"""Stage a checked export with the code and docs from this release. No upload.""" from __future__ import annotations import argparse import copy import hashlib import json import os from pathlib import Path import re import shutil import subprocess import sys import time RUNTIME = {'.cache', '__pycache__', '.git', '.venv', 'work'} def read_json(path): return json.loads(path.read_text(encoding='utf-8')) def write_json(path, value): path.parent.mkdir(parents=True, exist_ok=True) path.write_text(json.dumps(value, indent=2), encoding='utf-8') def inventory(folder): folder = folder.resolve(strict=True) paths = sorted( p for p in folder.rglob('*') if p.is_file() and not (set(p.relative_to(folder).parts) & RUNTIME) and p.name not in {'release_inventory.json', 'upload_state.json'} and p.suffix != '.pyc' ) entries = [] started = last = time.monotonic() for index, path in enumerate(paths, 1): with path.open('rb') as handle: digest = hashlib.file_digest(handle, 'sha256').hexdigest() entries.append({'path': path.relative_to(folder).as_posix(), 'bytes': path.stat().st_size, 'sha256': digest}) if time.monotonic() - last >= 20: print(json.dumps({'event': 'hash_progress', 'files': index, 'total': len(paths)}), flush=True) last = time.monotonic() result = { 'status': 'prepared', 'algorithm': 'SHA256', 'files': entries, 'file_count': len(entries), 'total_bytes': sum(i['bytes'] for i in entries), 'total_GiB': sum(i['bytes'] for i in entries) / 2**30, 'note': 'This inventory excludes itself and local runtime files.', } write_json(folder / 'release_inventory.json', result) print(json.dumps({'event': 'inventory_complete', 'files': len(entries), 'GiB': result['total_GiB'], 'seconds': round(time.monotonic()-started, 1)}), flush=True) def prepare(args): source = args.compressed_dir.resolve(strict=True) assets = args.assets_dir.resolve(strict=True) dest = args.output_dir.resolve() for input_path in (source, assets): if dest == input_path or dest.is_relative_to(input_path) or input_path.is_relative_to(dest): raise ValueError('Use a separate output directory outside the input folders') if dest.exists(): raise ValueError('Output directory already exists. Choose a new directory') if not re.fullmatch(r'[A-Za-z0-9_.-]+/[A-Za-z0-9_.-]+', args.repo_id): raise ValueError('Use username/dataset-name') manifest = read_json(source / 'manifest.json') report = read_json(source / 'validation.json') if manifest['status'] != 'complete' or report['status'] != 'passed': raise ValueError('Use a completed export with a passed validation report') dest.mkdir(parents=True) for path in assets.rglob('*'): relative = path.relative_to(assets) if not path.is_file() or set(relative.parts) & RUNTIME or path.suffix in {'.parquet', '.pyc'}: continue if path.name in {'manifest.json', 'validation.json', 'upload_plan.json', 'release_inventory.json', 'release_checks.json', 'upload_state.json'}: continue target = dest / relative target.parent.mkdir(parents=True, exist_ok=True) shutil.copy2(path, target) public_manifest = copy.deepcopy(manifest) public_manifest['source_folder'] = 'source SQLite databases (local path omitted)' write_json(dest / 'manifest.json', public_manifest) shutil.copy2(source / 'validation.json', dest / 'validation.json') count = 0 counts = [] for filename, database in manifest['databases'].items(): for table, info in database['tables'].items(): counts.append({'File': filename, 'Table': table, 'Rows': info['rows'], 'Columns': len(info['schema']) - 1, 'SizeBytes': database['source_identity']['bytes']}) for part in info['parts']: relative = Path(Path(filename).stem) / table / part['part'] src = (source / relative).resolve(strict=True) target = (dest / relative).resolve() if not src.is_relative_to(source) or not target.is_relative_to(dest): raise ValueError('Invalid manifest part path') if src.stat().st_size != part['bytes']: raise ValueError('Part size differs: ' + str(relative)) target.parent.mkdir(parents=True, exist_ok=True) if args.copy_data: shutil.copy2(src, target) else: os.link(src, target) # Requires both folders on one filesystem. count += 1 if count != report['files']: raise ValueError('Part count differs from validation report') write_json(dest / 'metadata/source_counts.json', counts) plan = { 'repo_id': args.repo_id, 'repo_type': 'dataset', 'public': True, 'status': 'prepared_not_uploaded', 'parquet_files': count, 'parquet_bytes': manifest['total_parquet_bytes'], 'parquet_GiB': manifest['total_parquet_bytes'] / 2**30, 'tables': report['tables'], 'rows': manifest['total_rows'], 'original_databases_modified': False, 'code_included': True, 'local_data_layout': 'copies' if args.copy_data else 'hard links to validated Parquet', 'source_license': 'No explicit redistribution license supplied; no new license assigned', } write_json(dest / 'upload_plan.json', plan) subprocess.run([sys.executable, str(dest / 'code/prepare_hf_card.py'), '--repo-id', args.repo_id], check=True) write_json(dest / 'release_checks.json', {'status': 'not_run_for_this_new_package', 'note': 'Run the documented local validation before publishing this newly staged package.'}) inventory(dest) print(json.dumps(plan, indent=2)) if __name__ == '__main__': parser = argparse.ArgumentParser(description=__doc__) parser.add_argument('--compressed-dir', type=Path) parser.add_argument('--assets-dir', type=Path, default=Path(__file__).resolve().parents[1]) parser.add_argument('--output-dir', type=Path) parser.add_argument('--repo-id', default='alpha31476/spotify-metadata-parquet') parser.add_argument('--copy-data', action='store_true', help='Copy data instead of same-filesystem hard links') parser.add_argument('--inventory-only', type=Path) args = parser.parse_args() if args.inventory_only: inventory(args.inventory_only) elif not args.compressed_dir or not args.output_dir: parser.error('--compressed-dir and --output-dir are required for preparation') else: prepare(args)