集成验证码 OCR 识别：添加 doLoginWithCaptcha 方法，通过 Ollama qwen3-vl:4b 自动识别验证码并登录

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

src/main/java/com/ets/scraper/EtsScraper.java

@@ -4,11 +4,17 @@ import com.microsoft.playwright.*;
 import com.microsoft.playwright.options.LoadState;
 import com.microsoft.playwright.options.WaitUntilState;
 
+import java.io.BufferedReader;
 import java.io.FileOutputStream;
 import java.io.IOException;
+import java.io.InputStreamReader;
+import java.net.HttpURLConnection;
+import java.net.URL;
+import java.nio.file.Files;
 import java.nio.file.Path;
 import java.time.LocalDateTime;
 import java.time.format.DateTimeFormatter;
+import java.util.Base64;
 
 import static java.nio.file.Files.createDirectories;
 
@@ -22,8 +28,10 @@ public class EtsScraper {
     private static final String USERNAME = "sccw";
     private static final String PASSWORD = "slife@123";
     private static final Path SCREENSHOT_DIR = Path.of("screenshots");
+    private static final String OLLAMA_URL = "http://10.0.1.39:11434";
+    private static final String OLLAMA_MODEL = "qwen3-vl:4b";
 
-    public static void main(String[] args) {
+    public static void main(String[] args) throws Exception {
         try {
             createDirectories(SCREENSHOT_DIR);
          } catch (Exception e) {
@@ -65,8 +73,8 @@ public class EtsScraper {
                   // Close dialog again after page reload
                 closeNotificationDialog(page);
 
-                // Perform login
+                 // Recognize captcha and perform login
-                boolean loggedin = doLogin(page);
+                boolean loggedin = doLoginWithCaptcha(page);
 
                 if (loggedin) {
                     System.out.println("[+] Login successful!");
@@ -147,6 +155,85 @@ public class EtsScraper {
             return true;
          }
      }
+
+
+    private static boolean doLoginWithCaptcha(Page page) throws Exception {
+          // Find and fill username
+        String usernameInput = findInput(page, new String[]{
+                    "input[placeholder*='用户名']",
+                    "input[placeholder*='username']",
+                    "input[placeholder*='账号']",
+                    "input[name*='user']",
+                    "input[name='username']",
+                    "input[type='text']",
+          });
+        if (usernameInput == null) {
+            System.out.println("[-] Could not find username input");
+            return false;
+          }
+
+          // Find and fill password
+        String passwordInput = findInput(page, new String[]{
+                    "input[placeholder*='密码']",
+                    "input[placeholder*='password']",
+                    "input[name*='pass']",
+                    "input[name='password']",
+                    "input[name='pwd']",
+                    "input[type='password']",
+          });
+        if (passwordInput == null) {
+            System.out.println("[-] Could not find password input");
+            return false;
+          }
+
+          // Find and fill captcha
+        String captchaInput = findInput(page, new String[]{
+                    "input[placeholder*='验证码']",
+                    "input[placeholder*='captcha']",
+                    "input[name*='captcha']",
+                    "input[name='code']",
+                    "input[type='text']",
+          });
+        if (captchaInput == null) {
+            System.out.println("[-] Could not find captcha input");
+            return false;
+          }
+
+          // Recognize captcha
+        Path captchaPath = SCREENSHOT_DIR.resolve("captcha.png");
+        System.out.println("[*] Recognizing captcha with Ollama...");
+        String captchaText = recognizeCaptcha(captchaPath);
+        if (captchaText == null || captchaText.isEmpty()) {
+            System.out.println("[-] Failed to recognize captcha");
+            return false;
+          }
+        System.out.println("[+] Captcha recognized: " + captchaText);
+
+        System.out.println("[*] Filling credentials...");
+        page.locator(usernameInput).first().fill(USERNAME);
+        page.locator(passwordInput).first().fill(PASSWORD);
+        page.locator(captchaInput).first().fill(captchaText);
+        sleep(500);
+
+          // Click submit or press Enter
+        String submitBtn = findSubmit(page);
+        if (submitBtn != null) {
+            System.out.println("[*] Clicking submit button: " + submitBtn);
+            page.locator(submitBtn).first().click();
+          } else {
+            System.out.println("[*] No submit button found, pressing Enter");
+            page.locator(captchaInput).first().press("Enter");
+          }
+
+        try {
+            page.waitForLoadState(LoadState.DOMCONTENTLOADED,
+                    new Page.WaitForLoadStateOptions().setTimeout(10000));
+            return true;
+          } catch (Exception e) {
+            System.out.println("[!] Navigation timed out, but credentials were submitted");
+            return true;
+          }
+      }
     private static void downloadCaptcha(Page page) {
         try {
              // Set up listener FIRST, then reload to trigger the request
@@ -170,8 +257,6 @@ public class EtsScraper {
             }
          }
 
-
-
     private static void closeNotificationDialog(Page page) {
           // Find the frame that contains the notification dialog
          Frame dialogFrame = null;
@@ -204,8 +289,6 @@ public class EtsScraper {
          System.out.println("[*] Notification dialog closed");
          }
 
-
-
     private static String findInput(Page page, String[] selectors) {
         for (String selector : selectors) {
             try {
@@ -259,4 +342,54 @@ public class EtsScraper {
             Thread.currentThread().interrupt();
          }
      }
+
+    public static String recognizeCaptcha(Path imagePath) throws Exception {
+         byte[] imageBytes = Files.readAllBytes(imagePath);
+         String base64 = Base64.getEncoder().encodeToString(imageBytes);
+
+         String json = "{"
+                 + "\"model\":\"" + OLLAMA_MODEL + "\","
+                 + "\"messages\":["
+                 + "      {"
+                 + "           \"role\":\"user\","
+                 + "           \"content\":\"识别图中的验证码文字，只返回文字内容，不要有其他解释\","
+                 + "           \"images\":[\"" + base64 + "\"]"
+                 + "      }"
+                 + "]"
+                 + "}";
+
+         URL url = new URL(OLLAMA_URL + "/api/chat");
+         HttpURLConnection conn = (HttpURLConnection) url.openConnection();
+         conn.setRequestMethod("POST");
+         conn.setConnectTimeout(15000);
+         conn.setReadTimeout(30000);
+         conn.setDoOutput(true);
+         conn.setRequestProperty("Content-Type", "application/json; charset=utf-8");
+
+         conn.getOutputStream().write(json.getBytes("utf-8"));
+         conn.getOutputStream().flush();
+         conn.getOutputStream().close();
+
+         try (BufferedReader reader = new BufferedReader(
+                 new InputStreamReader(conn.getInputStream(), "utf-8"))) {
+             StringBuilder sb = new StringBuilder();
+             String line;
+             while ((line = reader.readLine()) != null) {
+                 sb.append(line);
+                 }
+             String response = sb.toString();
+                 // Parse "content":"..." from the JSON response
+             int contentIdx = response.indexOf("\"content\":");
+             if (contentIdx >= 0) {
+                 int start = response.indexOf('"', contentIdx + 10) + 1;
+                 int end = response.indexOf('"', start);
+                 if (start > 0 && end > start) {
+                     return response.substring(start, end).trim();
+                     }
+                 }
+             return null;
+             } finally {
+             conn.disconnect();
+             }
+         }
      }