Fixed spelling mistakes in libcudf documentation

cpp/benchmarks/common/generate_benchmark_input.cpp

@@ -108,7 +108,8 @@ size_t avg_element_bytes(data_profile const& profile, cudf::type_id tid)
 /**
  * @brief Functor that computes a random column element with the given data profile.
  *
- * The implementation is SFINAEd for diffent type groups. Currently only used for fixed-width types.
+ * The implementation is SFINAEd for different type groups. Currently only used for fixed-width
+ * types.
  */
 template <typename T, typename Enable = void>
 struct random_value_fn;

cpp/benchmarks/common/generate_benchmark_input.hpp

@@ -137,7 +137,7 @@ struct distribution_params<
 };
 
 /**
- * @brief Boolens are parameterized with the probability of getting `true` value.
+ * @brief Booleans are parameterized with the probability of getting `true` value.
  */
 template <typename T>
 struct distribution_params<T, typename std::enable_if_t<std::is_same<T, bool>::value>> {
@@ -195,7 +195,7 @@ std::vector<cudf::type_id> get_type_or_group(int32_t id);
  *
  * If an element of the input vector is a `cudf::type_id` enumerator, function return value simply
  * includes this type. If an element of the input vector is a `type_group_id` enumerator, function
- * return value includes all types coresponding to the group enumerator.
+ * return value includes all types corresponding to the group enumerator.
  *
  * @param ids Vector of integers equal to either a `cudf::type_id` enumerator or a `type_group_id`
  * enumerator.

cpp/benchmarks/join/generate_input_tables.cuh

@@ -141,7 +141,7 @@ __global__ void init_probe_tbl(key_type* const probe_tbl,
  * (e.g. device memory, zero copy memory or unified memory). Each value in the build table
  * will be from [0,rand_max] and if uniq_build_tbl_keys is true it is ensured that each value
  * will be uniq in the build table. Each value in the probe table will be also in the build
- * table with a propability of selectivity and a random number from
+ * table with a probability of selectivity and a random number from
  * [0,rand_max] \setminus \{build_tbl\} otherwise.
  *
  * @param[out] build_tbl            The build table to generate. Usually the smaller table used to
@@ -150,7 +150,7 @@ __global__ void init_probe_tbl(key_type* const probe_tbl,
  * @param[out] probe_tbl            The probe table to generate. Usually the larger table used to
  *                                  probe into the hash table created from the build table.
  * @param[in] build_tbl_size        number of keys in the build table
- * @param[in] selectivity           propability with which an element of the probe table is
+ * @param[in] selectivity           probability with which an element of the probe table is
  *                                  present in the build table.
  * @param[in] rand_max              maximum random number to generate. I.e. random numbers are
  *                                  integers from [0,rand_max].
@@ -169,7 +169,7 @@ void generate_input_tables(key_type* const build_tbl,
   // expense of not being that accurate with applying the selectivity an especially more memory
   // efficient implementations would be to partition the random numbers into two intervals and then
   // let one table choose random numbers from only one interval and the other only select with
-  // selectivity propability from the same interval and from the other in the other cases.
+  // selectivity probability from the same interval and from the other in the other cases.
 
   static_assert(std::is_signed<key_type>::value, "key_type needs to be signed for lottery to work");
 

cpp/docs/DEVELOPER_GUIDE.md

@@ -470,7 +470,7 @@ libcudf, and you should not use it in new code in libcudf without careful consid
 use `rmm::device_uvector` along with the utility factories in `device_factories.hpp`. These 
 utilities enable creation of `uvector`s from host-side vectors, or creating zero-initialized
 `uvector`s, so that they are as convenient to use as `device_vector`. Avoiding `device_vector` has
-a number of benefits, as described in the folling section on `rmm::device_uvector`.
+a number of benefits, as described in the following section on `rmm::device_uvector`.
 
 #### `rmm::device_uvector<T>`
 

cpp/docs/TESTING.md

@@ -67,7 +67,7 @@ not necessary for your test fixtures to inherit from it.
 
 Example:
 ```c++
-class MyTestFiture : public cudf::test::BaseFixture {...};
+class MyTestFixture : public cudf::test::BaseFixture {...};
 ```
 
 ## Typed Tests

cpp/include/cudf/aggregation.hpp

@@ -275,7 +275,7 @@ std::unique_ptr<Base> make_collect_set_aggregation(null_policy null_handling = n
  * @brief Factory to create a MERGE_LISTS aggregation.
  *
  * Given a lists column, this aggregation merges all the lists corresponding to the same key value
- * into one list. It is designed specificly to merge the partial results of multiple (distributed)
+ * into one list. It is designed specifically to merge the partial results of multiple (distributed)
  * groupby `COLLECT_LIST` aggregations into a final `COLLECT_LIST` result. As such, it requires the
  * input lists column to be non-nullable (the child column containing list entries is not subjected
  * to this requirement).
@@ -290,7 +290,7 @@ std::unique_ptr<Base> make_merge_lists_aggregation();
  * value into one list, then it drops all the duplicate entries in each lists, producing a lists
  * column containing non-repeated entries.
  *
- * This aggregation is designed specificly to merge the partial results of multiple (distributed)
+ * This aggregation is designed specifically to merge the partial results of multiple (distributed)
  * groupby `COLLECT_LIST` or `COLLECT_SET` aggregations into a final `COLLECT_SET` result. As such,
  * it requires the input lists column to be non-nullable (the child column containing list entries
  * is not subjected to this requirement).

cpp/include/cudf/column/column.hpp

@@ -293,7 +293,7 @@ class column {
   /**
    * @brief Implicit conversion operator to a `mutable_column_view`.
    *
-   * This allows pasing a `column` object into a function that accepts a
+   * This allows passing a `column` object into a function that accepts a
    *`mutable_column_view`. The conversion is automatic.
 
    * @note Creating a mutable view of a `column` invalidates the `column`'s

cpp/include/cudf/column/column_device_view.cuh

@@ -37,7 +37,7 @@
 
 /**
  * @file column_device_view.cuh
- * @brief Column device view class definitons
+ * @brief Column device view class definitions
  */
 
 namespace cudf {
@@ -541,7 +541,7 @@ class alignas(16) column_device_view : public detail::column_device_view_base {
    *
    * optional_begin with mode `DYNAMIC` defers the assumption of nullability to
    * runtime, with the user stating on construction of the iterator if column has nulls.
-   * `DYNAMIC` mode is nice when an algorithm is going to execute on mutliple
+   * `DYNAMIC` mode is nice when an algorithm is going to execute on multiple
    * iterators and you don't want to compile all the combinations of iterator types
    *
    * Example:

cpp/include/cudf/column/column_factories.hpp

@@ -399,7 +399,7 @@ std::unique_ptr<column> make_strings_column(
  *  one more than the total number of strings so the `offsets.back()` is the total number of bytes
  *  in the strings array. `offsets.front()` must always be 0 to point to the beginning of `strings`.
  * @param[in] null_mask Device span containing the null element indicator bitmask. Arrow format for
- *  nulls is used for interpeting this bitmask.
+ *  nulls is used for interpreting this bitmask.
  * @param[in] null_count The number of null string entries. If equal to `UNKNOWN_NULL_COUNT`, the
  *  null count will be computed dynamically on the first invocation of `column::null_count()`
  * @param[in] stream CUDA stream used for device memory operations and kernel launches.
@@ -428,7 +428,7 @@ std::unique_ptr<column> make_strings_column(
  *  strings are identified by the offsets and the nullmask.
  * @param[in] null_count The number of null string entries.
  * @param[in] null_mask The bits specifying the null strings in device memory. Arrow format for
- *  nulls is used for interpeting this bitmask.
+ *  nulls is used for interpreting this bitmask.
  * @param[in] stream CUDA stream used for device memory operations and kernel launches.
  * @param[in] mr Device memory resource used for allocation of the column's `null_mask` and children
  * columns' device memory.
@@ -491,7 +491,7 @@ std::unique_ptr<column> make_strings_column(
  *                     further nested.
  * @param[in] null_count The number of null list entries.
  * @param[in] null_mask The bits specifying the null lists in device memory.
- *                  Arrow format for nulls is used for interpeting this bitmask.
+ *                  Arrow format for nulls is used for interpreting this bitmask.
  * @param[in] stream Optional stream for use with all memory allocation
  *               and device kernels
  * @param[in] mr Optional resource to use for device memory

cpp/include/cudf/column/column_view.hpp

@@ -22,7 +22,7 @@
 
 /**
  * @file column_view.hpp
- * @brief column view class definitons
+ * @brief column view class definitions
  */
 
 namespace cudf {

cpp/include/cudf/detail/aggregation/aggregation.cuh

@@ -643,7 +643,7 @@ struct identity_initializer {
  * The `i`th column will be initialized with the identity value of the `i`th
  * aggregation operation in `aggs`.
  *
- * @throw cudf::logic_error if column type and corresponging agg are incompatible
+ * @throw cudf::logic_error if column type and corresponding agg are incompatible
  * @throw cudf::logic_error if column type is not fixed-width
  *
  * @param table The table of columns to initialize.

cpp/include/cudf/detail/indexalator.cuh

@@ -29,7 +29,7 @@ namespace detail {
 /**
  * @brief The base class for the input or output index normalizing iterator.
  *
- * This implementation uses CTRP to define the `input_indexalator` and the
+ * This implementation uses CRTP to define the `input_indexalator` and the
  * `output_indexalator` classes. This is so this class can manipulate the
  * uniquely typed subclass member variable `p_` directly without requiring
  * virtual functions since iterator instances will be copied to device memory.
@@ -241,7 +241,7 @@ struct base_indexalator {
  */
 struct input_indexalator : base_indexalator<input_indexalator> {
   friend struct indexalator_factory;
-  friend struct base_indexalator<input_indexalator>;  // for CTRP
+  friend struct base_indexalator<input_indexalator>;  // for CRTP
 
   using reference = size_type const;  // this keeps STL and thrust happy
 
@@ -326,7 +326,7 @@ struct input_indexalator : base_indexalator<input_indexalator> {
  */
 struct output_indexalator : base_indexalator<output_indexalator> {
   friend struct indexalator_factory;
-  friend struct base_indexalator<output_indexalator>;  // for CTRP
+  friend struct base_indexalator<output_indexalator>;  // for CRTP
 
   using reference = output_indexalator const&;  // required for output iterators
 

cpp/include/cudf/detail/iterator.cuh

@@ -177,7 +177,7 @@ auto make_null_replacement_iterator(column_device_view const& column,
  *
  * make_optional_iterator with mode `DYNAMIC` defers the assumption of nullability to
  * runtime, with the user stating on construction of the iterator if column has nulls.
- * `DYNAMIC` mode is nice when an algorithm is going to execute on mutliple
+ * `DYNAMIC` mode is nice when an algorithm is going to execute on multiple
  * iterators and you don't want to compile all the combinations of iterator types
  *
  * Example:
@@ -819,7 +819,7 @@ auto inline make_pair_iterator(scalar const& scalar_value)
  *
  * Else, if the scalar is null, then the value of `p.first` is undefined and `p.second == false`.
  *
- * The behaviour is undefined if the scalar is destroyed before iterator dereferencing.
+ * The behavior is undefined if the scalar is destroyed before iterator dereferencing.
  *
  * @throws cudf::logic_error if scalar datatype and Element type mismatch.
  * @throws cudf::logic_error if the returned iterator is dereferenced in host

cpp/include/cudf/detail/nvtx/nvtx3.hpp

@@ -54,7 +54,7 @@
  * \code{.cpp}
  * #include "nvtx3.hpp"
  * void some_function(){
- *    // Begins a NVTX range with the messsage "some_function"
+ *    // Begins a NVTX range with the message "some_function"
  *    // The range ends when some_function() returns and `r` is destroyed
  *    nvtx3::thread_range r{"some_function"};
  *
@@ -322,7 +322,7 @@
  * Example:
  * \code{.cpp}
  * // Create an `event_attributes` with the custom message "my message"
- * nvtx3::event_attributes attr{nvtx3::Mesage{"my message"}};
+ * nvtx3::event_attributes attr{nvtx3::message{"my message"}};
  *
  * // strings and string literals implicitly assumed to be a `nvtx3::message`
  * nvtx3::event_attributes attr{"my message"};
@@ -1267,7 +1267,7 @@ class registered_message {
  * nvtx3::thread_range range1{attr1};
  *
  * // `range2` contains message "message 2"
- * nvtx3::thread_range range2{nvtx3::Mesage{"message 2"}};
+ * nvtx3::thread_range range2{nvtx3::message{"message 2"}};
  *
  * // `std::string` and string literals are implicitly assumed to be
  * // the contents of an `nvtx3::message`
@@ -1525,7 +1525,7 @@ class payload {
  *
  * // For convenience, the arguments that can be passed to the
  * `event_attributes`
- * // constructor may be passed to the `domain_thread_range` contructor where
+ * // constructor may be passed to the `domain_thread_range` constructor where
  * // they will be forwarded to the `EventAttribute`s constructor
  * nvtx3::thread_range r{nvtx3::payload{42}, nvtx3::category{1}, "message"};
  * \endcode

cpp/include/cudf/detail/unary.hpp

@@ -31,7 +31,7 @@ namespace detail {
  * doesn't.
  *
  * @tparam InputIterator Iterator type for `begin` and `end`
- * @tparam Predicate A predicator type which will be evaludated
+ * @tparam Predicate A predicator type which will be evaluated
  * @param begin Beginning of the sequence of elements
  * @param end End of the sequence of elements
  * @param p Predicate to be applied to each element in `[begin,end)`

cpp/include/cudf/detail/utilities/device_atomics.cuh

@@ -161,7 +161,7 @@ struct genericAtomicOperationImpl<T, Op, 8> {
 
 // -----------------------------------------------------------------------
 // specialized functions for operators
-// `atomicAdd` supports int32, float, double (signed int64 is not supproted.)
+// `atomicAdd` supports int32, float, double (signed int64 is not supported.)
 // `atomicMin`, `atomicMax` support int32_t, int64_t
 // `atomicAnd`, `atomicOr`, `atomicXor` support int32_t, int64_t
 template <>

cpp/include/cudf/detail/valid_if.cuh

@@ -117,7 +117,7 @@ std::pair<rmm::device_buffer, size_type> valid_if(
 *         input ranges.
 
  * Given a set of bitmasks, `masks`, the state of bit `j` in mask `i` is
- * determined by `p( *(begin1 + i), *(begin2 + j))`. If the predivate evaluates
+ * determined by `p( *(begin1 + i), *(begin2 + j))`. If the predicate evaluates
  * to true, the the bit is set to `1`. If false, set to `0`.
  *
  * Example Arguments:

cpp/include/cudf/groupby.hpp

@@ -116,7 +116,7 @@ class groupby {
   /**
    * @brief Performs grouped aggregations on the specified values.
    *
-   * The values to aggregate and the aggregations to perform are specifed in an
+   * The values to aggregate and the aggregations to perform are specified in an
    * `aggregation_request`. Each request contains a `column_view` of values to
    * aggregate and a set of `aggregation`s to perform on those elements.
    *
@@ -173,7 +173,7 @@ class groupby {
   /**
    * @brief Performs grouped scans on the specified values.
    *
-   * The values to aggregate and the aggregations to perform are specifed in an
+   * The values to aggregate and the aggregations to perform are specified in an
    * `aggregation_request`. Each request contains a `column_view` of values to
    * aggregate and a set of `aggregation`s to perform on those elements.
    *

cpp/include/cudf/io/parquet.hpp

@@ -973,7 +973,7 @@ class chunked_parquet_writer_options_builder {
    * @brief Set to true if timestamps should be written as
    * int96 types instead of int64 types. Even though int96 is deprecated and is
    * not an internal type for cudf, it needs to be written for backwards
-   * compatability reasons.
+   * compatibility reasons.
    *
    * @param enabled Boolean value to enable/disable int96 timestamps.
    * @return this for chaining.

cpp/include/cudf/scalar/scalar.hpp

@@ -78,7 +78,7 @@ class scalar {
   /**
    * @brief Indicates whether the scalar contains a valid value.
    *
-   * @note Using the value when `is_valid() == false` is undefined behaviour. In addition, this
+   * @note Using the value when `is_valid() == false` is undefined behavior. In addition, this
    * function does a stream synchronization.
    *
    * @param stream CUDA stream used for device memory operations.

cpp/include/cudf/scalar/scalar_device_view.cuh

@@ -21,7 +21,7 @@
 
 /**
  * @file scalar_device_view.cuh
- * @brief Scalar device view class definitons
+ * @brief Scalar device view class definitions
  */
 
 namespace cudf {

cpp/include/cudf/sorting.hpp

@@ -187,7 +187,7 @@ std::unique_ptr<column> rank(
 /**
  * @brief Returns sorted order after sorting each segment in the table.
  *
- * If segment_offsets contains values larger than number of rows, behaviour is undefined.
+ * If segment_offsets contains values larger than number of rows, behavior is undefined.
  * @throws cudf::logic_error if `segment_offsets` is not `size_type` column.
  *
  * @param keys The table that determines the ordering of elements in each segment
@@ -214,7 +214,7 @@ std::unique_ptr<column> segmented_sorted_order(
 /**
  * @brief Performs a lexicographic segmented sort of a table
  *
- * If segment_offsets contains values larger than number of rows, behaviour is undefined.
+ * If segment_offsets contains values larger than number of rows, behavior is undefined.
  * @throws cudf::logic_error if `values.num_rows() != keys.num_rows()`.
  * @throws cudf::logic_error if `segment_offsets` is not `size_type` column.
  *

cpp/include/cudf/strings/combine.hpp

@@ -272,7 +272,7 @@ std::unique_ptr<column> join_list_elements(
  * delimited by the @p separator provided.
  *
  * A null list row will always result in a null string in the output row. Any non-null list row
- * having a null elenent will result in the corresponding output row to be null unless a
+ * having a null element will result in the corresponding output row to be null unless a
  * @p narep string is specified to be used in its place.
  *
  * If @p separate_nulls is set to `NO` and @p narep is valid then separators are not added to the

cpp/include/cudf/strings/repeat_strings.hpp

@@ -31,7 +31,7 @@ namespace strings {
  * @brief Repeat the given string scalar by a given number of times.
  *
  * For a given string scalar, an output string scalar is generated by repeating the input string by
- * a number of times given by the @p `repeat_times` parameter. If `repeat_times` is not a positve
+ * a number of times given by the @p `repeat_times` parameter. If `repeat_times` is not a positive
  * value, an empty (valid) string scalar will be returned. An invalid input scalar will always
  * result in an invalid output scalar regardless of the value of `repeat_times` parameter.
  *
@@ -42,7 +42,7 @@ namespace strings {
  * out is '123XYZ-123XYZ-123XYZ-'
  * @endcode
  *
- * @throw cudf::logic_error if the size of the ouput string scalar exceeds the maximum value that
+ * @throw cudf::logic_error if the size of the output string scalar exceeds the maximum value that
  *        can be stored by the index type
  *        (i.e., `input.size() * repeat_times > numeric_limits<size_type>::max()`).
  *
@@ -61,7 +61,7 @@ std::unique_ptr<string_scalar> repeat_strings(
  *
  * For a given strings column, an output strings column is generated by repeating each string from
  * the input by a number of times given by the @p `repeat_times` parameter. If `repeat_times` is not
- * a positve value, all the rows of the output strings column will be an empty string. Any null row
+ * a positive value, all the rows of the output strings column will be an empty string. Any null row
  * will result in a null row regardless of the value of `repeat_times` parameter.
  *
  * Note that this function cannot handle the cases when the size of the output column exceeds the

cpp/include/cudf/strings/replace.hpp

@@ -36,7 +36,7 @@ namespace strings {
  * input string. If not found, the output entry is just a copy of the
  * corresponding input string.
  *
- * Specifing an empty string for repl will essentially remove the target
+ * Specifying an empty string for repl will essentially remove the target
  * string if found in each string.
  *
  * Null string entries will return null output string entries.