lidavidm commented on a change in pull request #10494:
URL: https://github.com/apache/arrow/pull/10494#discussion_r649203237
##########
File path: cpp/src/arrow/compute/kernels/scalar_string.cc
##########
@@ -2288,6 +2288,164 @@ const FunctionDoc replace_substring_regex_doc(
{"strings"}, "ReplaceSubstringOptions");
#endif
+// ----------------------------------------------------------------------
+// Replace slice
+
+struct ReplaceSliceTransformBase : public StringTransformBase {
+ using State = OptionsWrapper<ReplaceSliceOptions>;
+
+ const ReplaceSliceOptions* options;
+
+ explicit ReplaceSliceTransformBase(const ReplaceSliceOptions& options)
+ : options{&options} {}
+
+ int64_t MaxCodeunits(int64_t ninputs, int64_t input_ncodeunits) override {
+ return ninputs * options->replacement.size() + input_ncodeunits;
+ }
+};
+
+struct AsciiReplaceSliceTransform : ReplaceSliceTransformBase {
+ using ReplaceSliceTransformBase::ReplaceSliceTransformBase;
+ int64_t Transform(const uint8_t* input, int64_t input_string_ncodeunits,
+ uint8_t* output) {
+ const auto& opts = *options;
+ int64_t before_slice = 0;
+ int64_t after_slice = 0;
+ uint8_t* output_start = output;
+
+ if (opts.start >= 0) {
+ // Count from left
+ before_slice = std::min<int64_t>(input_string_ncodeunits, opts.start);
+ } else {
+ // Count from right
+ before_slice = std::max<int64_t>(0, input_string_ncodeunits +
opts.start);
+ }
+ // Mimic Pandas: if stop would be before start, treat as 0-length slice
+ if (opts.stop >= 0) {
+ // Count from left
+ after_slice =
+ std::min<int64_t>(input_string_ncodeunits, std::max(opts.start,
opts.stop));
Review comment:
Good catch. I added a test case for this too.
--
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
