"""Package only the explicitly selected saved jobs and referenced uploads."""
import argparse
import hashlib
import json
from pathlib import Path
import re
import zipfile


def strings(value):
    if isinstance(value, str): yield value
    elif isinstance(value, dict):
        for child in value.values(): yield from strings(child)
    elif isinstance(value, list):
        for child in value: yield from strings(child)


def build(data, selection, output):
    if output.exists(): raise ValueError('refusing to overwrite bundle')
    files = {}; missing = set()
    models = json.loads(selection.read_text())['models']
    for model in models:
        identity = model['job_id']
        if not re.fullmatch('[0-9a-f]{24}', identity): raise ValueError('invalid job ID')
        name = f'8097-jobs/{identity}.json'; path = data / name
        files[name] = path
        job = json.loads(path.read_text())
        if not (job.get('result') or {}).get('concrete_components'):
            raise ValueError('saved concrete missing')
        references = {match for s in strings(job)
                      for match in re.findall(r'drawing-[0-9a-f]+\.(?:dxf|dwg|pdf|png|jpg|jpeg)', s, re.I)}
        for reference in references:
            source = data / 'uploads' / reference
            if source.is_file(): files['uploads/' + reference] = source
            else: missing.add(reference)
    manifest = {'schema': 'dms-shared-cache-v1', 'models': models,
                'missing_referenced_uploads': sorted(missing), 'files': []}
    output.parent.mkdir(parents=True, exist_ok=True)
    with zipfile.ZipFile(output, 'x', compression=zipfile.ZIP_DEFLATED, compresslevel=6) as bundle:
        for name, path in sorted(files.items()):
            payload = path.read_bytes()
            manifest['files'].append({'path': name, 'bytes': len(payload),
                                      'sha256': hashlib.sha256(payload).hexdigest()})
            bundle.writestr(name, payload)
        bundle.writestr('manifest.json', json.dumps(manifest, ensure_ascii=False, indent=2))
    return {'models': len(models), 'files': len(files), 'archive_bytes': output.stat().st_size,
            'missing_referenced_uploads': sorted(missing)}


if __name__ == '__main__':
    p=argparse.ArgumentParser(description=__doc__)
    p.add_argument('--data',type=Path,required=True);p.add_argument('--selection',type=Path,required=True)
    p.add_argument('--output',type=Path,required=True);a=p.parse_args()
    print(json.dumps(build(a.data,a.selection,a.output),ensure_ascii=False))