(text: Any)
| 733 | |
| 734 | |
| 735 | def _split_sentences(text: Any) -> List[str]: |
| 736 | normalized = _clean_pdf_evidence_text(text) |
| 737 | if not normalized: |
| 738 | return [] |
| 739 | |
| 740 | parts = re.split(r"(?<=[。！？!?\.])\s+|(?<=;)\s+", normalized) |
| 741 | sentences: List[str] = [] |
| 742 | for part in parts: |
| 743 | sentence = re.sub(r"\s+", " ", part).strip() |
| 744 | if len(sentence) >= 12 and not _is_noisy_pdf_evidence_text(sentence, min_chars=12): |
| 745 | sentences.append(sentence) |
| 746 | return sentences |
| 747 | |
| 748 | |
| 749 | def _first_non_empty(*values: Any) -> str: |
no test coverage detected