From 7f1700400ba2b7464ace0c1896c736e151e3fb4b Mon Sep 17 00:00:00 2001
From: Hunter Stern <hunter@archive.org>
Date: Fri, 1 Aug 2014 17:37:31 -0700
Subject: [PATCH] Allow xml extractor to handle urls in CDATA.

---
 .../main/java/org/archive/modules/extractor/ExtractorXML.java  | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/modules/src/main/java/org/archive/modules/extractor/ExtractorXML.java b/modules/src/main/java/org/archive/modules/extractor/ExtractorXML.java
index 64a1d500..92baf805 100644
--- a/modules/src/main/java/org/archive/modules/extractor/ExtractorXML.java
+++ b/modules/src/main/java/org/archive/modules/extractor/ExtractorXML.java
@@ -56,7 +56,8 @@ public class ExtractorXML extends ContentExtractor {
         Logger.getLogger(ExtractorXML.class.getName());
 
     static final Pattern XML_URI_EXTRACTOR = Pattern
-            .compile("(?s)[\"\'>]\\s*([^<>\\s'\"@]+)\\s*[\"\'<]");
+            .compile("(?s)[\"\'>]\\s*(?:<!\\[CDATA\\[)?([^<>\\s\'\"@]+)\\s*(?:\\]\\]>)?[\"\'<]");    
+    
     // GROUPS:
     // (G1) possible uri