| 12345678910111213141516171819202122232425262728293031323334353637383940414243 |
- """Rebuild the runtime whitelist from the final UI tags, never legacy names."""
- import hashlib
- import json
- from pathlib import Path
- from openpyxl import load_workbook
- def build_taxonomy(source):
- source = Path(source)
- workbook = load_workbook(source, read_only=True, data_only=True)
- try:
- sheet = workbook['Sheet1']
- rows = iter(sheet.values)
- header = next(rows)
- category_index = header.index('TAG_CATEGORY')
- tag_index = header.index('最终使用TAG')
- categories = {}
- count = 0
- for row in rows:
- if not any(value is not None for value in row):
- continue
- category, tag = row[category_index], row[tag_index]
- if not all(isinstance(value, str) and value.strip() == value and value
- for value in (category, tag)):
- raise ValueError('invalid_final_tag_row')
- tags = categories.setdefault(category, [])
- if tag not in tags:
- tags.append(tag)
- count += 1
- return {'source': source.as_posix(),
- 'sha256': hashlib.sha256(source.read_bytes()).hexdigest(),
- 'sheet': 'Sheet1', 'tag_column': '最终使用TAG',
- 'row_count': count, 'categories': categories}
- finally:
- workbook.close()
- if __name__ == '__main__':
- snapshot = build_taxonomy(Path('docs/法人标签表.xlsx'))
- Path(__file__).with_name('taxonomy.json').write_text(
- json.dumps(snapshot, ensure_ascii=False, indent=2) + '\n', encoding='utf-8')
|
