Publish licensed data snapshots with LAN LFS transfer and content audits

2026-10-01 03:42:51 +08:00
parent 13cd0e0be7
commit 6e33e05ede
173 changed files with 3614 additions and 0 deletions
@@ -0,0 +1,52 @@
+"""One real archive chunk upload and fresh download; never modifies source data."""
+import argparse
+import json
+from pathlib import Path
+import time
+from datetime import datetime, timezone
+
+import lan_lfs_transfer as lan
+import publish_archives as archives
+
+
+def run(output):
+    root = Path.cwd()
+    receipt = archives.no_links(output)
+    if receipt.exists() or root / 'registry/publication-20261001' not in receipt.parents:
+        raise ValueError('New publication receipt required')
+    catalog = archives.json_read(root / 'artifacts/public-resources-20261001/malecns-synaptic-partners/catalog.json')
+    chunk = catalog['chunks'][0]
+    oid, size = chunk['sha256'], chunk['bytes']
+    source = root / '.git/lfs/objects' / oid[:2] / oid[2:4] / oid
+    records = []
+    for operation in ('upload', 'download'):
+        agent = lan.Agent(dict(event='init', operation=operation, remote='origin',
+                               concurrent=True, concurrenttransfers=2), lambda value: None)
+        original = agent.curl
+        methods = []
+        def observe(method, url, headers, **kwargs):
+            result = original(method, url, headers, **kwargs)
+            methods.append(method)
+            return result
+        agent.curl = observe
+        start = time.perf_counter()
+        result = agent.transfer(dict(event=operation, action=None, oid=oid, size=size,
+                                     **({'path': str(source)} if operation == 'upload' else {})))
+        record = dict(operation=operation, seconds=time.perf_counter()-start, methods=methods,
+                      sha256=oid, bytes=size, complete=result['event'] == 'complete')
+        if operation == 'download':
+            record['new_download_file'] = result['path']
+        records.append(record)
+    report = dict(schema='lfs-lan-roundtrip-smoke-v1', pass_all=True,
+                  observed_utc=datetime.now(timezone.utc).isoformat(),
+                  ip=lan.IP, hostname=lan.HOST, port=443, certificate_verification=True,
+                  proxy_used=False, transfers=records,
+                  adapter=archives.sha_file(lan.__file__))
+    archives.write_new(receipt, archives.canonical(report))
+    print(json.dumps(report), flush=True)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument('--receipt', required=True)
+    run(parser.parse_args().receipt)