[GitHub] [arrow] jpedroantunes commented on a change in pull request #10173: ARROW-12567: [C++][Gandiva] Implement LPAD and RPAD functions for string input values

[GitBox <gi...@apache.org>]

jpedroantunes commented on a change in pull request #10173:
URL: https://github.com/apache/arrow/pull/10173#discussion_r622568548



##########
File path: cpp/src/gandiva/precompiled/string_ops.cc
##########
@@ -1422,6 +1422,239 @@ const char* replace_utf8_utf8_utf8(gdv_int64 context, const char* text,
                                              out_len);
 }
 
+FORCE_INLINE
+const char* lpad(gdv_int64 context, const char* text, gdv_int32 text_len,
+                 gdv_int32 return_length, const char* fill_text, gdv_int32 fill_text_len,
+                 gdv_int32* out_len) {
+  // if the text length or the defined return length (number of characters to return)
+  // is <=0, then return an empty string.
+  if (text_len == 0 || return_length <= 0) {
+    *out_len = 0;
+    return "";
+  }
+
+  // initially counts the number of utf8 characters in the defined text and fill_text
+  int32_t text_char_count = utf8_length(context, text, text_len);
+  int32_t fill_char_count = utf8_length(context, fill_text, fill_text_len);
+  // text_char_count is zero if input has invalid utf8 char
+  // fill_char_count is zero if fill_text_len is > 0 and its value has invalid utf8 char
+  if (text_char_count == 0 || (fill_text_len > 0 && fill_char_count == 0)) {

Review comment:
       @projjal Do you think it should have a new function signature with the same behavior but with only this difference? 
   Or it is better to modify this function for this behavior?




-- 
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:
users@infra.apache.org