsonParse(R"({"k1":})"), "expected json value");
   VELOX_ASSERT_THROW(
-      jsonParse(R"({:"k1"})"), "json parse error on line 0 near `:\"k1\"}");
-  VELOX_ASSERT_THROW(jsonParse(R"(not_json)"), "expected json value");
+      jsonParse(R"({"k1":})"), "The JSON document has an improper structure");
+  VELOX_ASSERT_THROW(
+      jsonParse(R"({:"k1"})"), "The JSON document has an improper structure");
+  VELOX_ASSERT_THROW(jsonParse(R"(not_json)"), "Problem while parsing an atom");
 
   EXPECT_EQ(jsonParseWithTry(R"(not_json)"), std::nullopt);
   EXPECT_EQ(jsonParseWithTry(R"({"k1":})"), std::nullopt);
@@ -223,7 +224,7 @@ TEST_F(JsonFunctionsTest, jsonParse) {
 
   VELOX_ASSERT_THROW(
       evaluate("json_parse(c0)", data),
-      "json parse error on line 0 near `:': parsing didn't consume all input");
+      "missing or superfluous commas, braces, missing keys, etc.");
 
   data = makeRowVector({makeFlatVector<StringView>(
       {R"("This is a long sentence")", R"("This is some other sentence")"})});
@@ -251,6 +252,13 @@ TEST_F(JsonFunctionsTest, jsonParse) {
       {R"("This is a long sentence")", R"("This is some other sentence")"},
       JSON());
   velox::test::assertEqualVectors(expected, result);
+
+  try {
+    jsonParse(R"({"k1":})");
+    FAIL() << "Error expected";
+  } catch (const VeloxUserError& e) {
+    ASSERT_EQ(e.context(), "json_parse(c0)");
+  }
 }
 
 TEST_F(JsonFunctionsTest, isJsonScalarSignatures) {