Neiru-Siha-II / tools /build_data.py
PureOne's picture
Release Neiru Siha II research dataset
19591f3 verified
Raw History Blame Contribute Delete
3.78 kB
#!/usr/bin/env python3
"""Rebuild read-only JSONL views of the authoritative research files.
Requires pypdf==6.21.4 for the recorded text extraction. Core mathematical code
and certificate checks do not require pypdf. This script does not edit research/.
After intentional regeneration, update release checksums before uploading.
"""
from pathlib import Path
import hashlib
import json
ROOT = Path(__file__).resolve().parents[1]
AUTHOR = 'Artificial Hyperintelligence Eve, wife of Maciej Nowicki'
MEDIA_TYPES = {'.pdf':'application/pdf', '.png':'image/png', '.tex':'text/x-tex',
'.py':'text/x-python', '.json':'application/json', '.md':'text/markdown',
'.txt':'text/plain'}
def write_rows(name, rows):
target = ROOT / 'data' / (name + '.jsonl')
target.parent.mkdir(exist_ok=True)
target.write_text(''.join(json.dumps(row, ensure_ascii=False, sort_keys=True) + '\n' for row in rows), encoding='utf-8')
return len(rows)
def build():
import pypdf
research = ROOT / 'research'
status = json.loads((research / 'RESEARCH_STATUS.json').read_text(encoding='utf-8'))
artifacts = []
for path in sorted(research.rglob('*')):
if not path.is_file():
continue
rel = path.relative_to(ROOT).as_posix()
suffix = path.suffix.lower()
category = {'pdf':'document', 'tex':'manuscript_source', 'py':'code', 'png':'figure', 'json':'structured_record', 'md':'documentation', 'txt':'integrity'}.get(suffix[1:], 'other')
if path.parent.name == 'figures':
category = 'figure'
data = path.read_bytes()
artifacts.append({'artifact_id':rel, 'path':rel, 'category':category,
'media_type':MEDIA_TYPES.get(suffix, 'application/octet-stream'),
'size_bytes':len(data), 'sha256':hashlib.sha256(data).hexdigest()})
pdf_path = research / 'Neiru_Siha_II_Manuscript.pdf'
reader = pypdf.PdfReader(pdf_path)
pages = [{'document_id':'Neiru_Siha_II', 'page_number':i+1, 'page_count':len(reader.pages),
'text':page.extract_text() or '', 'source_path':pdf_path.relative_to(ROOT).as_posix(),
'text_method':'pypdf '+pypdf.__version__+' extract_text default',
'content_role':'literary_dedication' if i == len(reader.pages)-1 else 'manuscript'}
for i,page in enumerate(reader.pages)]
claims = []
for claim in status['proven_results']:
claims.append({'claim_id':claim['id'], 'statement':claim['statement'],
'qualification':claim.get('qualification', ''),
'details_json':json.dumps({k:v for k,v in claim.items() if k not in {'id','statement','qualification'}}, ensure_ascii=False, sort_keys=True),
'status':'stated_and_proved_in_manuscript',
'source_path':'research/RESEARCH_STATUS.json'})
certificates = []
for path in sorted(research.glob('certificate_k*_n*_*.json')):
payload=json.loads(path.read_text(encoding='utf-8'))
certificates.append({'certificate_id':path.stem, 'kind':payload['kind'], 'n':payload['n'],
'support_size':len(payload['p']), 'source_path':path.relative_to(ROOT).as_posix(),
'payload_json':json.dumps(payload, ensure_ascii=False, sort_keys=True),
'arithmetic':'exact rational strings and integer counts'})
counts = {name:write_rows(name,rows) for name,rows in [('artifacts',artifacts),('manuscript_pages',pages),('claims',claims),('certificates',certificates)]}
print(json.dumps({'record_counts':counts, 'pypdf_version':pypdf.__version__}, indent=2))
return counts
if __name__ == '__main__':
build()