bugfix httpresponse content length skipped when empty

DigitalPebble · Feb 6, 2014 · 9b0c4ff · 9b0c4ff
1 parent ff1d0db
commit 9b0c4ff
Showing 1 changed file with 2 additions and 1 deletion.
diff --git a/io/src/main/java/com/digitalpebble/behemoth/io/warc/HttpResponse.java b/io/src/main/java/com/digitalpebble/behemoth/io/warc/HttpResponse.java
@@ -24,6 +24,7 @@
 import java.io.InputStream;
 import java.io.PushbackInputStream;
 
+import org.apache.commons.lang.StringUtils;
 import org.apache.nutch.metadata.Metadata;
 import org.apache.nutch.net.protocols.Response;
 import org.apache.nutch.protocol.ProtocolException;
@@ -81,7 +82,7 @@ private void readPlainContent(InputStream in) throws IOException {
 
         int contentLength = Integer.MAX_VALUE; // get content length
         String contentLengthString = headers.get(Response.CONTENT_LENGTH);
-        if (contentLengthString != null) {
+        if (StringUtils.isNotBlank(contentLengthString)) {
             contentLengthString = contentLengthString.trim();
             try {
                 contentLength = Integer.parseInt(contentLengthString);