Initial SQL-only 1C adapter baseline

2026-07-22 03:03:47 +03:00
commit e2503b77e7
545 changed files with 184711 additions and 0 deletions
@@ -0,0 +1,88 @@
+from __future__ import annotations
+
+import argparse
+import os
+import sys
+from pathlib import Path
+
+try:
+    from huggingface_hub import snapshot_download
+except ImportError as exc:
+    raise SystemExit(
+        "Missing dependency: huggingface_hub. Install dependencies with `pip install -r requirements.txt`."
+    ) from exc
+
+from common import load_model_card
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description="Download a Hugging Face model from a model card.")
+    parser.add_argument("card_id", help="Model card id, for example qwen3-4b-instruct-2507")
+    parser.add_argument(
+        "--local-dir",
+        help="Override destination directory. Defaults to the card storage_path.",
+    )
+    parser.add_argument(
+        "--dry-run",
+        action="store_true",
+        help="Print what would be downloaded without downloading files.",
+    )
+    parser.add_argument(
+        "--allow-pattern",
+        action="append",
+        dest="allow_patterns",
+        help="Optional Hugging Face allow pattern. Can be passed multiple times.",
+    )
+    parser.add_argument(
+        "--ignore-pattern",
+        action="append",
+        dest="ignore_patterns",
+        default=["*.msgpack", "*.h5", "*.ot"],
+        help="Optional Hugging Face ignore pattern. Can be passed multiple times.",
+    )
+    args = parser.parse_args()
+
+    card = load_model_card(args.card_id)
+    repo_id = card.get("upstream_id")
+    if not repo_id:
+        print(f"Model card `{args.card_id}` has no upstream_id.", file=sys.stderr)
+        return 1
+
+    destination_raw = args.local_dir or card.get("storage_path") or ""
+    if not str(destination_raw):
+        print(f"Model card `{args.card_id}` has no storage_path.", file=sys.stderr)
+        return 1
+
+    token = os.environ.get("HF_TOKEN") or None
+
+    print(f"Model card: {args.card_id}")
+    print(f"Repository: {repo_id}")
+    print(f"Destination: {destination_raw}")
+    print(f"Token: {'set' if token else 'not set'}")
+
+    if args.dry_run:
+        return 0
+
+    if os.name == "nt" and not args.local_dir and str(destination_raw).startswith("/"):
+        print(
+            "The model card storage_path is a Linux path. On Windows, pass --local-dir "
+            "or run this script on the GPU/Linux host where /models exists.",
+            file=sys.stderr,
+        )
+        return 1
+
+    destination = Path(destination_raw)
+    destination.mkdir(parents=True, exist_ok=True)
+    downloaded_path = snapshot_download(
+        repo_id=repo_id,
+        local_dir=destination,
+        token=token,
+        allow_patterns=args.allow_patterns,
+        ignore_patterns=args.ignore_patterns,
+    )
+    print(f"Downloaded to: {downloaded_path}")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())