scripts/upload_space_urllib.py

"""
Upload WorldDisasterLM-8B as a public HuggingFace Space using urllib (no httpx/requests).
Works on Python 3.14 on Windows where httpx TLS may fail.

Usage
-----
  python scripts/upload_space_urllib.py --token hf_xxx --username drdeveloper88
"""

from __future__ import annotations

import argparse
import base64
import json
import os
import shutil
import ssl
import sys
import tempfile
import urllib.request
from pathlib import Path

ROOT = Path(__file__).parent.parent
SPACE_DIR = ROOT / "hf_space"
HF_API = "https://huggingface.co/api"

CTX = ssl.create_default_context()


def api(method: str, path: str, token: str, payload: dict | None = None) -> dict:
    url = f"{HF_API}{path}"
    data = json.dumps(payload).encode() if payload else None
    req = urllib.request.Request(
        url,
        data=data,
        method=method,
        headers={
            "Authorization": f"Bearer {token}",
            "Content-Type": "application/json",
        },
    )
    try:
        with urllib.request.urlopen(req, context=CTX, timeout=30) as r:
            return json.loads(r.read())
    except urllib.error.HTTPError as e:
        body = e.read().decode()
        raise RuntimeError(f"HTTP {e.code}: {body}") from e


def upload_file(repo_id: str, token: str, local_path: Path, repo_path: str, commit_msg: str) -> None:
    """Upload a single file via the HF /api/repos endpoint (LFS-aware preupload + commit)."""
    content = local_path.read_bytes()

    # Step 1: preupload (get upload URL or confirm non-LFS)
    preupload_url = f"https://huggingface.co/api/spaces/{repo_id}/preupload/main"
    payload = json.dumps([{"path": repo_path, "size": len(content)}]).encode()
    req = urllib.request.Request(
        preupload_url,
        data=payload,
        method="POST",
        headers={
            "Authorization": f"Bearer {token}",
            "Content-Type": "application/json",
        },
    )
    with urllib.request.urlopen(req, context=CTX, timeout=30) as r:
        preupload = json.loads(r.read())

    files_info = preupload.get("files", [{}])
    upload_mode = files_info[0].get("uploadMode", "regular") if files_info else "regular"

    if upload_mode == "lfs":
        # Upload to LFS URL
        upload_url = files_info[0]["uploadUrl"]
        put_req = urllib.request.Request(
            upload_url,
            data=content,
            method="PUT",
            headers={"Content-Type": "application/octet-stream"},
        )
        with urllib.request.urlopen(put_req, context=CTX, timeout=60) as r:
            r.read()
        oid = files_info[0]["oid"]
        size = files_info[0]["size"]
        lfs_content = (
            f"version https://git-lfs.github.com/spec/v1\n"
            f"oid sha256:{oid}\n"
            f"size {size}\n"
        ).encode()
        final_content = lfs_content
    else:
        final_content = content

    # Step 2: commit
    commit_url = f"https://huggingface.co/api/spaces/{repo_id}/commit/main"
    header = json.dumps({
        "summary": commit_msg,
        "files": [{"path": repo_path, "encoding": "base64"}],
    })
    boundary = "----HFUploadBoundary"
    body_parts = [
        f"--{boundary}\r\nContent-Disposition: form-data; name=\"header\"\r\n\r\n{header}\r\n".encode(),
        f"--{boundary}\r\nContent-Disposition: form-data; name=\"file\"; filename=\"{repo_path}\"\r\n\r\n".encode(),
        base64.b64encode(final_content),
        f"\r\n--{boundary}--\r\n".encode(),
    ]
    body = b"".join(body_parts)
    commit_req = urllib.request.Request(
        commit_url,
        data=body,
        method="POST",
        headers={
            "Authorization": f"Bearer {token}",
            "Content-Type": f"multipart/form-data; boundary={boundary}",
        },
    )
    with urllib.request.urlopen(commit_req, context=CTX, timeout=60) as r:
        r.read()


def upload_folder_simple(repo_id: str, token: str, folder: Path, readme_text: str) -> None:
    """Upload all Space files using the simpler single-commit API."""
    files_payload = []
    file_contents = {}

    for f in folder.iterdir():
        if f.is_file():
            if f.name == "README.md":
                content = readme_text.encode("utf-8")
            else:
                content = f.read_bytes()
            encoded = base64.b64encode(content).decode()
            files_payload.append({"path": f.name, "encoding": "base64"})
            file_contents[f.name] = encoded

    # Build multipart commit
    boundary = "HFSpaceUpload42"
    parts = []
    header = json.dumps({
        "summary": "Upload WorldDisasterLM-8B Space",
        "files": files_payload,
    })
    parts.append(
        f"--{boundary}\r\nContent-Disposition: form-data; name=\"header\"\r\n\r\n{header}\r\n".encode()
    )
    for item in files_payload:
        fname = item["path"]
        parts.append(
            f"--{boundary}\r\nContent-Disposition: form-data; name=\"file\"; filename=\"{fname}\"\r\n\r\n".encode()
        )
        parts.append(file_contents[fname].encode())
        parts.append(b"\r\n")
    parts.append(f"--{boundary}--\r\n".encode())
    body = b"".join(parts)

    commit_url = f"https://huggingface.co/api/spaces/{repo_id}/commit/main"
    req = urllib.request.Request(
        commit_url,
        data=body,
        method="POST",
        headers={
            "Authorization": f"Bearer {token}",
            "Content-Type": f"multipart/form-data; boundary={boundary}",
        },
    )
    try:
        with urllib.request.urlopen(req, context=CTX, timeout=120) as r:
            result = json.loads(r.read())
            return result
    except urllib.error.HTTPError as e:
        raise RuntimeError(f"Commit failed HTTP {e.code}: {e.read().decode()}") from e


def parse_args() -> argparse.Namespace:
    p = argparse.ArgumentParser()
    p.add_argument("--token", default=os.environ.get("HF_TOKEN"), help="HF token (or set HF_TOKEN env var)")
    p.add_argument("--username", required=True)
    p.add_argument("--space-name", default="WorldDisasterLM-8B")
    p.add_argument("--private", action="store_true")
    return p.parse_args()


def main() -> None:
    args = parse_args()
    token = args.token
    if not token:
        print("ERROR: provide --token or set HF_TOKEN"); sys.exit(1)

    repo_id = f"{args.username}/{args.space_name}"

    print(f"\n{'='*60}")
    print(f"  WorldDisasterLM-8B → HuggingFace Space")
    print(f"  Repo: {repo_id}")
    print(f"{'='*60}\n")

    # Verify token
    user = api("GET", "/whoami-v2", token)
    print(f"  Authenticated as: {user.get('name')}")

    # Step 1: Create Space
    print("\nStep 1/3 — Creating Space repository...")
    try:
        api("POST", "/repos/create", token, {
            "type": "space",
            "name": args.space_name,
            "sdk": "gradio",
            "private": args.private,
            "exist_ok": True,
        })
        print(f"  ✓ Space created")
    except RuntimeError as e:
        if "already exists" in str(e) or "409" in str(e):
            print(f"  ✓ Space already exists, updating")
        else:
            raise

    # Step 2: Patch README
    readme_text = (SPACE_DIR / "README.md").read_text(encoding="utf-8")
    readme_text = readme_text.replace("YOUR_HF_USERNAME", args.username)

    # Step 3: Upload files
    print("Step 2/3 — Uploading files...")
    result = upload_folder_simple(repo_id, token, SPACE_DIR, readme_text)
    print(f"  ✓ Files uploaded  (commit: {result.get('commitOid', 'ok')[:12] if result.get('commitOid') else 'done'})")

    print(f"\nStep 3/3 — Done!\n")
    print(f"{'='*60}")
    print(f"  Live at: https://huggingface.co/spaces/{repo_id}")
    print(f"  (Space may take 1-2 min to build)")
    print(f"{'='*60}\n")


if __name__ == "__main__":
    main()