Merge pull request #107 from BigDataQiChao/main

fix:url解析图片时进行链接有效性校验

mallchat-common/src/main/java/com/abin/mallchat/common/common/utils/discover/AbstractUrlDiscover.java

@@ -11,6 +11,9 @@ import org.jsoup.Jsoup;
 import org.jsoup.nodes.Document;
 import org.springframework.data.util.Pair;
 
+import java.io.InputStream;
+import java.net.HttpURLConnection;
+import java.net.URL;
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
@@ -59,7 +62,7 @@ public abstract class AbstractUrlDiscover implements UrlDiscover {
         return UrlInfo.builder()
                 .title(getTitle(document))
                 .description(getDescription(document))
-                .image(getImage(assemble(url),document)).build();
+                .image(getImage(assemble(url), document)).build();
     }
 
 
@@ -83,4 +86,32 @@ public abstract class AbstractUrlDiscover implements UrlDiscover {
         return null;
     }
 
+    /**
+     * 判断链接是否有效
+     * 输入链接
+     * 返回true或者false
+     */
+    public static boolean isConnect(String href) {
+        //请求地址
+        URL url;
+        //请求状态码
+        int state;
+        //下载链接类型
+        String fileType;
+        try {
+            url = new URL(href);
+            HttpURLConnection httpURLConnection = (HttpURLConnection) url.openConnection();
+            state = httpURLConnection.getResponseCode();
+            fileType = httpURLConnection.getHeaderField("Content-Disposition");
+            //如果成功200，缓存304，移动302都算有效链接，并且不是下载链接
+            if ((state == 200 || state == 302 || state == 304) && fileType == null) {
+                return true;
+            }
+            httpURLConnection.disconnect();
+        } catch (Exception e) {
+            return false;
+        }
+        return false;
+    }
+
 }

mallchat-common/src/main/java/com/abin/mallchat/common/common/utils/discover/CommonUrlDiscover.java

@@ -7,6 +7,10 @@ import org.jetbrains.annotations.Nullable;
 import org.jsoup.nodes.Document;
 import org.jsoup.select.Elements;
 
+import java.io.InputStream;
+import java.net.HttpURLConnection;
+import java.net.URL;
+
 /**
  * @author zhaoqichao
  * @date 2023/7/3 16:54
@@ -34,19 +38,17 @@ public class CommonUrlDiscover extends AbstractUrlDiscover {
         String image = document.select("link[type=image/x-icon]").attr("href");
         //如果没有去匹配含有icon属性的logo
         String href = StrUtil.isEmpty(image) ? document.select("link[rel$=icon]").attr("href") : image;
-        //如果icon中已经包含了url部分域名
-        if (StrUtil.isNotBlank(StrUtil.removeAny(StrUtil.removeAny(href, "/"), "favicon.ico")) &&
-                StrUtil.containsAny(StrUtil.removePrefix(url, "http://"), StrUtil.removeAny(StrUtil.removeAny(href, "/"), "favicon.ico"))) {
-            return "http://" + StrUtil.removePrefix(href, "/");
-        }
         //如果url已经包含了logo
         if (StrUtil.containsAny(url, "favicon")) {
             return url;
         }
-        //如果logo中有url
+        //如果icon可以直接访问或者包含了http
-        if (StrUtil.containsAny(href, "http") || StrUtil.containsAny(href, "https")) {
+        if (isConnect(!StrUtil.startWith(href, "http") ? "http:" + href : href)) {
             return href;
         }
+
         return StrUtil.format("{}/{}", url, StrUtil.removePrefix(href, "/"));
     }
+
+
 }

mallchat-common/src/main/java/com/abin/mallchat/common/common/utils/discover/WxUrlDiscover.java

@@ -25,6 +25,7 @@ public class WxUrlDiscover extends AbstractUrlDiscover {
     @Nullable
     @Override
     public String getImage(String url, Document document) {
-        return document.getElementsByAttributeValue("property", "og:image").attr("content");
+        String href = document.getElementsByAttributeValue("property", "og:image").attr("content");
+        return isConnect(href) ? href: null;
     }
 }