Replace message parsing with throwing more specific exceptions

cpp/include/cudf/binaryop.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -108,6 +108,7 @@ enum class binary_operator : int32_t {
  * @throw cudf::logic_error if @p output_type dtype isn't fixed-width
  * @throw cudf::logic_error if @p output_type dtype isn't boolean for comparison and logical
  * operations.
+ * @throw cudf::data_type_error if the operation is not supported for the types of @p lhs and @p rhs
  */
 std::unique_ptr<column> binary_operation(
   scalar const& lhs,
@@ -136,6 +137,7 @@ std::unique_ptr<column> binary_operation(
  * @throw cudf::logic_error if @p output_type dtype isn't fixed-width
  * @throw cudf::logic_error if @p output_type dtype isn't boolean for comparison and logical
  * operations.
+ * @throw cudf::data_type_error if the operation is not supported for the types of @p lhs and @p rhs
  */
 std::unique_ptr<column> binary_operation(
   column_view const& lhs,
@@ -163,6 +165,7 @@ std::unique_ptr<column> binary_operation(
  * @throw cudf::logic_error if @p output_type dtype isn't boolean for comparison and logical
  * operations.
  * @throw cudf::logic_error if @p output_type dtype isn't fixed-width
+ * @throw cudf::data_type_error if the operation is not supported for the types of @p lhs and @p rhs
  */
 std::unique_ptr<column> binary_operation(
   column_view const& lhs,

cpp/include/cudf/concatenate.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -50,8 +50,8 @@ rmm::device_buffer concatenate_masks(
 /**
  * @brief Concatenates multiple columns into a single column.
  *
- * @throws cudf::logic_error
- * If types of the input columns mismatch
+ * @throws cudf::logic_error If types of the input columns mismatch
+ * @throws std::overflow_error If the the total number of output rows exceeds cudf::size_type
  *
  * @param columns_to_concat host_span of column views to be concatenated into a single column
  * @param mr Device memory resource used to allocate the returned column's device memory
@@ -80,8 +80,8 @@ std::unique_ptr<column> concatenate(
  * column_view tc1 = (t->view()).column(1); //Contains {0,1,2,3,4,5,6,7}
  * ```
  *
- * @throws cudf::logic_error
- * If number of columns mismatch
+ * @throws cudf::logic_error If number of columns mismatch
+ * @throws std::overflow_error If the the total number of output rows exceeds cudf::size_type
  *
  * @param tables_to_concat host_span of table views to be concatenated into a single table
  * @param mr Device memory resource used to allocate the returned table's device memory

cpp/include/cudf/lists/combine.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -80,10 +80,10 @@ std::unique_ptr<column> concatenate_rows(
  * r is [ {1, 2, 3, 4, 5}, {6, 7, 8, 9} ]
  * @endcode
  *
- * @throws cudf::logic_error if the input column is not at least two-level depth lists column (i.e.,
- *         each row must be a list of list).
+ * @throws std::invalid_argument if the input column is not at least two-level depth lists column
+ * (i.e., each row must be a list of list).
  * @throws cudf::logic_error if the input lists column contains nested typed entries that are not
- *         lists.
+ * lists.
  *
  * @param input The lists column containing lists of list elements to concatenate.
  * @param null_policy The parameter to specify whether a null list element will be ignored from

cpp/include/cudf/lists/contains.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -126,9 +126,7 @@ enum class duplicate_find_option : int32_t {
  * @param mr Device memory resource used to allocate the returned column's device memory.
  * @return std::unique_ptr<column> INT32 column of `n` rows with the location of the `search_key`
  *
- * @throw cudf::logic_error If `search_key` type does not match the element type in `lists`
- * @throw cudf::logic_error If `search_key` is of a nested type, or `lists` contains nested
- * elements (LIST, STRUCT)
+ * @throw cudf::data_type_error If `search_keys` type does not match the element type in `lists`
  */
 std::unique_ptr<column> index_of(
   cudf::lists_column_view const& lists,
@@ -163,8 +161,7 @@ std::unique_ptr<column> index_of(
  * @return std::unique_ptr<column> INT32 column of `n` rows with the location of the `search_key`
  *
  * @throw cudf::logic_error If `search_keys` does not match `lists` in its number of rows
- * @throw cudf::logic_error If `search_keys` type does not match the element type in `lists`
- * @throw cudf::logic_error If `lists` or `search_keys` contains nested elements (LIST, STRUCT)
+ * @throw cudf::data_type_error If `search_keys` type does not match the element type in `lists`
  */
 std::unique_ptr<column> index_of(
   cudf::lists_column_view const& lists,

cpp/include/cudf/lists/gather.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -43,7 +43,7 @@ namespace lists {
  * @endcode
  *
  * @throws cudf::logic_error if `gather_map_list` size is not same as `source_column` size.
- * @throws cudf::logic_error if gather_map contains null values.
+ * @throws std::invalid_argument if gather_map contains null values.
  * @throws cudf::logic_error if gather_map is not list column of an index type.
  *
  * If indices in `gather_map_list` are outside the range `[-n, n)`, where `n` is the number of

cpp/include/cudf/strings/json.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -160,6 +160,8 @@ class get_json_object_options {
  * @param options Options for controlling the behavior of the function
  * @param mr Resource for allocating device memory.
  * @return New strings column containing the retrieved json object strings
+ *
+ * @throw std::invalid_argument if provided an invalid operator or an empty name
  */
 std::unique_ptr<cudf::column> get_json_object(
   cudf::strings_column_view const& col,

cpp/include/cudf/utilities/error.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -84,6 +84,29 @@ struct cuda_error : public std::runtime_error {
 struct fatal_cuda_error : public cuda_error {
   using cuda_error::cuda_error;  // Inherit constructors
 };
+
+/**
+ * @brief Exception thrown when an operation is attempted on an unsupported dtype.
+ *
+ * This exception should be thrown when an operation is attempted on an
+ * unsupported data_type. This exception should not be thrown directly and is
+ * instead thrown by the CUDF_EXPECTS or CUDF_FAIL macros.
+ */
+struct data_type_error : public std::invalid_argument {
+  /**
+   * @brief Constructs a data_type_error with the error message.
+   *
+   * @param message Message to be associated with the exception
+   */
+  data_type_error(char const* const message) : std::invalid_argument(message) {}
+
+  /**
+   * @brief Construct a new data_type_error object with error message
+   *
+   * @param message Message to be associated with the exception
+   */
+  data_type_error(std::string const& message) : std::invalid_argument(message) {}
+};
 /** @} */
 
 }  // namespace cudf

cpp/src/binaryop/binaryop.cpp

@@ -203,7 +203,7 @@ std::unique_ptr<column> binary_operation(LhsType const& lhs,
     return cudf::binops::compiled::string_null_min_max(lhs, rhs, op, output_type, stream, mr);
 
   if (not cudf::binops::compiled::is_supported_operation(output_type, lhs.type(), rhs.type(), op))
-    CUDF_FAIL("Unsupported operator for these types");
+    CUDF_FAIL("Unsupported operator for these types", cudf::data_type_error);
 
   if (cudf::is_fixed_point(lhs.type()) or cudf::is_fixed_point(rhs.type())) {
     cudf::binops::compiled::fixed_point_binary_operation_validation(

cpp/src/binaryop/compiled/equality_ops.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -26,7 +26,8 @@ void dispatch_equality_op(mutable_column_view& out,
                           rmm::cuda_stream_view stream)
 {
   CUDF_EXPECTS(op == binary_operator::EQUAL || op == binary_operator::NOT_EQUAL,
-               "Unsupported operator for these types");
+               "Unsupported operator for these types",
+               cudf::data_type_error);
   auto common_dtype = get_common_type(out.type(), lhs.type(), rhs.type());
   auto outd         = mutable_column_device_view::create(out, stream);
   auto lhsd         = column_device_view::create(lhs, stream);

cpp/src/binaryop/compiled/struct_binary_ops.cuh

@@ -149,7 +149,8 @@ void apply_struct_equality_op(mutable_column_view& out,
 {
   CUDF_EXPECTS(op == binary_operator::EQUAL || op == binary_operator::NOT_EQUAL ||
                  op == binary_operator::NULL_EQUALS,
-               "Unsupported operator for these types");
+               "Unsupported operator for these types",
+               cudf::data_type_error);
 
   auto tlhs = table_view{{lhs}};
   auto trhs = table_view{{rhs}};

cpp/src/copying/concatenate.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -228,7 +228,8 @@ std::unique_ptr<column> fused_concatenate(host_span<column_view const> views,
   auto const output_size  = std::get<3>(device_views);
 
   CUDF_EXPECTS(output_size <= static_cast<std::size_t>(std::numeric_limits<size_type>::max()),
-               "Total number of concatenated rows exceeds size_type range");
+               "Total number of concatenated rows exceeds size_type range",
+               std::overflow_error);
 
   // Allocate output
   auto const policy = has_nulls ? mask_policy::ALWAYS : mask_policy::NEVER;
@@ -398,7 +399,8 @@ void traverse_children::operator()<cudf::string_view>(host_span<column_view cons
     });
   // note:  output text must include "exceeds size_type range" for python error handling
   CUDF_EXPECTS(total_char_count <= static_cast<size_t>(std::numeric_limits<size_type>::max()),
-               "Total number of concatenated chars exceeds size_type range");
+               "Total number of concatenated chars exceeds size_type range",
+               std::overflow_error);
 }
 
 template <>
@@ -469,7 +471,8 @@ void bounds_and_type_check(host_span<column_view const> cols, rmm::cuda_stream_v
     });
   // note:  output text must include "exceeds size_type range" for python error handling
   CUDF_EXPECTS(total_row_count <= static_cast<size_t>(std::numeric_limits<size_type>::max()),
-               "Total number of concatenated rows exceeds size_type range");
+               "Total number of concatenated rows exceeds size_type range",
+               std::overflow_error);
 
   // traverse children
   cudf::type_dispatcher(cols.front().type(), traverse_children{}, cols, stream);

cpp/src/lists/combine/concatenate_list_elements.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -257,11 +257,13 @@ std::unique_ptr<column> concatenate_list_elements(column_view const& input,
                                                   rmm::mr::device_memory_resource* mr)
 {
   auto type = input.type();  // Column that is lists of lists.
-  CUDF_EXPECTS(type.id() == type_id::LIST, "Input column must be a lists column.");
+  CUDF_EXPECTS(
+    type.id() == type_id::LIST, "Input column must be a lists column.", std::invalid_argument);
 
   auto col = lists_column_view(input).child();  // Rows, which are lists.
   type     = col.type();
-  CUDF_EXPECTS(type.id() == type_id::LIST, "Rows of the input column must be lists.");
+  CUDF_EXPECTS(
+    type.id() == type_id::LIST, "Rows of the input column must be lists.", std::invalid_argument);
 
   col  = lists_column_view(col).child();  // The last level entries what we need to check.
   type = col.type();

cpp/src/lists/contains.cu

@@ -309,7 +309,8 @@ struct dispatch_index_of {
     auto const child = lists.child();
 
     CUDF_EXPECTS(child.type() == search_keys.type(),
-                 "Type/Scale of search key does not match list column element type.");
+                 "Type/Scale of search key does not match list column element type.",
+                 cudf::data_type_error);
     CUDF_EXPECTS(search_keys.type().id() != type_id::EMPTY, "Type cannot be empty.");
 
     auto constexpr search_key_is_scalar = std::is_same_v<SearchKeyType, cudf::scalar>;

cpp/src/lists/copying/segmented_gather.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -38,7 +38,7 @@ std::unique_ptr<column> segmented_gather(lists_column_view const& value_column,
 {
   CUDF_EXPECTS(is_index_type(gather_map.child().type()),
                "Gather map should be list column of index type");
-  CUDF_EXPECTS(!gather_map.has_nulls(), "Gather map contains nulls");
+  CUDF_EXPECTS(!gather_map.has_nulls(), "Gather map contains nulls", std::invalid_argument);
   CUDF_EXPECTS(value_column.size() == gather_map.size(),
                "Gather map and list column should be same size");
 

cpp/src/strings/json/json_path.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -588,7 +588,7 @@ class path_state : private parser {
         return path_operator{path_operator_type::CHILD_WILDCARD};
       } break;
 
-      default: CUDF_FAIL("Unrecognized JSONPath operator"); break;
+      default: CUDF_FAIL("Unrecognized JSONPath operator", std::invalid_argument); break;
     }
     return {path_operator_type::ERROR};
   }
@@ -624,7 +624,8 @@ class path_state : private parser {
     }
 
     // an empty name is not valid
-    CUDF_EXPECTS(name.size_bytes() > 0, "Invalid empty name in JSONPath query string");
+    CUDF_EXPECTS(
+      name.size_bytes() > 0, "Invalid empty name in JSONPath query string", std::invalid_argument);
 
     return true;
   }

cpp/tests/binaryop/binop-compiled-fixed_point-test.cpp

@@ -678,7 +678,7 @@ TYPED_TEST(FixedPointCompiledTest, FixedPointBinaryOpThrows)
   auto const col           = fp_wrapper<RepType>{{100, 300, 500, 700}, scale_type{-2}};
   auto const non_bool_type = cudf::data_type{cudf::type_to_id<decimalXX>(), -2};
   EXPECT_THROW(cudf::binary_operation(col, col, cudf::binary_operator::LESS, non_bool_type),
-               cudf::logic_error);
+               cudf::data_type_error);
 }
 
 TYPED_TEST(FixedPointCompiledTest, FixedPointBinaryOpModSimple)