ARROW-17693: [C++] Remove string_view backport (#14177)

Authored-by: Antoine Pitrou <antoine@python.org>
Signed-off-by: Antoine Pitrou <antoine@python.org>

.github/workflows/cpp.yml

@@ -146,8 +146,7 @@ jobs:
       ARROW_WITH_SNAPPY: ON
       ARROW_WITH_ZLIB: ON
       ARROW_WITH_ZSTD: ON
-      # System Abseil installed by Homebrew uses C++ 17
-      CMAKE_CXX_STANDARD: 17
+      GTest_SOURCE: BUNDLED
     steps:
       - name: Checkout Arrow
         uses: actions/checkout@v3

LICENSE.txt

@@ -653,34 +653,6 @@ SOFTWARE.
 
 --------------------------------------------------------------------------------
 
-The file cpp/src/arrow/vendored/string_view.hpp has the following license
-
-Boost Software License - Version 1.0 - August 17th, 2003
-
-Permission is hereby granted, free of charge, to any person or organization
-obtaining a copy of the software and accompanying documentation covered by
-this license (the "Software") to use, reproduce, display, distribute,
-execute, and transmit the Software, and to prepare derivative works of the
-Software, and to permit third-parties to whom the Software is furnished to
-do so, all subject to the following:
-
-The copyright notices in the Software and this entire statement, including
-the above license grant, this restriction and the following disclaimer,
-must be included in all copies of the Software, in whole or in part, and
-all derivative works of the Software, unless such copies or derivative
-works are solely in the form of machine-executable object code generated by
-a source language processor.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
-SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
-FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
-ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
-DEALINGS IN THE SOFTWARE.
-
---------------------------------------------------------------------------------
-
 The files in cpp/src/arrow/vendored/xxhash/ have the following license
 (BSD 2-Clause License)
 

c_glib/arrow-glib/compute.cpp

@@ -5121,7 +5121,7 @@ GArrowFunctionOptions *
 garrow_function_options_new_raw(
   const arrow::compute::FunctionOptions *arrow_options)
 {
-  arrow::util::string_view arrow_type_name(arrow_options->type_name());
+  std::string_view arrow_type_name(arrow_options->type_name());
   if (arrow_type_name == "CastOptions") {
     auto arrow_cast_options =
       static_cast<const arrow::compute::CastOptions *>(arrow_options);

c_glib/arrow-glib/input-stream.cpp

@@ -20,7 +20,6 @@
 #include <arrow/io/interfaces.h>
 #include <arrow/io/memory.h>
 #include <arrow/ipc/reader.h>
-#include <arrow/util/string_view.h>
 
 #include <arrow-glib/buffer.hpp>
 #include <arrow-glib/codec.hpp>
@@ -34,6 +33,7 @@
 #include <arrow-glib/tensor.hpp>
 
 #include <mutex>
+#include <string_view>
 
 G_BEGIN_DECLS
 
@@ -855,7 +855,7 @@ namespace garrow {
       }
     }
 
-    arrow::Result<arrow::util::string_view> Peek(int64_t nbytes) override {
+    arrow::Result<std::string_view> Peek(int64_t nbytes) override {
       if (!G_IS_BUFFERED_INPUT_STREAM(input_stream_)) {
         std::string message("[gio-input-stream][peek] "
                             "not peekable input stream: <");
@@ -882,8 +882,7 @@ namespace garrow {
       if (data_size > static_cast<gsize>(nbytes)) {
         data_size = nbytes;
       }
-      return arrow::util::string_view(static_cast<const char *>(data),
-                                      data_size);
+      return std::string_view(static_cast<const char *>(data), data_size);
     }
 
     arrow::Status Seek(int64_t position) override {

c_glib/arrow-glib/scalar.cpp

@@ -250,9 +250,8 @@ garrow_scalar_parse(GArrowDataType *data_type,
                     GError **error)
 {
   const auto arrow_data_type = garrow_data_type_get_raw(data_type);
-  auto arrow_data =
-    arrow::util::string_view(reinterpret_cast<const char *>(data),
-                             size);
+  auto arrow_data = std::string_view(reinterpret_cast<const char *>(data),
+                                     size);
   auto arrow_scalar_result = arrow::Scalar::Parse(arrow_data_type, arrow_data);
   if (garrow::check(error, arrow_scalar_result, "[scalar][parse]")) {
     auto arrow_scalar = *arrow_scalar_result;

ci/docker/debian-11-cpp.dockerfile

@@ -106,6 +106,7 @@ ENV absl_SOURCE=BUNDLED \
     CC=gcc \
     CXX=g++ \
     google_cloud_cpp_storage_SOURCE=BUNDLED \
+    GTest_SOURCE=BUNDLED \
     ORC_SOURCE=BUNDLED \
     PATH=/usr/lib/ccache/:$PATH \
     Protobuf_SOURCE=BUNDLED \

ci/scripts/java_jni_macos_build.sh

@@ -74,6 +74,7 @@ cmake \
   -DCMAKE_INSTALL_LIBDIR=lib \
   -DCMAKE_INSTALL_PREFIX=${install_dir} \
   -DCMAKE_UNITY_BUILD=${CMAKE_UNITY_BUILD} \
+  -DGTest_SOURCE=BUNDLED \
   -DPARQUET_BUILD_EXAMPLES=OFF \
   -DPARQUET_BUILD_EXECUTABLES=OFF \
   -DPARQUET_REQUIRE_ENCRYPTION=OFF \

ci/scripts/java_jni_manylinux_build.sh

@@ -84,6 +84,7 @@ cmake \
   -DCMAKE_INSTALL_LIBDIR=lib \
   -DCMAKE_INSTALL_PREFIX=${ARROW_HOME} \
   -DCMAKE_UNITY_BUILD=${CMAKE_UNITY_BUILD} \
+  -DGTest_SOURCE=BUNDLED \
   -DORC_SOURCE=BUNDLED \
   -DORC_PROTOBUF_EXECUTABLE=${VCPKG_ROOT}/installed/${VCPKG_TARGET_TRIPLET}/tools/protobuf/protoc \
   -DPARQUET_BUILD_EXAMPLES=OFF \

cpp/cmake_modules/ThirdpartyToolchain.cmake

@@ -1974,9 +1974,8 @@ macro(build_gtest)
   set(dummy ">")
 
   set(GTEST_CMAKE_ARGS
-      ${EP_COMMON_TOOLCHAIN}
+      ${EP_COMMON_CMAKE_ARGS}
       -DBUILD_SHARED_LIBS=ON
-      -DCMAKE_BUILD_TYPE=${CMAKE_BUILD_TYPE}
       -DCMAKE_CXX_FLAGS=${GTEST_CMAKE_CXX_FLAGS}
       -DCMAKE_CXX_FLAGS_${UPPERCASE_BUILD_TYPE}=${GTEST_CMAKE_CXX_FLAGS}
       -DCMAKE_INSTALL_LIBDIR=lib

cpp/examples/arrow/join_example.cc

@@ -63,7 +63,7 @@ arrow::Result<std::shared_ptr<arrow::dataset::Dataset>> CreateDataSetFromCSVData
   std::shared_ptr<arrow::io::InputStream> input;
   std::string csv_data = is_left ? kLeftRelationCsvData : kRightRelationCsvData;
   std::cout << csv_data << std::endl;
-  arrow::util::string_view sv = csv_data;
+  std::string_view sv = csv_data;
   input = std::make_shared<arrow::io::BufferReader>(sv);
   auto read_options = arrow::csv::ReadOptions::Defaults();
   auto parse_options = arrow::csv::ParseOptions::Defaults();

cpp/examples/arrow/rapidjson_row_converter.cc

@@ -97,7 +97,7 @@ class RowBatchBuilder {
     for (int64_t i = 0; i < array.length(); ++i) {
       if (!array.IsNull(i)) {
         rapidjson::Value str_key(field_->name(), rows_[i].GetAllocator());
-        arrow::util::string_view value_view = array.Value(i);
+        std::string_view value_view = array.Value(i);
         rapidjson::Value value;
         value.SetString(value_view.data(),
                         static_cast<rapidjson::SizeType>(value_view.size()),

cpp/examples/parquet/parquet_stream_api/stream_reader_writer.cc

@@ -135,10 +135,10 @@ struct TestData {
     if (i % 2 == 0) return {};
     return "Str #" + std::to_string(i);
   }
-  static arrow::util::string_view GetStringView(const int i) {
+  static std::string_view GetStringView(const int i) {
     static std::string string;
     string = "StringView #" + std::to_string(i);
-    return arrow::util::string_view(string);
+    return std::string_view(string);
   }
   static const char* GetCharPtr(const int i) {
     static std::string string;
@@ -190,7 +190,7 @@ void WriteParquetFile() {
   os.SetMaxRowGroupSize(1000);
 
   for (auto i = 0; i < TestData::num_rows; ++i) {
-    // Output string using 3 different types: std::string, arrow::util::string_view and
+    // Output string using 3 different types: std::string, std::string_view and
     // const char *.
     switch (i % 3) {
       case 0:

cpp/gdb_arrow.py

@@ -456,7 +456,7 @@ def value(self):
 
 class StdString:
     """
-    A `std::string` (or possibly `string_view`) value.
+    A `std::string` (or possibly `std::string_view`) value.
     """
 
     def __init__(self, val):
@@ -2163,23 +2163,6 @@ def to_string(self):
         return f"arrow::Result<{data_type}>({inner})"
 
 
-class StringViewPrinter:
-    """
-    Pretty-printer for arrow::util::string_view.
-    """
-
-    def __init__(self, name, val):
-        self.val = val
-
-    def to_string(self):
-        size = int(self.val['size_'])
-        if size == 0:
-            return f"arrow::util::string_view of size 0"
-        else:
-            data = bytes_literal(self.val['data_'], size)
-            return f"arrow::util::string_view of size {size}, {data}"
-
-
 class FieldPrinter:
     """
     Pretty-printer for arrow::Field.
@@ -2397,8 +2380,6 @@ def to_string(self):
     "arrow::SimpleTable": TablePrinter,
     "arrow::Status": StatusPrinter,
     "arrow::Table": TablePrinter,
-    "arrow::util::string_view": StringViewPrinter,
-    "nonstd::sv_lite::basic_string_view": StringViewPrinter,
 }
 
 

cpp/src/arrow/adapters/orc/util.cc

@@ -19,6 +19,7 @@
 
 #include <cmath>
 #include <string>
+#include <string_view>
 #include <vector>
 
 #include "arrow/array/builder_base.h"
@@ -30,7 +31,6 @@
 #include "arrow/util/checked_cast.h"
 #include "arrow/util/decimal.h"
 #include "arrow/util/range.h"
-#include "arrow/util/string_view.h"
 #include "arrow/visit_data_inline.h"
 
 #include "orc/Exceptions.hh"
@@ -462,7 +462,7 @@ struct Appender<DataType, liborc::StringVectorBatch> {
     running_arrow_offset++;
     return Status::OK();
   }
-  Status VisitValue(util::string_view v) {
+  Status VisitValue(std::string_view v) {
     batch->notNull[running_orc_offset] = true;
     COffsetType data_length = 0;
     batch->data[running_orc_offset] = reinterpret_cast<char*>(
@@ -486,7 +486,7 @@ struct Appender<Decimal128Type, liborc::Decimal64VectorBatch> {
     running_arrow_offset++;
     return Status::OK();
   }
-  Status VisitValue(util::string_view v) {
+  Status VisitValue(std::string_view v) {
     batch->notNull[running_orc_offset] = true;
     const Decimal128 dec_value(array.GetValue(running_arrow_offset));
     batch->values[running_orc_offset] = static_cast<int64_t>(dec_value.low_bits());
@@ -507,7 +507,7 @@ struct Appender<Decimal128Type, liborc::Decimal128VectorBatch> {
     running_arrow_offset++;
     return Status::OK();
   }
-  Status VisitValue(util::string_view v) {
+  Status VisitValue(std::string_view v) {
     batch->notNull[running_orc_offset] = true;
     const Decimal128 dec_value(array.GetValue(running_arrow_offset));
     batch->values[running_orc_offset] =
@@ -557,7 +557,7 @@ struct FixedSizeBinaryAppender {
     running_arrow_offset++;
     return Status::OK();
   }
-  Status VisitValue(util::string_view v) {
+  Status VisitValue(std::string_view v) {
     batch->notNull[running_orc_offset] = true;
     batch->data[running_orc_offset] = reinterpret_cast<char*>(
         const_cast<uint8_t*>(array.GetValue(running_arrow_offset)));

cpp/src/arrow/array/array_binary.h

@@ -23,6 +23,7 @@
 #include <cstdint>
 #include <memory>
 #include <string>
+#include <string_view>
 #include <vector>
 
 #include "arrow/array/array_base.h"
@@ -32,7 +33,6 @@
 #include "arrow/type.h"
 #include "arrow/util/checked_cast.h"
 #include "arrow/util/macros.h"
-#include "arrow/util/string_view.h"  // IWYU pragma: export
 #include "arrow/util/visibility.h"
 
 namespace arrow {
@@ -67,15 +67,15 @@ class BaseBinaryArray : public FlatArray {
   ///
   /// \param i the value index
   /// \return the view over the selected value
-  util::string_view GetView(int64_t i) const {
+  std::string_view GetView(int64_t i) const {
     // Account for base offset
     i += data_->offset;
     const offset_type pos = raw_value_offsets_[i];
-    return util::string_view(reinterpret_cast<const char*>(raw_data_ + pos),
-                             raw_value_offsets_[i + 1] - pos);
+    return std::string_view(reinterpret_cast<const char*>(raw_data_ + pos),
+                            raw_value_offsets_[i + 1] - pos);
   }
 
-  std::optional<util::string_view> operator[](int64_t i) const {
+  std::optional<std::string_view> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 
@@ -84,7 +84,7 @@ class BaseBinaryArray : public FlatArray {
   ///
   /// \param i the value index
   /// \return the view over the selected value
-  util::string_view Value(int64_t i) const { return GetView(i); }
+  std::string_view Value(int64_t i) const { return GetView(i); }
 
   /// \brief Get binary value as a std::string
   ///
@@ -236,11 +236,11 @@ class ARROW_EXPORT FixedSizeBinaryArray : public PrimitiveArray {
   const uint8_t* GetValue(int64_t i) const;
   const uint8_t* Value(int64_t i) const { return GetValue(i); }
 
-  util::string_view GetView(int64_t i) const {
-    return util::string_view(reinterpret_cast<const char*>(GetValue(i)), byte_width());
+  std::string_view GetView(int64_t i) const {
+    return std::string_view(reinterpret_cast<const char*>(GetValue(i)), byte_width());
   }
 
-  std::optional<util::string_view> operator[](int64_t i) const {
+  std::optional<std::string_view> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 

cpp/src/arrow/array/array_binary_test.cc

@@ -19,6 +19,7 @@
 #include <cstring>
 #include <memory>
 #include <string>
+#include <string_view>
 #include <vector>
 
 #include <gmock/gmock-matchers.h>
@@ -37,7 +38,6 @@
 #include "arrow/util/bit_util.h"
 #include "arrow/util/bitmap_builders.h"
 #include "arrow/util/checked_cast.h"
-#include "arrow/util/string_view.h"
 #include "arrow/visit_data_inline.h"
 
 namespace arrow {
@@ -63,7 +63,7 @@ void CheckStringArray(const ArrayType& array, const std::vector<std::string>& st
       auto view = array.GetView(i);
       ASSERT_EQ(value_pos, array.value_offset(i));
       ASSERT_EQ(strings[j].size(), view.size());
-      ASSERT_EQ(util::string_view(strings[j]), view);
+      ASSERT_EQ(std::string_view(strings[j]), view);
       value_pos += static_cast<int32_t>(view.size());
     } else {
       ASSERT_TRUE(array.IsNull(i));
@@ -256,7 +256,7 @@ class TestStringArray : public ::testing::Test {
   }
 
   Status ValidateFull(int64_t length, std::vector<offset_type> offsets,
-                      util::string_view data, int64_t offset = 0) {
+                      std::string_view data, int64_t offset = 0) {
     ArrayType arr(length, Buffer::Wrap(offsets), std::make_shared<Buffer>(data),
                   /*null_bitmap=*/nullptr, /*null_count=*/0, offset);
     return arr.ValidateFull();
@@ -373,7 +373,7 @@ class TestUTF8Array : public ::testing::Test {
   using ArrayType = typename TypeTraits<TypeClass>::ArrayType;
 
   Status ValidateUTF8(int64_t length, std::vector<offset_type> offsets,
-                      util::string_view data, int64_t offset = 0) {
+                      std::string_view data, int64_t offset = 0) {
     ArrayType arr(length, Buffer::Wrap(offsets), std::make_shared<Buffer>(data),
                   /*null_bitmap=*/nullptr, /*null_count=*/0, offset);
     return arr.ValidateUTF8();
@@ -867,12 +867,12 @@ struct BinaryAppender {
     return Status::OK();
   }
 
-  Status VisitValue(util::string_view v) {
+  Status VisitValue(std::string_view v) {
     data.push_back(v);
     return Status::OK();
   }
 
-  std::vector<util::string_view> data;
+  std::vector<std::string_view> data;
 };
 
 template <typename T>