Merge branch 'main' of https://github.com/R-JunmingChen/arrow into AR…

…ROW-36831

.github/dependabot.yml

@@ -23,6 +23,12 @@ updates:
       interval: "weekly"
     commit-message:
       prefix: "MINOR: [CI] "
+  - package-ecosystem: "npm"
+    directory: "/js/"
+    schedule:
+      interval: "monthly"
+    commit-message:
+      prefix: "MINOR: [JS] "
   - package-ecosystem: "nuget"
     directory: "/csharp/"
     schedule:

.github/workflows/cpp.yml

@@ -164,8 +164,7 @@ jobs:
       ARROW_HDFS: ON
       ARROW_HOME: /usr/local
       ARROW_JEMALLOC: ON
-      # TODO(kszucs): link error in the tests
-      ARROW_ORC: OFF
+      ARROW_ORC: ON
       ARROW_PARQUET: ON
       ARROW_S3: ON
       ARROW_WITH_BROTLI: ON
@@ -273,7 +272,6 @@ jobs:
       CMAKE_INSTALL_LIBDIR: bin
       CMAKE_INSTALL_PREFIX: /usr
       CMAKE_UNITY_BUILD: ON
-      NPROC: 3
     steps:
       - name: Disable Crash Dialogs
         run: |

.github/workflows/docs.yml

@@ -30,17 +30,20 @@ env:
 jobs:
 
   complete:
-    name: AMD64 Ubuntu 20.04 Complete Documentation
+    name: AMD64 Ubuntu 22.04 Complete Documentation
     runs-on: ubuntu-latest
     if: ${{ !contains(github.event.pull_request.title, 'WIP') }}
     timeout-minutes: 150
     env:
-      UBUNTU: "20.04"
+      UBUNTU: "22.04"
     steps:
       - name: Checkout Arrow
         uses: actions/checkout@v3
         with:
           fetch-depth: 0
+      - name: Free up disk space
+        run: |
+          ci/scripts/util_free_space.sh
       - name: Cache Docker Volumes
         uses: actions/cache@v3
         with:

.github/workflows/java.yml

@@ -55,7 +55,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        jdk: [8, 11, 17, 18]
+        jdk: [8, 11, 17, 20]
         include:
         - jdk: 8
           title: AMD64 Debian 9 Java JDK 8 Maven 3.5.4
@@ -66,13 +66,13 @@ jobs:
           maven: 3.6.2
           image: debian-java
         - jdk: 17
-          title: AMD64 Oracle Linux Server 8.5 Java JDK 17 Maven 3.8.5
-          maven: 3.8.5
-          image: oracle-java
-        - jdk: 18
-          title: AMD64 Oracle Linux Server 8.6 Java JDK 18 Maven 3.8.5
-          maven: 3.8.5
-          image: oracle-java
+          title: AMD64 Ubuntu 22.04 Java JDK 17 Maven 3.9.3
+          maven: 3.9.3
+          image: eclipse-java
+        - jdk: 20
+          title: AMD64 Ubuntu 22.04 Java JDK 20 Maven 3.9.3
+          maven: 3.9.3
+          image: eclipse-java
     env:
       JDK: ${{ matrix.jdk }}
       MAVEN: ${{ matrix.maven }}

c_glib/arrow-flight-glib/client.cpp

@@ -166,12 +166,17 @@ gaflight_call_options_foreach_header(GAFlightCallOptions *options,
 }
 
 
-typedef struct GAFlightClientOptionsPrivate_ {
+struct GAFlightClientOptionsPrivate {
   arrow::flight::FlightClientOptions options;
-} GAFlightClientOptionsPrivate;
+};
 
 enum {
-  PROP_DISABLE_SERVER_VERIFICATION = 1,
+  PROP_TLS_ROOT_CERTIFICATES = 1,
+  PROP_OVERRIDE_HOST_NAME,
+  PROP_CERTIFICATE_CHAIN,
+  PROP_PRIVATE_KEY,
+  PROP_WRITE_SIZE_LIMIT_BYTES,
+  PROP_DISABLE_SERVER_VERIFICATION,
 };
 
 G_DEFINE_TYPE_WITH_PRIVATE(GAFlightClientOptions,
@@ -202,6 +207,21 @@ gaflight_client_options_set_property(GObject *object,
   auto priv = GAFLIGHT_CLIENT_OPTIONS_GET_PRIVATE(object);
 
   switch (prop_id) {
+  case PROP_TLS_ROOT_CERTIFICATES:
+    priv->options.tls_root_certs = g_value_get_string(value);
+    break;
+  case PROP_OVERRIDE_HOST_NAME:
+    priv->options.override_hostname = g_value_get_string(value);
+    break;
+  case PROP_CERTIFICATE_CHAIN:
+    priv->options.cert_chain = g_value_get_string(value);
+    break;
+  case PROP_PRIVATE_KEY:
+    priv->options.private_key = g_value_get_string(value);
+    break;
+  case PROP_WRITE_SIZE_LIMIT_BYTES:
+    priv->options.write_size_limit_bytes = g_value_get_int64(value);
+    break;
   case PROP_DISABLE_SERVER_VERIFICATION:
     priv->options.disable_server_verification = g_value_get_boolean(value);
     break;
@@ -220,6 +240,21 @@ gaflight_client_options_get_property(GObject *object,
   auto priv = GAFLIGHT_CLIENT_OPTIONS_GET_PRIVATE(object);
 
   switch (prop_id) {
+  case PROP_TLS_ROOT_CERTIFICATES:
+    g_value_set_string(value, priv->options.tls_root_certs.c_str());
+    break;
+  case PROP_OVERRIDE_HOST_NAME:
+    g_value_set_string(value, priv->options.override_hostname.c_str());
+    break;
+  case PROP_CERTIFICATE_CHAIN:
+    g_value_set_string(value, priv->options.cert_chain.c_str());
+    break;
+  case PROP_PRIVATE_KEY:
+    g_value_set_string(value, priv->options.private_key.c_str());
+    break;
+  case PROP_WRITE_SIZE_LIMIT_BYTES:
+    g_value_set_int64(value, priv->options.write_size_limit_bytes);
+    break;
   case PROP_DISABLE_SERVER_VERIFICATION:
     g_value_set_boolean(value, priv->options.disable_server_verification);
     break;
@@ -248,6 +283,93 @@ gaflight_client_options_class_init(GAFlightClientOptionsClass *klass)
 
   auto options = arrow::flight::FlightClientOptions::Defaults();
   GParamSpec *spec;
+  /**
+   * GAFlightClientOptions:tls-root-certificates:
+   *
+   * Root certificates to use for validating server certificates.
+   *
+   * Since: 14.0.0
+   */
+  spec = g_param_spec_string("tls-root-certificates",
+                             nullptr,
+                             nullptr,
+                             options.tls_root_certs.c_str(),
+                             static_cast<GParamFlags>(G_PARAM_READWRITE));
+  g_object_class_install_property(gobject_class,
+                                  PROP_TLS_ROOT_CERTIFICATES,
+                                  spec);
+
+  /**
+   * GAFlightClientOptions:override-host-name:
+   *
+   * Override the host name checked by TLS. Use with caution.
+   *
+   * Since: 14.0.0
+   */
+  spec = g_param_spec_string("override-host-name",
+                             nullptr,
+                             nullptr,
+                             options.override_hostname.c_str(),
+                             static_cast<GParamFlags>(G_PARAM_READWRITE));
+  g_object_class_install_property(gobject_class,
+                                  PROP_OVERRIDE_HOST_NAME,
+                                  spec);
+
+  /**
+   * GAFlightClientOptions:certificate-chain:
+   *
+   * The client certificate to use if using Mutual TLS.
+   *
+   * Since: 14.0.0
+   */
+  spec = g_param_spec_string("certificate-chain",
+                             nullptr,
+                             nullptr,
+                             options.cert_chain.c_str(),
+                             static_cast<GParamFlags>(G_PARAM_READWRITE));
+  g_object_class_install_property(gobject_class,
+                                  PROP_CERTIFICATE_CHAIN,
+                                  spec);
+
+  /**
+   * GAFlightClientOptions:private-key:
+   *
+   * The private key associated with the client certificate for Mutual
+   * TLS.
+   *
+   * Since: 14.0.0
+   */
+  spec = g_param_spec_string("private-key",
+                             nullptr,
+                             nullptr,
+                             options.private_key.c_str(),
+                             static_cast<GParamFlags>(G_PARAM_READWRITE));
+  g_object_class_install_property(gobject_class,
+                                  PROP_PRIVATE_KEY,
+                                  spec);
+
+  /**
+   * GAFlightClientOptions:write-size-limit-bytes:
+   *
+   * A soft limit on the number of bytes to write in a single batch
+   * when sending Arrow data to a server.
+   *
+   * Used to help limit server memory consumption. Only enabled if
+   * positive. When enabled, @GARROW_ERROR_IO may be yielded.
+   *
+   * Since: 14.0.0
+   */
+  spec = g_param_spec_int64("write-size-limit-bytes",
+                            nullptr,
+                            nullptr,
+                            INT64_MIN,
+                            INT64_MAX,
+                            options.write_size_limit_bytes,
+                            static_cast<GParamFlags>(G_PARAM_READWRITE));
+  g_object_class_install_property(gobject_class,
+                                  PROP_WRITE_SIZE_LIMIT_BYTES,
+                                  spec);
+
   /**
    * GAFlightClientOptions:disable-server-verification:
    *
@@ -567,16 +689,14 @@ gaflight_client_do_get(GAFlightClient *client,
   if (options) {
     flight_options = gaflight_call_options_get_raw(options);
   }
-  std::unique_ptr<arrow::flight::FlightStreamReader> flight_reader;
   auto result = flight_client->DoGet(*flight_options, *flight_ticket);
-  auto status = std::move(result).Value(&flight_reader);
-  if (garrow::check(error,
-                    status,
-                    "[flight-client][do-get]")) {
-    return gaflight_stream_reader_new_raw(flight_reader.release());
-  } else {
-    return NULL;
+  if (!garrow::check(error,
+                     result,
+                     "[flight-client][do-get]")) {
+    return nullptr;
   }
+  auto flight_reader = std::move(*result);
+  return gaflight_stream_reader_new_raw(flight_reader.release(), TRUE);
 }
 
 
@@ -585,11 +705,13 @@ G_END_DECLS
 
 GAFlightStreamReader *
 gaflight_stream_reader_new_raw(
-  arrow::flight::FlightStreamReader *flight_reader)
+  arrow::flight::FlightStreamReader *flight_reader,
+  gboolean is_owner)
 {
   return GAFLIGHT_STREAM_READER(
     g_object_new(GAFLIGHT_TYPE_STREAM_READER,
                  "reader", flight_reader,
+                 "is-owner", is_owner,
                  NULL));
 }
 

c_glib/arrow-flight-glib/client.hpp

@@ -25,7 +25,9 @@
 
 
 GAFlightStreamReader *
-gaflight_stream_reader_new_raw(arrow::flight::FlightStreamReader *flight_reader);
+gaflight_stream_reader_new_raw(
+  arrow::flight::FlightStreamReader *flight_reader,
+  gboolean is_owner);
 
 arrow::flight::FlightCallOptions *
 gaflight_call_options_get_raw(GAFlightCallOptions *options);

c_glib/arrow-flight-glib/common.cpp

@@ -1223,10 +1223,12 @@ gaflight_stream_chunk_get_metadata(GAFlightStreamChunk *chunk)
 
 typedef struct GAFlightRecordBatchReaderPrivate_ {
   arrow::flight::MetadataRecordBatchReader *reader;
+  bool is_owner;
 } GAFlightRecordBatchReaderPrivate;
 
 enum {
   PROP_READER = 1,
+  PROP_IS_OWNER,
 };
 
 G_DEFINE_TYPE_WITH_PRIVATE(GAFlightRecordBatchReader,
@@ -1242,9 +1244,9 @@ static void
 gaflight_record_batch_reader_finalize(GObject *object)
 {
   auto priv = GAFLIGHT_RECORD_BATCH_READER_GET_PRIVATE(object);
-
-  delete priv->reader;
-
+  if (priv->is_owner) {
+    delete priv->reader;
+  }
   G_OBJECT_CLASS(gaflight_info_parent_class)->finalize(object);
 }
 
@@ -1262,6 +1264,9 @@ gaflight_record_batch_reader_set_property(GObject *object,
       static_cast<arrow::flight::MetadataRecordBatchReader *>(
         g_value_get_pointer(value));
     break;
+  case PROP_IS_OWNER:
+    priv->is_owner = g_value_get_boolean(value);
+    break;
   default:
     G_OBJECT_WARN_INVALID_PROPERTY_ID(object, prop_id, pspec);
     break;
@@ -1283,11 +1288,19 @@ gaflight_record_batch_reader_class_init(GAFlightRecordBatchReaderClass *klass)
 
   GParamSpec *spec;
   spec = g_param_spec_pointer("reader",
-                              "Reader",
-                              "The raw arrow::flight::MetadataRecordBatchReader *",
+                              nullptr,
+                              nullptr,
                               static_cast<GParamFlags>(G_PARAM_WRITABLE |
                                                        G_PARAM_CONSTRUCT_ONLY));
   g_object_class_install_property(gobject_class, PROP_READER, spec);
+
+  spec = g_param_spec_boolean("is-owner",
+                              nullptr,
+                              nullptr,
+                              TRUE,
+                              static_cast<GParamFlags>(G_PARAM_WRITABLE |
+                                                       G_PARAM_CONSTRUCT_ONLY));
+  g_object_class_install_property(gobject_class, PROP_IS_OWNER, spec);
 }
 
 /**