Method Reset

be/src/exprs/string-functions-ir.cc:700–733

}

void StringFunctions::TrimContext::Reset(const StringVal& chars_to_trim) {
  single_byte_chars_.reset();
  double_byte_chars_.clear();
  triple_byte_chars_.clear();
  quadruple_byte_chars_.clear();

  if (!utf8_mode_) {
    for (size_t i = 0; i < chars_to_trim.len; ++i) {
      single_byte_chars_.set(chars_to_trim.ptr[i], true);
    }
    return;
  }

  for (size_t i = 0, char_size = 0; i < chars_to_trim.len; i += char_size) {
    char_size = BitUtil::NumBytesInUtf8Encoding(chars_to_trim.ptr[i]);

    // If the remaining number of bytes does not match the number of bytes specified by
    // the UTF-8 character, we may have encountered an illegal UTF-8 character.
    // In order to prevent subsequent data access from going out of bounds, restrictions
    // are placed here to ensure that accessing pointers to multi-byte characters is
    // always safe.
    if (UNLIKELY(i + char_size > chars_to_trim.len)) {
      char_size = chars_to_trim.len - i;
    }

    switch (char_size) {
      case 1: single_byte_chars_.set(chars_to_trim.ptr[i], true); break;
      case 2: double_byte_chars_.push_back(&chars_to_trim.ptr[i]); break;
      case 3: triple_byte_chars_.push_back(&chars_to_trim.ptr[i]); break;
      case 4: quadruple_byte_chars_.push_back(&chars_to_trim.ptr[i]); break;
      default: DCHECK(false); break;
    }
  }
}

bool StringFunctions::TrimContext::Contains(const uint8_t* utf8_char, int len) const {
  auto eq = [&](const uint8_t* c){ return memcmp(c, utf8_char, len) == 0; };