PIG-1936: doc updates (chandec via olgan)
git-svn-id: https://svn.apache.org/repos/asf/pig/branches/branch-0.8@1136201 13f79535-47bb-0310-9956-ffa450edef68
diff --git a/CHANGES.txt b/CHANGES.txt
index 5ccbf39..9b1d570 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -28,6 +28,8 @@
IMPROVEMENTS
+PIG-1936: doc updates (chandec via olgan)
+
PIG-1830: Type mismatch error in key from map, when doing GROUP on PigStorageSchema() variable (dvryaboy)
PIG-1886: Add zookeeper jar to list of jars shipped when HBaseStorage used (dvryaboy)
diff --git a/src/docs/src/documentation/content/xdocs/udf.xml b/src/docs/src/documentation/content/xdocs/udf.xml
index ef47850..04bf777 100644
--- a/src/docs/src/documentation/content/xdocs/udf.xml
+++ b/src/docs/src/documentation/content/xdocs/udf.xml
@@ -189,7 +189,9 @@
</source>
<p>For a function to be algebraic, it needs to implement <code>Algebraic</code> interface that consist of definition of three classes derived from <code>EvalFunc</code>. The contract is that the <code>exec</code> function of the <code>Initial</code> class is called once and is passed the original input tuple. Its output is a tuple that contains partial results. The <code>exec</code> function of the <code>Intermed</code> class can be called zero or more times and takes as its input a tuple that contains partial results produced by the <code>Initial</code> class or by prior invocations of the <code>Intermed</code> class and produces a tuple with another partial result. Finally, the <code>exec</code> function of the <code>Final</code> class is called and produces the final result as a scalar type. </p>
-<p>Here's the way to think about this in the Hadoop world. The <code>exec</code> function of the <code>Initial</code> class is invoked once by the <code>map</code> process and produces partial results. The <code>exec</code> function of the <code>Intermed</code> class is invoked once by each <code>combiner</code> invocation (which can happen zero or more times) and also produces partial results. The <code>exec</code> function of the <code>Final</code> class is invoked once by the reducer and produces the final result. </p>
+
+<p>Here's the way to think about this in the Hadoop world. The <code>exec</code> function of the <code>Initial</code> class is invoked once for each input tuple by the <code>map</code> process and produces partial results. The <code>exec</code> function of the <code>Intermed</code> class is invoked once by each <code>combiner</code> invocation (which can happen zero or more times) and also produces partial results. The <code>exec</code> function of the <code>Final</code> class is invoked once by the reducer and produces the final result. </p>
+
<p>Take a look at the <code>COUNT</code> implementation to see how this is done. Note that the <code>exec</code> function of the <code>Initial</code> and <code>Intermed</code> classes is parameterized with <code>Tuple</code> and the <code>exec</code> of the <code>Final</code> class is parameterized with the real type of the function, which in the case of the <code>COUNT</code> is <code>Long</code>. Also, note that the fully-qualified name of the class needs to be returned from <code>getInitial</code>, <code>getIntermed</code>, and <code>getFinal</code> methods. </p>
</section>
