HADOOP-14835. mvn site build throws SAX errors. Contributed by Andrew Wang and Sean Mackrory.

(cherry picked from commit 3cf3540f19)

BUILDING.txt

@@ -308,6 +308,8 @@ Create a local staging version of the website (in /tmp/hadoop-site)
 
   $ mvn clean site -Preleasedocs; mvn site:stage -DstagingDirectory=/tmp/hadoop-site
 
+Note that the site needs to be built in a second pass after other artifacts.
+
 ----------------------------------------------------------------------------------
 Installing Hadoop
 

dev-support/bin/create-release

@@ -564,6 +564,7 @@ function makearelease
     "${MVN}" "${MVN_ARGS[@]}" install \
       site site:stage \
       -DskipTests \
+      -DskipShade \
       -Pdist,src \
       "${DOCFLAGS}"
 

hadoop-mapreduce-project/hadoop-mapreduce-client/pom.xml

@@ -196,6 +196,13 @@
         <jdiff.stability>-unstable</jdiff.stability>
         <jdiff.javadoc.maxmemory>512m</jdiff.javadoc.maxmemory>
       </properties>
+      <dependencies>
+        <dependency>
+          <groupId>xerces</groupId>
+          <artifactId>xercesImpl</artifactId>
+          <version>${xerces.jdiff.version}</version>
+        </dependency>
+      </dependencies>
       <build>
         <plugins>
           <plugin>
@@ -238,6 +245,14 @@
                       <outputDirectory>${project.build.directory}</outputDirectory>
                       <destFileName>hadoop-annotations.jar</destFileName>
                     </artifactItem>
+                    <artifactItem>
+                      <groupId>xerces</groupId>
+                      <artifactId>xercesImpl</artifactId>
+                      <version>${xerces.version.jdiff}</version>
+                      <overWrite>false</overWrite>
+                      <outputDirectory>${project.build.directory}</outputDirectory>
+                      <destFileName>xerces.jar</destFileName>
+                    </artifactItem>
                   </artifactItems>
                 </configuration>
               </execution>
@@ -275,7 +290,7 @@
                            sourceFiles="${dev-support.relative.dir}/jdiff/Null.java"
                            maxmemory="${jdiff.javadoc.maxmemory}">
                     <doclet name="org.apache.hadoop.classification.tools.IncludePublicAnnotationsJDiffDoclet"
-                            path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar">
+                            path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar:${project.build.directory}/xerces.jar">
                       <param name="-oldapi" value="${project.name} ${jdiff.stable.api}"/>
                       <param name="-newapi" value="${project.name} ${project.version}"/>
                       <param name="-oldapidir" value="${basedir}/${dev-support.relative.dir}/jdiff"/>

hadoop-project-dist/pom.xml

@@ -152,6 +152,13 @@
         <jdiff.compatibility></jdiff.compatibility>
         <jdiff.javadoc.maxmemory>512m</jdiff.javadoc.maxmemory>
       </properties>
+      <dependencies>
+        <dependency>
+          <groupId>xerces</groupId>
+          <artifactId>xercesImpl</artifactId>
+          <version>${xerces.jdiff.version}</version>
+        </dependency>
+      </dependencies>
       <build>
         <plugins>
           <plugin>
@@ -194,6 +201,14 @@
                       <outputDirectory>${project.build.directory}</outputDirectory>
                       <destFileName>hadoop-annotations.jar</destFileName>
                     </artifactItem>
+                    <artifactItem>
+                      <groupId>xerces</groupId>
+                      <artifactId>xercesImpl</artifactId>
+                      <version>${xerces.jdiff.version}</version>
+                      <overWrite>false</overWrite>
+                      <outputDirectory>${project.build.directory}</outputDirectory>
+                      <destFileName>xerces.jar</destFileName>
+                    </artifactItem>
                   </artifactItems>
                 </configuration>
               </execution>
@@ -259,7 +274,7 @@
                              sourceFiles="${basedir}/dev-support/jdiff/Null.java"
                              maxmemory="${jdiff.javadoc.maxmemory}">
                       <doclet name="org.apache.hadoop.classification.tools.IncludePublicAnnotationsJDiffDoclet"
-                              path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar">
+                              path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar:${project.build.directory}/xerces.jar">
                         <param name="-oldapi" value="${project.name} ${jdiff.stable.api}"/>
                         <param name="-newapi" value="${project.name} ${project.version}"/>
                         <param name="-oldapidir" value="${basedir}/dev-support/jdiff"/>

hadoop-project/pom.xml

@@ -45,6 +45,8 @@
     <!-- These 2 versions are defined here because they are used -->
     <!-- JDIFF generation from embedded ant in the antrun plugin -->
     <jdiff.version>1.0.9</jdiff.version>
+    <!-- Version number for xerces used by JDiff -->
+    <xerces.jdiff.version>2.11.0</xerces.jdiff.version>
 
     <kafka.version>0.8.2.1</kafka.version>
     <hbase.version>1.2.6</hbase.version>

hadoop-yarn-project/hadoop-yarn/pom.xml

@@ -113,6 +113,13 @@
         <jdiff.compatibility></jdiff.compatibility>
         <jdiff.javadoc.maxmemory>512m</jdiff.javadoc.maxmemory>
       </properties>
+      <dependencies>
+        <dependency>
+          <groupId>xerces</groupId>
+          <artifactId>xercesImpl</artifactId>
+          <version>${xerces.jdiff.version}</version>
+        </dependency>
+      </dependencies>
       <build>
         <plugins>
           <plugin>
@@ -155,6 +162,14 @@
                       <outputDirectory>${project.build.directory}</outputDirectory>
                       <destFileName>hadoop-annotations.jar</destFileName>
                     </artifactItem>
+                    <artifactItem>
+                      <groupId>xerces</groupId>
+                      <artifactId>xercesImpl</artifactId>
+                      <version>${xerces.version.jdiff}</version>
+                      <overWrite>false</overWrite>
+                      <outputDirectory>${project.build.directory}</outputDirectory>
+                      <destFileName>xerces.jar</destFileName>
+                    </artifactItem>
                   </artifactItems>
                 </configuration>
               </execution>
@@ -193,7 +208,7 @@
                       sourceFiles="${dev-support.relative.dir}/jdiff/Null.java"
                              maxmemory="${jdiff.javadoc.maxmemory}">
                       <doclet name="org.apache.hadoop.classification.tools.IncludePublicAnnotationsJDiffDoclet"
-                              path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar">
+                              path="${project.build.directory}/hadoop-annotations.jar:${project.build.directory}/jdiff.jar:${project.build.directory}/xerces.jar">
                         <param name="-oldapi" value="${project.name} ${jdiff.stable.api}"/>
                         <param name="-newapi" value="${project.name} ${project.version}"/>
                         <param name="-oldapidir" value="${basedir}/${dev-support.relative.dir}/jdiff"/>