Set page status code for SeleniumDownloader. Fixes #1185

code4craft · Dec 18, 2024 · b1cf7ae · b1cf7ae
1 parent 9bb2417
commit b1cf7ae
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 1 deletion.
diff --git a/webmagic-extension/src/main/java/us/codecraft/webmagic/downloader/PhantomJSDownloader.java b/webmagic-extension/src/main/java/us/codecraft/webmagic/downloader/PhantomJSDownloader.java
@@ -6,6 +6,7 @@
 import us.codecraft.webmagic.Request;
 import us.codecraft.webmagic.Task;
 import us.codecraft.webmagic.selector.PlainText;
+import us.codecraft.webmagic.utils.HttpConstant;
 
 import java.io.*;
 
@@ -96,7 +97,7 @@ public Page download(Request request, Task task) {
                 page.setRawText(content);
                 page.setUrl(new PlainText(request.getUrl()));
                 page.setRequest(request);
-                page.setStatusCode(200);
+                page.setStatusCode(HttpConstant.StatusCode.CODE_200);
             }
             onSuccess(page, task);
         } catch (Exception e) {

diff --git a/...-selenium/src/main/java/us/codecraft/webmagic/downloader/selenium/SeleniumDownloader.java b/...-selenium/src/main/java/us/codecraft/webmagic/downloader/selenium/SeleniumDownloader.java
@@ -14,9 +14,11 @@
 import us.codecraft.webmagic.downloader.AbstractDownloader;
 import us.codecraft.webmagic.selector.Html;
 import us.codecraft.webmagic.selector.PlainText;
+import us.codecraft.webmagic.utils.HttpConstant;
 
 import java.io.Closeable;
 import java.io.IOException;
+import java.net.http.HttpRequest;
 import java.util.Map;
 
 /**
@@ -111,6 +113,7 @@ public Page download(Request request, Task task) {
             page.setHtml(new Html(content, request.getUrl()));
             page.setUrl(new PlainText(request.getUrl()));
             page.setRequest(request);
+            page.setStatusCode(HttpConstant.StatusCode.CODE_200);
             onSuccess(page, task);
         } catch (Exception e) {
             logger.warn("download page {} error", request.getUrl(), e);