From 36d7ab3489dc2c90b8ebc87c38b8524062f2e94f Mon Sep 17 00:00:00 2001
From: Harsh J <harsh@apache.org>
Date: Fri, 23 Sep 2011 07:26:00 +0000
Subject: [PATCH] HADOOP-7542. Change Configuration XML format to 1.1 to
 support for serializing additional characters Contributed by Christopher
 Egner.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1174562 13f79535-47bb-0310-9956-ffa450edef68
---
 hadoop-common-project/hadoop-common/CHANGES.txt  |  2 ++
 .../org/apache/hadoop/conf/Configuration.java    |  4 ++++
 .../apache/hadoop/conf/TestConfiguration.java    | 16 ++++++++++++++--
 3 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/hadoop-common-project/hadoop-common/CHANGES.txt b/hadoop-common-project/hadoop-common/CHANGES.txt
index 6245bcf8f9..c69ec9d3ee 100644
--- a/hadoop-common-project/hadoop-common/CHANGES.txt
+++ b/hadoop-common-project/hadoop-common/CHANGES.txt
@@ -26,6 +26,8 @@ Trunk (unreleased changes)
     HADOOP-7621. alfredo config should be in a file not readable by users
                  (Alejandro Abdelnur via atm)
 
+    HADOOP-7542. Change Configuration XML format to 1.1 to support for serializing additional characters (Christopher Egner via harsh)
+
 Release 0.23.0 - Unreleased
 
   INCOMPATIBLE CHANGES
diff --git a/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java b/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java
index 4fb1d19066..c310aa65e6 100644
--- a/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java
+++ b/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java
@@ -1632,6 +1632,10 @@ private synchronized Document asXmlDocument() throws IOException {
     try {
       doc =
         DocumentBuilderFactory.newInstance().newDocumentBuilder().newDocument();
+
+      // Allow a broader set of control characters to appear in job confs.
+      // cf https://issues.apache.org/jira/browse/MAPREDUCE-109 
+      doc.setXmlVersion( "1.1" );
     } catch (ParserConfigurationException pe) {
       throw new IOException(pe);
     }
diff --git a/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/conf/TestConfiguration.java b/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/conf/TestConfiguration.java
index f9f14fb848..5842db199d 100644
--- a/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/conf/TestConfiguration.java
+++ b/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/conf/TestConfiguration.java
@@ -58,7 +58,7 @@ protected void tearDown() throws Exception {
   }
   
   private void startConfig() throws IOException{
-    out.write("<?xml version=\"1.0\"?>\n");
+    out.write("<?xml version=\"1.1\"?>\n");
     out.write("<configuration>\n");
   }
 
@@ -221,6 +221,18 @@ public void testCommentsInValue() throws IOException {
     assertEquals("this  contains a comment", conf.get("my.comment"));
   }
   
+  public void testControlAInValue() throws IOException {
+    out = new BufferedWriter(new FileWriter(CONFIG));
+    startConfig();
+    appendProperty("my.char", "&#1;");
+    appendProperty("my.string", "some&#1;string");
+    endConfig();
+    Path fileResource = new Path(CONFIG);
+    conf.addResource(fileResource);
+    assertEquals("\u0001", conf.get("my.char"));
+    assertEquals("some\u0001string", conf.get("my.string"));
+  }
+
   public void testTrim() throws IOException {
     out=new BufferedWriter(new FileWriter(CONFIG));
     startConfig();
@@ -298,7 +310,7 @@ public void testWriteXml() throws IOException {
     conf.writeXml(baos);
     String result = baos.toString();
     assertTrue("Result has proper header", result.startsWith(
-        "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?><configuration>"));
+        "<?xml version=\"1.1\" encoding=\"UTF-8\" standalone=\"no\"?><configuration>"));
     assertTrue("Result has proper footer", result.endsWith("</configuration>"));
   }