ARROW-12670: [C++] Fix extract_regex output after non-matching values

Closes #10287 from pitrou/ARROW-12670-extract-regex-nulls Authored-by: Antoine Pitrou <antoine@python.org> Signed-off-by: Benjamin Kietzman <bengilgit@gmail.com>
apache · May 17, 2021 · af476c6 · af476c6
1 parent 6a2915e
commit af476c6
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 7 deletions.
diff --git a/cpp/src/arrow/compute/kernels/scalar_string.cc b/cpp/src/arrow/compute/kernels/scalar_string.cc
@@ -1622,20 +1622,15 @@ struct ExtractRegex : public ExtractRegexBase {
             checked_cast<BuilderType*>(struct_builder->field_builder(i)));
       }
 
-      auto visit_null = [&]() {
-        for (int i = 0; i < group_count; i++) {
-          RETURN_NOT_OK(field_builders[i]->AppendEmptyValue());
-        }
-        return struct_builder->AppendNull();
-      };
+      auto visit_null = [&]() { return struct_builder->AppendNull(); };
       auto visit_value = [&](util::string_view s) {
         if (Match(s)) {
           for (int i = 0; i < group_count; i++) {
             RETURN_NOT_OK(field_builders[i]->Append(ToStringView(found_values[i])));
           }
           return struct_builder->Append();
         } else {
-          return visit_null();
+          return struct_builder->AppendNull();
         }
       };
       const ArrayData& input = *batch[0].array();

diff --git a/cpp/src/arrow/compute/kernels/scalar_string_test.cc b/cpp/src/arrow/compute/kernels/scalar_string_test.cc
@@ -539,6 +539,10 @@ TYPED_TEST(TestStringKernels, ExtractRegex) {
       "extract_regex", R"(["a1", "b2", "c3", null])", type,
       R"([{"letter": "a", "digit": "1"}, {"letter": "b", "digit": "2"}, null, null])",
       &options);
+  this->CheckUnary(
+      "extract_regex", R"(["a1", "c3", null, "b2"])", type,
+      R"([{"letter": "a", "digit": "1"}, null, null, {"letter": "b", "digit": "2"}])",
+      &options);
   this->CheckUnary("extract_regex", R"(["a1", "b2"])", type,
                    R"([{"letter": "a", "digit": "1"}, {"letter": "b", "digit": "2"}])",
                    &options);