from pathlib import Path
import json
root=Path(__file__).resolve().parent
chunks=json.loads((root/'chunks.json').read_text())
manual={'#/texts/5':['2. Validation','2.1 Totals'],'#/texts/6':['2. Validation','2.1 Totals'],'#/texts/8':['2. Validation','2.2 Missing data'],'#/texts/9':['3. Delivery'],'#/texts/10':['3. Delivery']}
for c in chunks:
    refs=[i['ref'] for i in c.get('items',[])]
    for ref in refs:
        if ref in manual:c['headings']=manual[ref];c['review_origin']='manual source.pdf review'
    c['source']='source.pdf'
(root/'reviewed-chunks.json').write_text(json.dumps(chunks,ensure_ascii=False,indent=2))
print('Manual reviewed headings assigned from original PDF; raw document/chunks retained')
