SBT Support

.gitignore

@@ -6,3 +6,5 @@ target
 tf-sandbox
 spark-warehouse/
 metastore_db/
+project/project/
+test-output.tfr

.travis.yml

@@ -0,0 +1,21 @@
+language: scala
+
+# Cache settings here are based on latest SBT documentation.
+cache:
+  directories:
+    - $HOME/.ivy2/cache
+    - $HOME/.sbt/boot/
+
+before_cache:
+  # Tricks to avoid unnecessary cache updates
+  - find $HOME/.ivy2 -name "ivydata-*.properties" -delete
+  - find $HOME/.sbt -name "*.lock" -delete
+
+scala:
+   - 2.11.8
+
+jdk:
+  - oraclejdk8
+
+script:
+  - sbt ++$TRAVIS_SCALA_VERSION clean publish-local

README.md

@@ -1,3 +1,5 @@
+[![Build Status](https://travis-ci.org/tapanalyticstoolkit/spark-tensorflow-connector.svg?branch=sbt)](https://travis-ci.org/tapanalyticstoolkit/spark-tensorflow-connector)
+
 # spark-tensorflow-connector
 
 This repo contains a library for loading and storing TensorFlow records with [Apache Spark](http://spark.apache.org/).
@@ -19,19 +21,34 @@ None.
 2. [Apache Maven](https://maven.apache.org/)
 
 ## Building the library
-Build the library using Maven as shown below.
+You can build library using both Maven and SBT build tools
+
+#### Maven
+Build the library using Maven(3.3) as shown below
 
 ```sh
 mvn clean install
 ```
 
+#### SBT 
+Build the library using SBT(0.13.13) as show below
+```sh
+sbt clean assembly
+```
+
 ## Using Spark Shell
 Run this library in Spark using the `--jars` command line option in `spark-shell` or `spark-submit`. For example:
 
+Maven Jars
 ```sh
 $SPARK_HOME/bin/spark-shell --jars target/spark-tensorflow-connector-1.0-SNAPSHOT.jar,target/lib/tensorflow-hadoop-1.0-01232017-SNAPSHOT-shaded-protobuf.jar
 ```
 
+SBT Jars
+```sh
+$SPARK_HOME/bin/spark-shell --jars target/scala-2.11/spark-tensorflow-connector-assembly-1.0-SNAPSHOT.jar
+```
+
 The following code snippet demonstrates usage.
 
 ```scala
@@ -40,7 +57,7 @@ import org.apache.spark.sql.{ DataFrame, Row }
 import org.apache.spark.sql.catalyst.expressions.GenericRow
 import org.apache.spark.sql.types._
 
-val path = s"$TF_SANDBOX_DIR/test-output.tfr"
+val path = "test-output.tfr"
 val testRows: Array[Row] = Array(
 new GenericRow(Array[Any](11, 1, 23L, 10.0F, 14.0, List(1.0, 2.0), "r1")),
 new GenericRow(Array[Any](21, 2, 24L, 12.0F, 15.0, List(2.0, 2.0), "r2")))

build.sbt

@@ -0,0 +1,52 @@
+scalaVersion in Global := "2.11.8"
+
+def ProjectName(name: String,path:String): Project =  Project(name, file(path))
+
+resolvers in Global ++= Seq("https://tap.jfrog.io/tap/public" at "https://tap.jfrog.io/tap/public" ,
+           "https://tap.jfrog.io/tap/public-snapshots" at "https://tap.jfrog.io/tap/public-snapshots" ,
+           "https://repo.maven.apache.org/maven2" at "https://repo.maven.apache.org/maven2" )
+
+val `junit_junit` = "junit" % "junit" % "4.12"
+
+val `org.apache.hadoop_hadoop-yarn-api` = "org.apache.hadoop" % "hadoop-yarn-api" % "2.7.3"
+
+val `org.apache.spark_spark-core_2.11` = "org.apache.spark" % "spark-core_2.11" % "2.1.0"
+
+val `org.apache.spark_spark-sql_2.11` = "org.apache.spark" % "spark-sql_2.11" % "2.1.0"
+
+val `org.apache.spark_spark-mllib_2.11` = "org.apache.spark" % "spark-mllib_2.11" % "2.1.0"
+
+val `org.scalatest_scalatest_2.11` = "org.scalatest" % "scalatest_2.11" % "2.2.6"
+
+val `org.tensorflow_tensorflow-hadoop` = "org.tensorflow" % "tensorflow-hadoop" % "1.0-01232017-SNAPSHOT"
+
+
+spName := "spark-tensorflow-connector"
+
+sparkVersion := "2.1.0"
+
+sparkComponents ++= Seq("sql", "mllib")
+
+spIgnoreProvided := true
+
+version := "1.0-SNAPSHOT"
+
+name := "spark-tensorflow-connector"
+
+organization := "org.trustedanalytics"
+
+libraryDependencies in Global ++= Seq(`org.tensorflow_tensorflow-hadoop` classifier "shaded-protobuf",
+   `org.scalatest_scalatest_2.11` % "test" ,
+   `org.apache.spark_spark-sql_2.11` % "provided" ,
+   `org.apache.spark_spark-mllib_2.11` % "test" classifier "tests",
+   `org.apache.spark_spark-core_2.11` % "provided" ,
+   `org.apache.hadoop_hadoop-yarn-api` % "provided" ,
+   `junit_junit` % "test" )
+
+assemblyExcludedJars in assembly := {
+  val cp = (fullClasspath in assembly).value
+  cp filterNot {x => List("spark-tensorflow-connector-1.0-SNAPSHOT.jar",
+    "tensorflow-hadoop-1.0-01232017-SNAPSHOT-shaded-protobuf.jar").contains(x.data.getName)}
+}
+
+licenses := Seq("Apache License 2.0" -> url("http://www.apache.org/licenses/LICENSE-2.0.html"))

project/build.properties

@@ -0,0 +1 @@
+sbt.version=0.13.13

project/plugins.sbt

@@ -0,0 +1,5 @@
+resolvers += "bintray-spark-packages" at "https://dl.bintray.com/spark-packages/maven/"
+
+addSbtPlugin("com.eed3si9n" % "sbt-assembly" % "0.14.3")
+
+addSbtPlugin("org.spark-packages" % "sbt-spark-package" % "0.2.5")