GH-42247: [C++] Support casting to and from utf8_view/binary_view (#4…

…3302) ### Rationale for this change We need casts between string (binary) and string-view (binary-view) types since they are semantically equivalent. ### What changes are included in this PR? - Add `is_binary_view_like()` type predicate - Add `BinaryViewTypes()` list including `STRING_VIEW/BINARY_VIEW` - New cast kernels ### Are these changes tested? Yes, but test coverage might be improved. ### Are there any user-facing changes? More casts are available. * GitHub Issue: #42247 Lead-authored-by: Felipe Oliveira Carvalho <[email protected]> Co-authored-by: mwish <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
apache · Sep 12, 2024 · 85fc3eb · 85fc3eb
1 parent 0c2891d
commit 85fc3eb
Show file tree

Hide file tree

Showing 12 changed files with 473 additions and 76 deletions.
diff --git a/cpp/src/arrow/compute/kernels/codegen_internal.h b/cpp/src/arrow/compute/kernels/codegen_internal.h
@@ -133,7 +133,8 @@ struct GetViewType<Type, enable_if_has_c_type<Type>> {
 
 template <typename Type>
 struct GetViewType<Type, enable_if_t<is_base_binary_type<Type>::value ||
-                                     is_fixed_size_binary_type<Type>::value>> {
+                                     is_fixed_size_binary_type<Type>::value ||
+                                     is_binary_view_like_type<Type>::value>> {
   using T = std::string_view;
   using PhysicalType = T;
 
@@ -1265,6 +1266,22 @@ ArrayKernelExec GenerateVarBinary(detail::GetTypeId get_id) {
   }
 }
 
+// Generate a kernel given a templated functor for binary-view types. Generates a
+// single kernel for binary/string-view.
+//
+// See "Numeric" above for description of the generator functor
+template <template <typename...> class Generator, typename Type0, typename... Args>
+ArrayKernelExec GenerateVarBinaryViewBase(detail::GetTypeId get_id) {
+  switch (get_id.id) {
+    case Type::BINARY_VIEW:
+    case Type::STRING_VIEW:
+      return Generator<Type0, BinaryViewType, Args...>::Exec;
+    default:
+      DCHECK(false);
+      return nullptr;
+  }
+}
+
 // Generate a kernel given a templated functor for temporal types
 //
 // See "Numeric" above for description of the generator functor

diff --git a/cpp/src/arrow/compute/kernels/scalar_cast_boolean.cc b/cpp/src/arrow/compute/kernels/scalar_cast_boolean.cc
@@ -63,6 +63,12 @@ std::vector<std::shared_ptr<CastFunction>> GetBooleanCasts() {
                                                  BooleanType, ParseBooleanString>(*ty);
     DCHECK_OK(func->AddKernel(ty->id(), {ty}, boolean(), exec));
   }
+  for (const auto& ty : BinaryViewTypes()) {
+    ArrayKernelExec exec =
+        GenerateVarBinaryViewBase<applicator::ScalarUnaryNotNull, BooleanType,
+                                  ParseBooleanString>(*ty);
+    DCHECK_OK(func->AddKernel(ty->id(), {ty}, boolean(), exec));
+  }
   return {func};
 }
 

diff --git a/cpp/src/arrow/compute/kernels/scalar_cast_internal.cc b/cpp/src/arrow/compute/kernels/scalar_cast_internal.cc
@@ -188,8 +188,6 @@ void CastNumberToNumberUnsafe(Type::type in_type, Type::type out_type,
 // ----------------------------------------------------------------------
 
 Status UnpackDictionary(KernelContext* ctx, const ExecSpan& batch, ExecResult* out) {
-  // TODO: is there an implementation more friendly to the "span" data structures?
-
   DictionaryArray dict_arr(batch[0].array.ToArrayData());
   const CastOptions& options = checked_cast<const CastState&>(*ctx->state()).options;
 
@@ -281,6 +279,8 @@ void AddZeroCopyCast(Type::type in_type_id, InputType in_type, OutputType out_ty
 }
 
 static bool CanCastFromDictionary(Type::type type_id) {
+  /// TODO(GH-43010): add is_binary_view_like() here once array_take
+  /// can handle string-views
   return (is_primitive(type_id) || is_base_binary_like(type_id) ||
           is_fixed_size_binary(type_id));
 }
@@ -297,9 +297,6 @@ void AddCommonCasts(Type::type out_type_id, OutputType out_ty, CastFunction* fun
   // From dictionary to this type
   if (CanCastFromDictionary(out_type_id)) {
     // Dictionary unpacking not implemented for boolean or nested types.
-    //
-    // XXX: Uses Take and does its own memory allocation for the moment. We can
-    // fix this later.
     DCHECK_OK(func->AddKernel(Type::DICTIONARY, {InputType(Type::DICTIONARY)}, out_ty,
                               UnpackDictionary, NullHandling::COMPUTED_NO_PREALLOCATE,
                               MemAllocation::NO_PREALLOCATE));

diff --git a/cpp/src/arrow/compute/kernels/scalar_cast_numeric.cc b/cpp/src/arrow/compute/kernels/scalar_cast_numeric.cc
@@ -313,7 +313,9 @@ struct ParseString {
 
 template <typename O, typename I>
 struct CastFunctor<
-    O, I, enable_if_t<(is_number_type<O>::value && is_base_binary_type<I>::value)>> {
+    O, I,
+    enable_if_t<(is_number_type<O>::value && (is_base_binary_type<I>::value ||
+                                              is_binary_view_like_type<I>::value))>> {
   static Status Exec(KernelContext* ctx, const ExecSpan& batch, ExecResult* out) {
     return applicator::ScalarUnaryNotNull<O, I, ParseString<O>>::Exec(ctx, batch, out);
   }
@@ -658,11 +660,15 @@ struct DecimalCastFunctor {
 };
 
 template <typename I>
-struct CastFunctor<Decimal128Type, I, enable_if_t<is_base_binary_type<I>::value>>
+struct CastFunctor<
+    Decimal128Type, I,
+    enable_if_t<is_base_binary_type<I>::value || is_binary_view_like_type<I>::value>>
     : public DecimalCastFunctor<Decimal128Type, I> {};
 
 template <typename I>
-struct CastFunctor<Decimal256Type, I, enable_if_t<is_base_binary_type<I>::value>>
+struct CastFunctor<
+    Decimal256Type, I,
+    enable_if_t<is_base_binary_type<I>::value || is_binary_view_like_type<I>::value>>
     : public DecimalCastFunctor<Decimal256Type, I> {};
 
 // ----------------------------------------------------------------------
@@ -708,6 +714,10 @@ void AddCommonNumberCasts(const std::shared_ptr<DataType>& out_ty, CastFunction*
     auto exec = GenerateVarBinaryBase<CastFunctor, OutType>(*in_ty);
     DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, out_ty, exec));
   }
+  for (const std::shared_ptr<DataType>& in_ty : BinaryViewTypes()) {
+    auto exec = GenerateVarBinaryViewBase<CastFunctor, OutType>(*in_ty);
+    DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, out_ty, exec));
+  }
 }
 
 template <typename OutType>
@@ -793,6 +803,10 @@ std::shared_ptr<CastFunction> GetCastToDecimal128() {
     auto exec = GenerateVarBinaryBase<CastFunctor, Decimal128Type>(in_ty->id());
     DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, sig_out_ty, std::move(exec)));
   }
+  for (const std::shared_ptr<DataType>& in_ty : BinaryViewTypes()) {
+    auto exec = GenerateVarBinaryViewBase<CastFunctor, Decimal128Type>(in_ty->id());
+    DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, sig_out_ty, std::move(exec)));
+  }
 
   // Cast from other decimal
   auto exec = CastFunctor<Decimal128Type, Decimal128Type>::Exec;
@@ -828,6 +842,10 @@ std::shared_ptr<CastFunction> GetCastToDecimal256() {
     auto exec = GenerateVarBinaryBase<CastFunctor, Decimal256Type>(in_ty->id());
     DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, sig_out_ty, std::move(exec)));
   }
+  for (const std::shared_ptr<DataType>& in_ty : BinaryViewTypes()) {
+    auto exec = GenerateVarBinaryViewBase<CastFunctor, Decimal256Type>(in_ty->id());
+    DCHECK_OK(func->AddKernel(in_ty->id(), {in_ty}, sig_out_ty, std::move(exec)));
+  }
 
   // Cast from other decimal
   auto exec = CastFunctor<Decimal256Type, Decimal128Type>::Exec;