Revamped Deltastreamer (#93)

* Add analytics to site

* Fix ugly favicon

* New & Improved HoodieDeltaStreamer

 - Can incrementally consume from HDFS or Kafka, with exactly-once semantics!
 - Supports Json/Avro data, Source can also do custom things
 - Source is totally pluggable, via reflection
 - Key generation is pluggable, currently added SimpleKeyGenerator
 - Schema provider is pluggable, currently Filebased schemas
 - Configurable field to break ties during preCombine
 - Finally, can also plugin the HoodieRecordPayload, to get other merge types than overwriting
 - Handles efficient avro serialization in Spark

 Pending :
 - Rewriting of HiveIncrPullSource
 - Hive sync via hoodie-hive
 - Cleanup & tests

* Minor fixes from master rebase

* Implementation of HiveIncrPullSource
 - Copies commit by commit from source to target

* Adding TimestampBasedKeyGenerator
 - Supports unix time & date strings

hoodie-client/src/main/java/com/uber/hoodie/config/HoodieWriteConfig.java

@@ -27,6 +27,7 @@ import javax.annotation.concurrent.Immutable;
 import java.io.File;
 import java.io.FileReader;
 import java.io.IOException;
+import java.io.InputStream;
 import java.util.Properties;
 
 /**
@@ -230,6 +231,15 @@ public class HoodieWriteConfig extends DefaultHoodieConfig {
             }
         }
 
+        public Builder fromInputStream(InputStream inputStream) throws IOException {
+            try {
+                this.props.load(inputStream);
+                return this;
+            } finally {
+                inputStream.close();
+            }
+        }
+
 
         public Builder withPath(String basePath) {
             props.setProperty(BASE_PATH_PROP, basePath);