Spaces:
Running
Running
| import os | |
| import shutil | |
| import zipfile | |
| import threading | |
| import time | |
| import humanize | |
| from flask import Flask, request, jsonify, render_template_string | |
| import gdown | |
| from huggingface_hub import HfApi, login, upload_folder, hf_hub_url | |
| from huggingface_hub.utils import HfHubHTTPError | |
| # --- Configuration & Initialization --- | |
| # Ensure Hugging Face cache and other temp data writes to /tmp | |
| os.environ["HF_HOME"] = "/tmp/hf_home" | |
| DOWNLOAD_DIR = "/tmp/backups" | |
| EXTRACT_DIR = "/tmp/extracted_backups" | |
| # Environment variables (set these in your Space secrets) | |
| FOLDER_URL = os.getenv("FOLDER_URL") | |
| REPO_ID = os.getenv("REPO_ID") | |
| TOKEN = os.getenv("HF_TOKEN") | |
| # --- Global State Management --- | |
| # Using a dictionary to hold state is thread-safe in CPython for simple reads/writes | |
| app_state = { | |
| "backup_status": "idle", # idle, running, success, error | |
| "backup_log": ["Awaiting first run."], | |
| "last_backup_time": "Never", | |
| "next_backup_time": "Scheduler disabled", | |
| "schedule_interval_minutes": 0, # 0 means disabled | |
| "scheduler_thread": None | |
| } | |
| # --- Flask App Setup --- | |
| app = Flask(__name__) | |
| api = HfApi() | |
| # --- HTML, CSS, JS Template --- | |
| HTML_TEMPLATE = """ | |
| <!DOCTYPE html> | |
| <html lang="en" data-bs-theme="dark"> | |
| <head> | |
| <meta charset="utf-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1"> | |
| <title>Backup & Dataset Controller</title> | |
| <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet"> | |
| <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/all.min.css"> | |
| <style> | |
| body { | |
| background-color: #212529; /* Dark background */ | |
| } | |
| .log-box { | |
| height: 300px; | |
| overflow-y: auto; | |
| font-family: 'Courier New', Courier, monospace; | |
| font-size: 0.85rem; | |
| color: #f8f9fa; | |
| background-color: #111315 !important; | |
| border-top: 1px solid #495057; | |
| } | |
| .log-box div { | |
| padding: 2px 5px; | |
| border-bottom: 1px solid #343a40; | |
| } | |
| .status-badge { | |
| padding: 0.35em 0.65em; | |
| font-size: .75em; | |
| font-weight: 700; | |
| line-height: 1; | |
| color: #fff; | |
| text-align: center; | |
| white-space: nowrap; | |
| vertical-align: baseline; | |
| border-radius: 0.25rem; | |
| transition: background-color 0.3s ease-in-out; | |
| } | |
| .status-idle { background-color: #6c757d; } /* Gray */ | |
| .status-running { background-color: #0d6efd; } /* Blue */ | |
| .status-success { background-color: #198754; } /* Green */ | |
| .status-error { background-color: #dc3545; } /* Red */ | |
| #files-list-container { | |
| max-height: 450px; | |
| overflow-y: auto; | |
| } | |
| .btn i, .btn .spinner-border { | |
| pointer-events: none; /* Make clicks on icons pass through to the button */ | |
| } | |
| .card { | |
| border: 1px solid rgba(255, 255, 255, 0.1); | |
| } | |
| </style> | |
| </head> | |
| <body> | |
| <div class="container my-4"> | |
| <header class="d-flex align-items-center pb-3 mb-4 border-bottom border-secondary"> | |
| <i class="fas fa-server fa-2x me-3 text-info"></i> | |
| <span class="fs-4">Minecraft Backup & Dataset Controller</span> | |
| </header> | |
| <div class="row g-4"> | |
| <!-- Left Panel: Backup Controls --> | |
| <div class="col-lg-6"> | |
| <div class="card h-100 shadow-sm"> | |
| <div class="card-header d-flex justify-content-between align-items-center"> | |
| <h5 class="mb-0"><i class="fas fa-shield-alt me-2"></i>Backup Controls</h5> | |
| <div id="backup-status-indicator" class="status-badge" data-bs-toggle="tooltip" title="Current Status">Idle</div> | |
| </div> | |
| <div class="card-body"> | |
| <div class="d-grid gap-2 mb-4"> | |
| <button id="run-now-btn" class="btn btn-lg btn-success"> | |
| <i class="fas fa-play-circle me-2"></i>Run Backup Now | |
| </button> | |
| </div> | |
| <form id="schedule-form" class="row g-2 align-items-center"> | |
| <div class="col"> | |
| <label for="interval-input" class="form-label">Schedule Interval (minutes)</label> | |
| <input type="number" class="form-control" id="interval-input" placeholder="0 to disable" min="0"> | |
| </div> | |
| <div class="col-auto align-self-end"> | |
| <button type="submit" class="btn btn-primary"> | |
| <i class="fas fa-save me-2"></i>Set | |
| </button> | |
| </div> | |
| </form> | |
| <ul class="list-group list-group-flush mt-4"> | |
| <li class="list-group-item d-flex justify-content-between bg-transparent"> | |
| <span>Last Backup:</span> | |
| <strong id="last-run-time">Never</strong> | |
| </li> | |
| <li class="list-group-item d-flex justify-content-between bg-transparent"> | |
| <span>Next Scheduled:</span> | |
| <strong id="next-run-time">N/A</strong> | |
| </li> | |
| </ul> | |
| </div> | |
| <div class="card-footer"> | |
| <strong><i class="fas fa-clipboard-list me-2"></i>Live Log</strong> | |
| </div> | |
| <div id="log-output" class="log-box card-body"> | |
| <!-- Logs will be populated here by JavaScript --> | |
| </div> | |
| </div> | |
| </div> | |
| <!-- Right Panel: Dataset Management --> | |
| <div class="col-lg-6"> | |
| <div class="card h-100 shadow-sm"> | |
| <div class="card-header d-flex justify-content-between align-items-center"> | |
| <h5 class="mb-0"><i class="fas fa-database me-2"></i>Dataset Management</h5> | |
| <a href="https://huggingface.co/datasets/{{ repo_id }}" target="_blank" class="btn btn-sm btn-outline-info"> | |
| View on Hub <i class="fas fa-external-link-alt"></i> | |
| </a> | |
| </div> | |
| <div class="card-body"> | |
| <div class="d-flex justify-content-between align-items-center mb-3"> | |
| <p class="text-muted mb-0">Files in <strong>{{ repo_id }}</strong></p> | |
| <button id="refresh-files-btn" class="btn btn-sm btn-secondary"> | |
| <i class="fas fa-sync-alt me-1"></i> Refresh | |
| </button> | |
| </div> | |
| <div id="files-list-container"> | |
| <div id="files-loader" class="text-center p-4" style="display: none;"> | |
| <div class="spinner-border text-primary" role="status"> | |
| <span class="visually-hidden">Loading...</span> | |
| </div> | |
| </div> | |
| <table class="table table-hover"> | |
| <thead> | |
| <tr> | |
| <th>File Path</th> | |
| <th>Size</th> | |
| <th>Actions</th> | |
| </tr> | |
| </thead> | |
| <tbody id="files-list"> | |
| <!-- File list will be populated here --> | |
| </tbody> | |
| </table> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"></script> | |
| <script> | |
| document.addEventListener('DOMContentLoaded', () => { | |
| // --- DOM Elements --- | |
| const runNowBtn = document.getElementById('run-now-btn'); | |
| const scheduleForm = document.getElementById('schedule-form'); | |
| const intervalInput = document.getElementById('interval-input'); | |
| const logOutput = document.getElementById('log-output'); | |
| const statusIndicator = document.getElementById('backup-status-indicator'); | |
| const lastRunTimeEl = document.getElementById('last-run-time'); | |
| const nextRunTimeEl = document.getElementById('next-run-time'); | |
| const refreshFilesBtn = document.getElementById('refresh-files-btn'); | |
| const filesListBody = document.getElementById('files-list'); | |
| const filesLoader = document.getElementById('files-loader'); | |
| let statusInterval; | |
| // --- API Helper --- | |
| async function apiCall(endpoint, options = {}) { | |
| try { | |
| const response = await fetch(endpoint, options); | |
| if (!response.ok) { | |
| const errorData = await response.json(); | |
| throw new Error(errorData.message || `HTTP error! status: ${response.status}`); | |
| } | |
| return response.json(); | |
| } catch (error) { | |
| console.error(`API call to ${endpoint} failed:`, error); | |
| alert(`Error: ${error.message}`); | |
| return null; | |
| } | |
| } | |
| // --- UI Update Functions --- | |
| function updateLog(logs) { | |
| logOutput.innerHTML = logs.map(line => `<div>${line}</div>`).join(''); | |
| logOutput.scrollTop = logOutput.scrollHeight; | |
| } | |
| function updateStatusUI(data) { | |
| // Status Badge | |
| statusIndicator.textContent = data.backup_status.charAt(0).toUpperCase() + data.backup_status.slice(1); | |
| statusIndicator.className = 'status-badge'; // Reset classes | |
| statusIndicator.classList.add(`status-${data.backup_status}`); | |
| // Timestamps | |
| lastRunTimeEl.textContent = data.last_backup_time; | |
| nextRunTimeEl.textContent = data.next_backup_time; | |
| // Interval Input | |
| if (document.activeElement !== intervalInput) { | |
| intervalInput.value = data.schedule_interval_minutes > 0 ? data.schedule_interval_minutes : ''; | |
| } | |
| // Button state | |
| runNowBtn.disabled = data.backup_status === 'running'; | |
| if (data.backup_status === 'running') { | |
| runNowBtn.innerHTML = `<span class="spinner-border spinner-border-sm" role="status" aria-hidden="true"></span> Running...`; | |
| } else { | |
| runNowBtn.innerHTML = `<i class="fas fa-play-circle me-2"></i>Run Backup Now`; | |
| } | |
| } | |
| // --- Core Functions --- | |
| async function fetchStatus() { | |
| const data = await apiCall('/api/status'); | |
| if (data) { | |
| updateLog(data.backup_log); | |
| updateStatusUI(data); | |
| } | |
| } | |
| async function runBackup() { | |
| if (runNowBtn.disabled) return; | |
| const data = await apiCall('/api/run-backup', { method: 'POST' }); | |
| if (data) { | |
| fetchStatus(); // Immediately update status | |
| } | |
| } | |
| async function setSchedule(event) { | |
| event.preventDefault(); | |
| const interval = intervalInput.value; | |
| await apiCall('/api/set-schedule', { | |
| method: 'POST', | |
| headers: { 'Content-Type': 'application/json' }, | |
| body: JSON.stringify({ interval: parseInt(interval, 10) || 0 }), | |
| }); | |
| fetchStatus(); | |
| } | |
| async function listFiles() { | |
| filesLoader.style.display = 'block'; | |
| filesListBody.innerHTML = ''; | |
| refreshFilesBtn.disabled = true; | |
| const data = await apiCall('/api/list-files'); | |
| filesLoader.style.display = 'none'; | |
| refreshFilesBtn.disabled = false; | |
| if (data && data.files) { | |
| if (data.files.length === 0) { | |
| filesListBody.innerHTML = '<tr><td colspan="3" class="text-center text-muted">No files found in repository.</td></tr>'; | |
| return; | |
| } | |
| data.files.forEach(file => { | |
| const row = document.createElement('tr'); | |
| row.innerHTML = ` | |
| <td class="text-break"> | |
| <a href="${file.url}" target="_blank" title="${file.name}"> | |
| ${file.name} | |
| </a> | |
| </td> | |
| <td>${file.size}</td> | |
| <td> | |
| <button class="btn btn-sm btn-outline-danger delete-btn" data-filename="${file.name}" title="Delete File"> | |
| <i class="fas fa-trash-alt"></i> | |
| </button> | |
| </td> | |
| `; | |
| filesListBody.appendChild(row); | |
| }); | |
| } | |
| } | |
| async function deleteFile(event) { | |
| const button = event.target.closest('.delete-btn'); | |
| if (!button) return; | |
| const filename = button.dataset.filename; | |
| if (!confirm(`Are you sure you want to permanently delete "${filename}"?`)) { | |
| return; | |
| } | |
| button.disabled = true; | |
| button.innerHTML = `<span class="spinner-border spinner-border-sm"></span>`; | |
| const data = await apiCall('/api/delete-file', { | |
| method: 'POST', | |
| headers: { 'Content-Type': 'application/json' }, | |
| body: JSON.stringify({ filename }), | |
| }); | |
| if (data) { | |
| listFiles(); | |
| } else { | |
| button.disabled = false; | |
| button.innerHTML = `<i class="fas fa-trash-alt"></i>`; | |
| } | |
| } | |
| // --- Event Listeners & Initializers --- | |
| runNowBtn.addEventListener('click', runBackup); | |
| scheduleForm.addEventListener('submit', setSchedule); | |
| refreshFilesBtn.addEventListener('click', listFiles); | |
| filesListBody.addEventListener('click', deleteFile); | |
| // Initialize tooltips | |
| const tooltipTriggerList = [].slice.call(document.querySelectorAll('[data-bs-toggle="tooltip"]')); | |
| tooltipTriggerList.map(function (tooltipTriggerEl) { | |
| return new bootstrap.Tooltip(tooltipTriggerEl); | |
| }); | |
| // Initial data load and periodic polling | |
| fetchStatus(); | |
| listFiles(); | |
| statusInterval = setInterval(fetchStatus, 3000); // Poll for status every 3 seconds | |
| }); | |
| </script> | |
| </body> | |
| </html> | |
| """ | |
| # --- Core Backup Logic --- | |
| def run_backup_job(): | |
| """The main backup logic, designed to be run in a background thread.""" | |
| global app_state | |
| app_state["backup_status"] = "running" | |
| app_state["backup_log"] = ["Starting backup process..."] | |
| def log(message): | |
| print(message) | |
| app_state["backup_log"].append(message) | |
| try: | |
| # 1. Clean up old directories | |
| log("Resetting temporary directories...") | |
| shutil.rmtree(DOWNLOAD_DIR, ignore_errors=True) | |
| shutil.rmtree(EXTRACT_DIR, ignore_errors=True) | |
| os.makedirs(DOWNLOAD_DIR, exist_ok=True) | |
| os.makedirs(EXTRACT_DIR, exist_ok=True) | |
| log("Directories reset.") | |
| # 2. Download from Google Drive | |
| log(f"Downloading from Google Drive folder...") | |
| gdown.download_folder(url=FOLDER_URL, output=DOWNLOAD_DIR, use_cookies=False, quiet=True) | |
| log("Download finished.") | |
| # 3. Extract downloaded zip files | |
| log("Extracting zip archives...") | |
| extracted_count = 0 | |
| for root, _, files in os.walk(DOWNLOAD_DIR): | |
| for f in files: | |
| if f.endswith(".zip"): | |
| zp = os.path.join(root, f) | |
| with zipfile.ZipFile(zp) as z: | |
| z.extractall(EXTRACT_DIR) | |
| log(f"Extracted: {f}") | |
| extracted_count += 1 | |
| if extracted_count == 0: | |
| log("Warning: No .zip files found to extract.") | |
| # 4. Fix potential folder name typo | |
| bad_path = os.path.join(EXTRACT_DIR, "world_nither") | |
| good_path = os.path.join(EXTRACT_DIR, "world_nether") | |
| if os.path.exists(bad_path) and not os.path.exists(good_path): | |
| os.rename(bad_path, good_path) | |
| log("Fixed folder name typo: 'world_nither' -> 'world_nether'") | |
| # 5. Log in to Hugging Face | |
| log("Logging into Hugging Face Hub...") | |
| login(token=TOKEN) | |
| log("Login successful.") | |
| # 6. Ensure repository exists | |
| log(f"Ensuring dataset repository '{REPO_ID}' exists...") | |
| api.create_repo(repo_id=REPO_ID, repo_type="dataset", private=False, exist_ok=True) | |
| log("Repository is ready.") | |
| # 7. Upload specified subfolders | |
| subfolders_to_upload = { | |
| "world": os.path.join(EXTRACT_DIR, "world"), | |
| "world_nether": os.path.join(EXTRACT_DIR, "world_nether"), | |
| "world_the_end": os.path.join(EXTRACT_DIR, "world_the_end"), | |
| "plugins": os.path.join(EXTRACT_DIR, "plugins") | |
| } | |
| for name, path in subfolders_to_upload.items(): | |
| if os.path.exists(path): | |
| log(f"Uploading '{name}'...") | |
| upload_folder( | |
| repo_id=REPO_ID, | |
| folder_path=path, | |
| repo_type="dataset", | |
| path_in_repo=name, | |
| commit_message=f"Backup update for {name}" | |
| ) | |
| log(f"'{name}' uploaded successfully.") | |
| else: | |
| log(f"Skipping '{name}' - directory not found.") | |
| app_state["last_backup_time"] = time.strftime("%Y-%m-%d %H:%M:%S %Z") | |
| log(f"Backup completed successfully at {app_state['last_backup_time']}.") | |
| app_state["backup_status"] = "success" | |
| except Exception as e: | |
| log(f"AN ERROR OCCURRED: {str(e)}") | |
| app_state["backup_status"] = "error" | |
| # --- Scheduler Thread --- | |
| def scheduler_loop(): | |
| """Periodically triggers the backup job based on the set interval.""" | |
| global app_state | |
| while True: | |
| interval = app_state["schedule_interval_minutes"] | |
| if interval > 0: | |
| if app_state["backup_status"] != "running": | |
| print(f"Scheduler triggering backup. Interval: {interval} mins.") | |
| run_backup_job() | |
| next_run_timestamp = time.time() + interval * 60 | |
| app_state["next_backup_time"] = time.strftime("%Y-%m-%d %H:%M:%S %Z", time.localtime(next_run_timestamp)) | |
| time.sleep(interval * 60) | |
| else: | |
| app_state["next_backup_time"] = "Scheduler disabled" | |
| time.sleep(5) | |
| # --- Flask Routes (API Endpoints) --- | |
| def index(): | |
| """Serves the main HTML page by rendering the template string.""" | |
| return render_template_string(HTML_TEMPLATE, repo_id=REPO_ID) | |
| def get_status(): | |
| """Provides the current status of the application to the frontend.""" | |
| return jsonify(dict(app_state)) | |
| def start_backup(): | |
| """Triggers a manual backup run in a background thread.""" | |
| if app_state["backup_status"] == "running": | |
| return jsonify({"status": "error", "message": "A backup is already in progress."}), 409 | |
| threading.Thread(target=run_backup_job, daemon=True).start() | |
| return jsonify({"status": "ok", "message": "Backup process started."}) | |
| def set_schedule(): | |
| """Sets the backup interval.""" | |
| try: | |
| interval = int(request.json.get("interval", 0)) | |
| if interval < 0: | |
| raise ValueError("Interval must be non-negative.") | |
| app_state["schedule_interval_minutes"] = interval | |
| if interval > 0: | |
| next_run_timestamp = time.time() + interval * 60 | |
| app_state["next_backup_time"] = time.strftime("%Y-%m-%d %H:%M:%S %Z", time.localtime(next_run_timestamp)) | |
| else: | |
| app_state["next_backup_time"] = "Scheduler disabled" | |
| return jsonify({"status": "ok", "message": f"Schedule set to {interval} minutes."}) | |
| except (ValueError, TypeError): | |
| return jsonify({"status": "error", "message": "Invalid interval value."}), 400 | |
| def list_repo_files(): | |
| """Lists all files in the dataset repository.""" | |
| try: | |
| repo_files = api.list_repo_files(repo_id=REPO_ID, repo_type="dataset") | |
| files_details = [] | |
| for filename in repo_files: | |
| try: | |
| info = api.get_repo_file_info(repo_id=REPO_ID, path_in_repo=filename, repo_type="dataset") | |
| size = humanize.naturalsize(info.size) if info.size else "0 B" | |
| except HfHubHTTPError: | |
| size = "N/A" | |
| files_details.append({ | |
| "name": filename, | |
| "size": size, | |
| "url": hf_hub_url(repo_id=REPO_ID, filename=filename, repo_type="dataset") | |
| }) | |
| return jsonify({"status": "ok", "files": files_details}) | |
| except Exception as e: | |
| return jsonify({"status": "error", "message": str(e)}), 500 | |
| def delete_repo_file(): | |
| """Deletes a specific file from the dataset repository.""" | |
| filename = request.json.get("filename") | |
| if not filename: | |
| return jsonify({"status": "error", "message": "Filename not provided."}), 400 | |
| try: | |
| api.delete_file( | |
| repo_id=REPO_ID, | |
| path_in_repo=filename, | |
| repo_type="dataset", | |
| commit_message=f"Deleted file: {filename}" | |
| ) | |
| return jsonify({"status": "ok", "message": f"Successfully deleted '{filename}'."}) | |
| except HfHubHTTPError as e: | |
| return jsonify({"status": "error", "message": f"File not found or permission error: {e}"}), 404 | |
| except Exception as e: | |
| return jsonify({"status": "error", "message": str(e)}), 500 | |
| # --- Main Execution --- | |
| if __name__ == "__main__": | |
| # Start the scheduler in a background thread | |
| app_state["scheduler_thread"] = threading.Thread(target=scheduler_loop, daemon=True) | |
| app_state["scheduler_thread"].start() | |
| # Start the Flask web server | |
| app.run(host="0.0.0.0", port=7860) |