From afc2a55e495c2ce184d57266b53f1d56dbd86ed2 Mon Sep 17 00:00:00 2001
From: szznax <szznax@daa5b2f2-a927-0410-8b2d-f5f262fa301a>
Date: Fri, 13 Aug 2010 00:37:55 +0000
Subject: [PATCH] fix for [HER-1808] uncaught URLDecoder
 IllegalArgumentException in ExtractorHTML * ExtractorHTML.java     still
 consider values rejected by URLDecoder

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

diff --git a/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java b/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
index 0584aa36..94945a0c 100644
--- a/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
+++ b/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
@@ -538,6 +538,9 @@ public class ExtractorHTML extends ContentExtractor implements InitializingBean
             if (encodedKeyVal.length == 2) try {
                 String value = URLDecoder.decode(encodedKeyVal[1], "UTF-8");
                 considerIfLikelyUri(curi, value, valueContext, hop);
+            } catch (IllegalArgumentException e) {
+                // still consider values rejected by URLDecoder
+                considerIfLikelyUri(curi, encodedKeyVal[1], valueContext, hop);
             } catch (UnsupportedEncodingException e) {
                 throw new AssertionError("all jvms must support UTF-8, and yet somehow this happened: " + e);
             }