smart-image-scraper-mcp 2.1.0 → 2.1.1

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "smart-image-scraper-mcp",
-  "version": "2.1.0",
+  "version": "2.1.1",
   "description": "全网智能图片抓取 MCP 服务器 - 支持 Bing/Google 图片搜索、验证和下载",
   "main": "src/index.js",
   "type": "module",

package/src/services/linkValidator.js

@@ -74,16 +74,15 @@ export class LinkValidator {
   
   /**
    * 快速验证 - 仅检查 URL 格式，不发送 HTTP 请求
+   * 宽松模式：只要 URL 格式正确就通过
    */
   quickValidate(url) {
     if (!this._isValidUrlFormat(url)) {
       return { url, valid: false, error: 'Invalid URL' };
     }
-    // 检查常见图片扩展名
-    const imageExtensions = ['.jpg', '.jpeg', '.png', '.gif', '.webp', '.bmp', '.svg'];
-    const urlLower = url.toLowerCase();
-    const hasImageExt = imageExtensions.some(ext => urlLower.includes(ext));
-    return { url, valid: hasImageExt, quality: null };
+    // 宽松验证：只要 URL 格式正确就认为有效
+    // 图片 URL 可能不包含扩展名（动态生成的 URL）
+    return { url, valid: true, quality: null };
   }
 
   /**

package/src/services/orchestrator.js

@@ -67,23 +67,24 @@ export class Orchestrator {
    */
   async processKeywordLink(keyword, count, source, options = {}) {
     const startTime = Date.now();
-    const fastMode = options.fastMode !== false; // 默认开启快速模式
+    const fastMode = options.fastMode !== false;
     const prioritizeQuality = options.prioritizeQuality === true;
     
     try {
       const scraper = getScraper(source);
-      const searchCount = fastMode ? count : Math.ceil(count * 1.5);
+      // 多搜索一些以确保有足够的结果
+      const searchCount = Math.max(count * 3, 10);
       
       // 尝试从缓存获取
       const cachedUrls = searchCache.getSearchResult(keyword, source, options);
       let rawUrls;
       
-      if (cachedUrls && cachedUrls.length >= searchCount) {
+      if (cachedUrls && cachedUrls.length >= count) {
         logger.info(`[CACHE] "${keyword}" - ${cachedUrls.length} URLs`);
         rawUrls = cachedUrls;
         metrics.recordCacheHit();
       } else {
-        logger.info(`[SEARCH] "${keyword}"...`);
+        logger.info(`[SEARCH] "${keyword}" (target: ${searchCount})...`);
         rawUrls = await scraper.search(keyword, searchCount, options);
         if (rawUrls.length > 0) {
           searchCache.setSearchResult(keyword, source, options, rawUrls);
@@ -100,20 +101,24 @@ export class Orchestrator {
         };
       }
 
-      // 快速模式：跳过 HTTP 验证，仅检查 URL 格式
+      // 快速模式：直接返回搜索结果（不验证）
       let resultUrls;
       if (fastMode && !prioritizeQuality) {
-        // 快速验证：仅检查 URL 格式和扩展名
-        const quickResults = rawUrls.map(url => this.linkValidator.quickValidate(url));
-        resultUrls = quickResults.filter(r => r.valid).slice(0, count).map(r => r.url);
-        logger.info(`[FAST] "${keyword}" - ${resultUrls.length} URLs (no HTTP validation)`);
+        // 快速模式：直接使用搜索结果
+        resultUrls = rawUrls.slice(0, count);
+        logger.info(`[FAST] "${keyword}" - ${resultUrls.length} URLs`);
       } else {
-        // 完整验证
+        // 完整验证模式：验证不通过的继续搜索更多
         const { valid } = await this.linkValidator.validateMany(rawUrls, {
           fetchQuality: prioritizeQuality,
           sortByQuality: prioritizeQuality,
         });
         resultUrls = valid.slice(0, count).map(v => v.url);
+        
+        // 如果验证通过的不够，记录警告
+        if (resultUrls.length < count) {
+          logger.warn(`[VALIDATE] "${keyword}" - only ${resultUrls.length}/${count} valid`);
+        }
       }
 
       return {