(arrow) branch main updated: GH-50654: [C++] Support simdjson without exceptions (#50672)

[kou]

This is an automated email from the ASF dual-hosted git repository.

kou pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow.git


The following commit(s) were added to refs/heads/main by this push:
     new bb0cf77107 GH-50654: [C++] Support simdjson without exceptions (#50672)
bb0cf77107 is described below

commit bb0cf7710738a0e09e149c2f7e27e4f509b42a61
Author: Aaditya Srinivasan <156181482+rerank...@users.noreply.github.com>
AuthorDate: Thu Jul 30 07:59:35 2026 +0530

    GH-50654: [C++] Support simdjson without exceptions (#50672)
    
    ### Rationale for this change
    
    This change enables building Arrow with `SIMDJSON_EXCEPTIONS=0` by updating 
the JSON object parser to use simdjson's non-throwing API.
    
    ### What changes are included?
    
    - Enable `SIMDJSON_EXCEPTIONS=0` for the bundled simdjson dependency.
    - Replace uses of `simdjson_result::value()` in `ObjectParser` with the 
non-throwing `get()` API.
    - Preserve existing error handling by returning Arrow `Status` values on 
simdjson errors.
    
    ### Are these changes tested?
    
    Yes.
    
    Fixes: #50654
    * GitHub Issue: #50654
    
    Authored-by: Aaditya Srinivasan <aadityasr...@gmail.com>
    Signed-off-by: Sutou Kouhei <k...@clear-code.com>
---
 cpp/cmake_modules/ThirdpartyToolchain.cmake |  2 ++
 cpp/src/arrow/json/object_parser.cc         | 38 ++++++++++++++++++++---------
 2 files changed, 29 insertions(+), 11 deletions(-)

diff --git a/cpp/cmake_modules/ThirdpartyToolchain.cmake 
b/cpp/cmake_modules/ThirdpartyToolchain.cmake
index 9fc1e3d0aa..c4548e788f 100644
--- a/cpp/cmake_modules/ThirdpartyToolchain.cmake
+++ b/cpp/cmake_modules/ThirdpartyToolchain.cmake
@@ -2823,6 +2823,8 @@ function(build_simdjson)
 
   fetchcontent_makeavailable(simdjson)
 
+  target_compile_definitions(simdjson PUBLIC SIMDJSON_EXCEPTIONS=0)
+
   # The macOS 11.3 SDK has incomplete C++20 concepts support, which prevents
   # simdjson headers from compiling. Disable simdjson concepts for this SDK.
   if(CMAKE_OSX_SYSROOT AND CMAKE_OSX_SYSROOT MATCHES "MacOSX11\\.3\\.sdk$")
diff --git a/cpp/src/arrow/json/object_parser.cc 
b/cpp/src/arrow/json/object_parser.cc
index c730988c1e..4aad2aca67 100644
--- a/cpp/src/arrow/json/object_parser.cc
+++ b/cpp/src/arrow/json/object_parser.cc
@@ -30,7 +30,9 @@ class ObjectParser::Impl {
     padded_json_ = simdjson::padded_string(json);
 
     // Store parsed document
-    document_ = parser_.iterate(padded_json_);
+    if (auto error = parser_.iterate(padded_json_).get(document_)) {
+      return Status::Invalid("JSON parse error: ", 
simdjson::error_message(error));
+    }
 
     // Validate root is an object
     auto object = document_.get_object();
@@ -69,7 +71,12 @@ class ObjectParser::Impl {
                              "': ", 
simdjson::error_message(str_result.error()));
     }
 
-    return std::string(str_result.value());
+    std::string_view str;
+    if (auto error = std::move(str_result).get(str)) {
+      return Status::Invalid("Error getting string for key '", key,
+                             "': ", simdjson::error_message(error));
+    }
+    return std::string(str);
   }
 
   Result<std::unordered_map<std::string, std::string>> GetStringMap() {
@@ -80,13 +87,10 @@ class ObjectParser::Impl {
     auto object = document_.get_object();
 
     for (auto field : object) {
-      auto key_result = field.unescaped_key();
-
-      auto key = key_result.value();
-
-      if (key_result.error()) {
-        return Status::Invalid("Error getting value for key '", 
std::string(key),
-                               "': ", 
simdjson::error_message(key_result.error()));
+      std::string_view key;
+      if (auto error = field.unescaped_key().get(key)) {
+        return Status::Invalid("Error getting object key: ",
+                               simdjson::error_message(error));
       }
 
       auto value = field.value();
@@ -102,7 +106,13 @@ class ObjectParser::Impl {
                                "': (code=", 
static_cast<int>(str_result.error()), ")");
       }
 
-      map.emplace(std::string(key), std::string(str_result.value()));
+      std::string_view str;
+      if (auto error = std::move(str_result).get(str)) {
+        return Status::Invalid("Error getting value for key '", 
std::string(key),
+                               "': ", simdjson::error_message(error));
+      }
+
+      map.emplace(std::string(key), std::string(str));
     }
 
     return map;
@@ -132,7 +142,13 @@ class ObjectParser::Impl {
                              "': ", 
simdjson::error_message(bool_result.error()));
     }
 
-    return bool_result.value();
+    bool value;
+    if (auto error = std::move(bool_result).get(value)) {
+      return Status::Invalid("Error getting bool for key '", key,
+                             "': ", simdjson::error_message(error));
+    }
+
+    return value;
   }
 
  private: