Function ExtDetectLanguageSummaryCheckUTF8

cld2/internal/compact_lang_det.cc:317–355

Use this one. Hints are collected into a struct. Flags are passed in (normally zero). Also returns 3 internal language scores as a ratio to normal score for real text in that language. Scores close to 1.0 indicate normal text, while scores far away from 1.0 indicate badly-skewed text or gibberish Returns a vector of chunks in different languages, so that caller may spell-check, translate, or ot

// bytes that are valid UTF-8. If this is < buffer_length, there is invalid
// input starting at the following byte.
Language ExtDetectLanguageSummaryCheckUTF8(
                        const char* buffer,
                        int buffer_length,
                        bool is_plain_text,
                        const CLDHints* cld_hints,
                        int flags,
                        Language* language3,
                        int* percent3,
                        double* normalized_score3,
                        ResultChunkVector* resultchunkvector,
                        int* text_bytes,
                        bool* is_reliable,
                        int* valid_prefix_bytes) {
  *valid_prefix_bytes = SpanInterchangeValid(buffer, buffer_length);
  if (*valid_prefix_bytes < buffer_length) {
    *is_reliable = false;
    return UNKNOWN_LANGUAGE;
  }

  bool allow_extended_lang = true;
  Language plus_one = UNKNOWN_LANGUAGE;

  Language lang = DetectLanguageSummaryV2(
                          buffer,
                          buffer_length,
                          is_plain_text,
                          cld_hints,
                          allow_extended_lang,
                          flags,
                          plus_one,
                          language3,
                          percent3,
                          normalized_score3,
                          resultchunkvector,
                          text_bytes,
                          is_reliable);
  // Do not default to English
  return lang;
}

// Use this one ONLY if you can prove the the input text is valid UTF-8 by
// design because it went through a known-good conversion program.