feat: map local dataset names to hub names in download commands

Author: Ryu-Yang

docs/js/app.js

@@ -13,6 +13,7 @@ import SelectionPanelManager from './modules/selection-panel.js';
 import UIUtils from './modules/ui-utils.js';
 import EventHandlers from './modules/event-handlers.js';
 import RobotAliasManager from './modules/robot-aliases.js';
+import HubDatasetNamesManager from './modules/hub-dataset-names.js';
 import ErrorNotifier from './modules/error-notifier.js';
 import DownloadManager from './modules/download-manager.js';
 import NewsMenu from './modules/news-menu.js';
@@ -70,8 +71,9 @@ class Application {
                 console.warn('[APP] Failed to load download stats:', err);
             });
 
-            // Load robot alias map (non-blocking for core data, but awaited before UI init)
+            // Load robot alias map and hub dataset name mappings before UI init
             await RobotAliasManager.load(this.config);
+            await HubDatasetNamesManager.load(this.config);
 
             // Load datasets
             const loadingProgress = document.getElementById('loadingProgress');

docs/js/config/config.json

@@ -2,7 +2,7 @@
   "assets": {
     "useLocalAssets": false,
     "localAssetsPath": "../pageAssets",
-    "defaultRemoteAssetsRoot": "https://huggingface.co/datasets/RoboCOIN/pageAssets/resolve/main"
+    "defaultRemoteAssetsRoot": "https://huggingface.co/datasets/RoboCOIN/pageAssets/tree/main"
   },
   "downloadCommand": {
     "command": "robocoin-download",

docs/js/modules/config.js

@@ -3,6 +3,8 @@
  * @description Centralized configuration management for the RoboCOIN application
  */
 
+import HubDatasetNamesManager from './hub-dataset-names.js';
+
 /**
  * @typedef {Object} GridConfig
  * @property {number} minCardWidth - Minimum card width in pixels
@@ -190,7 +192,11 @@ class ConfigManager {
             return fallback;
         }
 
-        return normalized;
+        // HuggingFace browser URLs use /tree/main; raw file fetches need /resolve/main.
+        if (/huggingface\.co\/datasets\/[^/]+\/[^/]+$/i.test(normalized)) {
+            return `${normalized}/resolve/main`;
+        }
+        return normalized.replace(/\/tree\/main$/i, '/resolve/main');
     }
 
     /**
@@ -290,7 +296,9 @@ class ConfigManager {
             assets: {
                 useLocalAssets: this.getJsonValue('assets.useLocalAssets', false),
                 localAssetsPath: this.getJsonValue('assets.localAssetsPath', '../robocoin_datamanager_assets'),
-                defaultRemoteAssetsRoot: this.getJsonValue('assets.defaultRemoteAssetsRoot', 'https://huggingface.co/datasets/RogersPyke/robocoin_datamanager_assets/resolve/main')
+                defaultRemoteAssetsRoot: this.getJsonValue('assets.defaultRemoteAssetsRoot', 'https://huggingface.co/datasets/RogersPyke/robocoin_datamanager_assets/resolve/main'),
+                useBundledConsolidated: this.getJsonValue('assets.useBundledConsolidated', false),
+                bundledConsolidatedPath: this.getJsonValue('assets.bundledConsolidatedPath', './assets/info/consolidated_datasets.json')
             },
             layout: {
                 contentPadding: getValue('--content-padding', 'defaults.layout.contentPadding', 12)
@@ -347,7 +355,8 @@ class ConfigManager {
                 assetsRoot,
                 info: infoPath,  // JSON index files following standard structure
                 datasetInfo: datasetInfoPath,
-                videos: videosPath
+                videos: videosPath,
+                bundledConsolidated: this.getJsonValue('assets.bundledConsolidatedPath', './assets/info/consolidated_datasets.json')
             },
             // Download command format configuration
             // Loaded from config.json for easy editing
@@ -509,7 +518,8 @@ class ConfigManager {
      */
     static generateDownloadCommand(hub, datasets, datasetMap = undefined) {
         const config = this.getConfig().downloadCommand;
-        
+        const hubDatasets = HubDatasetNamesManager.resolveHubNames(datasets, hub);
+
         // First line: storage estimate comment (no blank line)
         const storageComment = this.buildRequiredStorageComment(datasets, datasetMap);
 
@@ -521,8 +531,8 @@ class ConfigManager {
 
         // Format: --ds_lists + each dataset on new line with continuation
         command += `${config.datasetsParam} `;
-        if (datasets.length > 0) {
-            const dsListContent = datasets.join(config.datasetSeparator);
+        if (hubDatasets.length > 0) {
+            const dsListContent = hubDatasets.join(config.datasetSeparator);
             command += `${dsListContent}${config.lineBreak}`;
         } else {
             command += `${config.lineBreak}`;

docs/js/modules/hub-dataset-names.js

@@ -0,0 +1,89 @@
+import ConfigManager from './config.js';
+
+/**
+ * @file Hub Dataset Names Module
+ * @description Maps local dataset names to hub-specific remote repository names
+ *              for robocoin-download --ds_lists (HF / ModelScope naming differs from UI).
+ */
+
+class HubDatasetNamesManager {
+    constructor() {
+        /** @type {{ huggingface?: Record<string, string>, modelscope?: Record<string, string> }} */
+        this.hubMaps = { huggingface: {}, modelscope: {} };
+
+        /** @type {boolean} */
+        this.loaded = false;
+
+        /** @type {Promise<Object>|null} */
+        this.loadingPromise = null;
+    }
+
+    /**
+     * @param {import('./config.js').ConfigManager['getConfig'] extends () => infer C ? C : any} config
+     * @returns {Promise<Object>}
+     */
+    async load(config) {
+        if (this.loaded) {
+            return this.hubMaps;
+        }
+        if (this.loadingPromise) {
+            return this.loadingPromise;
+        }
+
+        const infoPath =
+            config?.paths?.info ||
+            `${ConfigManager.getDefaultRemoteAssetsRoot()}/info`;
+        const url = `${infoPath}/hub_dataset_names.json`;
+
+        this.loadingPromise = (async () => {
+            try {
+                const res = await fetch(url);
+                if (!res.ok) {
+                    console.warn(`[HubDatasetNames] Mapping file not found at ${url}. Status: ${res.status}`);
+                    this.hubMaps = { huggingface: {}, modelscope: {} };
+                } else {
+                    const data = await res.json();
+                    this.hubMaps = {
+                        huggingface: data?.huggingface && typeof data.huggingface === 'object' ? data.huggingface : {},
+                        modelscope: data?.modelscope && typeof data.modelscope === 'object' ? data.modelscope : {},
+                    };
+                }
+            } catch (err) {
+                console.warn('[HubDatasetNames] Failed to load hub_dataset_names.json:', err);
+                this.hubMaps = { huggingface: {}, modelscope: {} };
+            } finally {
+                this.loaded = true;
+            }
+            return this.hubMaps;
+        })();
+
+        return this.loadingPromise;
+    }
+
+    /**
+     * Resolve local UI name to the name expected by robocoin-download on the given hub.
+     * @param {string} localName
+     * @param {string} hub - 'huggingface' or 'modelscope'
+     * @returns {string}
+     */
+    resolveHubName(localName, hub) {
+        if (!localName) return localName;
+        const map = this.hubMaps[hub];
+        if (map && map[localName]) {
+            return map[localName];
+        }
+        return localName;
+    }
+
+    /**
+     * @param {string[]} localNames
+     * @param {string} hub
+     * @returns {string[]}
+     */
+    resolveHubNames(localNames, hub) {
+        return localNames.map(name => this.resolveHubName(name, hub));
+    }
+}
+
+const instance = new HubDatasetNamesManager();
+export default instance;