zhangstar333 commented on code in PR #42055:
URL: https://github.com/apache/doris/pull/42055#discussion_r1811910032
##########
be/src/vec/functions/function_string.h:
##########
@@ -2369,6 +2369,124 @@ class FunctionSplitByString : public IFunction {
}
};
+class FunctionCountSubString : public IFunction {
+public:
+ static constexpr auto name = "count_substrings";
+
+ static FunctionPtr create() { return
std::make_shared<FunctionCountSubString>(); }
+ using NullMapType = PaddedPODArray<UInt8>;
+
+ String get_name() const override { return name; }
+
+ bool is_variadic() const override { return false; }
+
+ size_t get_number_of_arguments() const override { return 2; }
+
+ DataTypePtr get_return_type_impl(const DataTypes& arguments) const
override {
+ DCHECK(is_string(arguments[0]))
+ << "first argument for function: " << name << " should be
string"
+ << " and arguments[0] is " << arguments[0]->get_name();
+ DCHECK(is_string(arguments[1]))
+ << "second argument for function: " << name << " should be
string"
+ << " and arguments[1] is " << arguments[1]->get_name();
+ return std::make_shared<DataTypeInt32>();
+ }
+
+ Status execute_impl(FunctionContext* /*context*/, Block& block, const
ColumnNumbers& arguments,
+ size_t result, size_t input_rows_count) const override
{
+ DCHECK_EQ(arguments.size(), 2);
+ const auto& [src_column, left_const] =
+ unpack_if_const(block.get_by_position(arguments[0]).column);
+ const auto& [right_column, right_const] =
+ unpack_if_const(block.get_by_position(arguments[1]).column);
+
+ const auto* col_left =
check_and_get_column<ColumnString>(src_column.get());
+ if (!col_left) {
+ return Status::InternalError("Left operator of function {} can not
be {}", get_name(),
+
block.get_by_position(arguments[0]).type->get_name());
+ }
+
+ const auto* col_right =
check_and_get_column<ColumnString>(right_column.get());
+ if (!col_right) {
+ return Status::InternalError("Right operator of function {} can
not be {}", get_name(),
+
block.get_by_position(arguments[1]).type->get_name());
+ }
+
+ auto dest_column_ptr = ColumnInt32::create(input_rows_count, 0);
+ // count_substring(ColumnString, "xxx")
+ if (right_const) {
+ _execute_constant_pattern(*col_left, col_right->get_data_at(0),
+ dest_column_ptr->get_data(),
input_rows_count);
+ } else if (left_const) {
+ // count_substring("xxx", ColumnString)
+ _execute_constant_src_string(col_left->get_data_at(0), *col_right,
+ dest_column_ptr->get_data(),
input_rows_count);
+ } else {
+ // count_substring(ColumnString, ColumnString)
+ _execute_vector(*col_left, *col_right,
dest_column_ptr->get_data(), input_rows_count);
+ }
+
+ block.replace_by_position(result, std::move(dest_column_ptr));
+ return Status::OK();
+ }
+
+private:
+ void _execute_constant_pattern(const ColumnString& src_column_string,
+ const StringRef& pattern_ref,
+ ColumnInt32::Container& dest_column_data,
+ size_t input_rows_count) const {
+ for (size_t i = 0; i < input_rows_count; i++) {
+ const StringRef str_ref = src_column_string.get_data_at(i);
+ dest_column_data[i] = find_str_count(str_ref, pattern_ref);
+ }
+ }
+
+ void _execute_vector(const ColumnString& src_column_string, const
ColumnString& pattern_column,
+ ColumnInt32::Container& dest_column_data, size_t
input_rows_count) const {
+ for (size_t i = 0; i < input_rows_count; i++) {
+ const StringRef pattern_ref = pattern_column.get_data_at(i);
+ const StringRef str_ref = src_column_string.get_data_at(i);
+ dest_column_data[i] = find_str_count(str_ref, pattern_ref);
+ }
+ }
+
+ void _execute_constant_src_string(const StringRef& str_ref, const
ColumnString& pattern_col,
+ ColumnInt32::Container& dest_column_data,
+ size_t input_rows_count) const {
+ for (size_t i = 0; i < input_rows_count; ++i) {
+ const StringRef pattern_ref = pattern_col.get_data_at(i);
+ dest_column_data[i] = find_str_count(str_ref, pattern_ref);
+ }
+ }
+
+ size_t find_pos(size_t pos, const StringRef str_ref, const StringRef
pattern_ref) const {
+ size_t old_size = pos;
+ size_t str_size = str_ref.size;
+ while (pos < str_size && memcmp_small_allow_overflow15(str_ref.data +
pos, pattern_ref.data,
Review Comment:
Almost the same
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
