Method GetUTF8Text

deps/tesseract/ccmain/resultiterator.cpp:556–597

* Returns the null terminated UTF-8 encoded text string for the current * object at the given level. Use delete [] to free after use. */

 * object at the given level. Use delete [] to free after use.
 */
char* ResultIterator::GetUTF8Text(PageIteratorLevel level) const {
  if (it_->word() == NULL) return NULL;  // Already at the end!
  STRING text;
  switch (level) {
    case RIL_BLOCK:
      {
        ResultIterator pp(*this);
        do {
          pp.AppendUTF8ParagraphText(&text);
        } while (pp.Next(RIL_PARA) && pp.it_->block() == it_->block());
      }
      break;
    case RIL_PARA:
      AppendUTF8ParagraphText(&text);
      break;
    case RIL_TEXTLINE:
      {
        ResultIterator it(*this);
        it.MoveToLogicalStartOfTextline();
        it.IterateAndAppendUTF8TextlineText(&text);
      }
      break;
    case RIL_WORD:
      AppendUTF8WordText(&text);
      break;
    case RIL_SYMBOL:
      {
        bool reading_direction_is_ltr =
          current_paragraph_is_ltr_ ^ in_minor_direction_;
        if (at_beginning_of_minor_run_) {
          text += reading_direction_is_ltr ? kLRM : kRLM;
        }
        text = it_->word()->BestUTF8(blob_index_, !reading_direction_is_ltr);
        if (IsAtFinalSymbolOfWord()) AppendSuffixMarks(&text);
      }
      break;
  }
  int length = text.length() + 1;
  char* result = new char[length];
  strncpy(result, text.string(), length);
  return result;
}

void ResultIterator::AppendUTF8WordText(STRING *text) const {
  if (!it_->word()) return;