TAJO-2183: Add Reading Functionality from MongoDB to Tajo Storage

tajo-common/src/main/java/org/apache/tajo/BuiltinStorages.java

@@ -31,6 +31,7 @@ public class BuiltinStorages {
   public static final String SEQUENCE_FILE = "SEQUENCEFILE";
   public static final String AVRO = "AVRO";
   public static final String HBASE = "HBASE";
+  public static final String MONGODB = "MONGODB";
   public static final String SYSTEM = "SYSTEM";
   public static final String EX_HTTP_JSON = "EX_HTTP_JSON";
   public static final String KAFKA = "KAFKA";

tajo-docs/src/main/sphinx/storage_plugins/mongodb.rst

@@ -0,0 +1,49 @@
+*******************
+MongoDB Integration
+*******************
+
+Overview
+========
+
+Apache Tajo™ storage modules contains a storage plugin for MongoDB which allows Apache Tajo to access databases and collections in MongoDB. In order to use MongoDB storage plugin user needs to follow following steps.
+
+* Download Mongo Java Driver 3.2.2 and add to the class path.
+* Setup a MongoDB server in a localhost or another. 
+* Configure the cluster using storage-site.json
+
+Download MongoDB Java Driver
+==========================
+
+You can download latest version of MongoDB java driver from https://mongodb.github.io/mongo-java-driver/ 
+Download the .jar file and place it in the class path. You can place the file in TAJO_HOME or you can place the .jar file somewhere else and configure the CLASS_PATH variable from conf/tajo-env.sh (or tajo-env.cmd). 
+
+Setup a Mongo Server
+====================
+In order to proceed you need to have a running mongo server instance. If you are new to MongoDB just follow the Getting Started Guide(https://docs.mongodb.com/getting-started/shell/). After setting up the server, find the host and port to connect to the particular server. If it is a local installation by default, the port is 27017.
+
+
+Configure the cluster
+=====================
+
+MongoDB storage handler configuration will be like this.
+First you have to register the StorageHandler.Then register the space. Update the storage-site.json as following.
+
+.. code-block:: json
+
+    {
+      "storages": {
+        "mongodb": {
+            "handler": "org.apache.tajo.storage.mongodb.MongoDBTableSpace",
+        "default-format": "text"
+        }
+      },
+        "spaces": {
+            "space_name": {
+            "uri": "mongodb://<host>:<port>/<database_name>?user=<username>&password=<password>",
+            "configs": {
+                "mapped_database": "<mapped_database_name>"
+                  }
+            }
+        }
+    }
+

tajo-plan/src/main/java/org/apache/tajo/plan/verifier/PreLogicalPlanVerifier.java

@@ -255,6 +255,7 @@ public Expr visitCreateTable(Context context, Stack<Expr> stack, CreateTable exp
         if (expr.hasSelfDescSchema()) {
           // TODO: support other types like Parquet and ORC.
           if (!expr.getStorageType().equalsIgnoreCase(BuiltinStorages.JSON) &&
+                  !expr.getStorageType().equalsIgnoreCase(BuiltinStorages.MONGODB) &&
               !expr.getStorageType().equalsIgnoreCase(BuiltinStorages.EX_HTTP_JSON)) {
             if (expr.getStorageType().equalsIgnoreCase(BuiltinStorages.PARQUET) ||
                 expr.getStorageType().equalsIgnoreCase(BuiltinStorages.ORC)) {

tajo-storage/pom.xml

@@ -40,6 +40,7 @@
     <module>tajo-storage-jdbc</module>
     <module>tajo-storage-pgsql</module>
     <module>tajo-storage-s3</module>
+    <module>tajo-storage-mongodb</module>
     <module>tajo-storage-kafka</module>
   </modules>
 
@@ -138,8 +139,8 @@
                       run cp -r ${basedir}/tajo-storage-hdfs/target/tajo-storage-hdfs-${project.version}*.jar .
                       run cp -r ${basedir}/tajo-storage-hbase/target/tajo-storage-hbase-${project.version}*.jar .
                       run cp -r ${basedir}/tajo-storage-s3/target/tajo-storage-s3-${project.version}*.jar .
+                      run cp -r ${basedir}/tajo-storage-mongodb/target/tajo-storage-mongodb-${project.version}*.jar .
                       run cp -r ${basedir}/tajo-storage-kafka/target/tajo-storage-kafka-${project.version}*.jar .
-
                       echo
                       echo "Tajo Storage dist layout available at: ${project.build.directory}/tajo-storage-${project.version}"
                       echo

tajo-storage/tajo-storage-common/src/main/resources/storage-default.json

@@ -32,6 +32,10 @@
       "handler": "org.apache.tajo.storage.s3.S3TableSpace",
       "default-format": "text"
     },
+    "mongodb": {
+      "handler": "org.apache.tajo.storage.mongodb.MongoDBTableSpace",
+      "default-format": "text"
+    },
     "kafka": {
       "handler": "org.apache.tajo.storage.kafka.KafkaTablespace",
       "default-format": "kafka"

tajo-storage/tajo-storage-common/src/main/resources/storage-default.xml

@@ -39,7 +39,7 @@
   <!--- Registered Scanner Handler -->
   <property>
     <name>tajo.storage.scanner-handler</name>
-    <value>text,json,regex,raw,draw,rcfile,row,parquet,orc,sequencefile,avro,hbase,ex_http_json,kafka</value>
+    <value>text,json,regex,raw,draw,rcfile,row,parquet,orc,sequencefile,avro,hbase,mongodb,ex_http_json,kafka</value>
   </property>
 
   <!--- Fragment Class Configurations -->
@@ -55,6 +55,10 @@
     <name>tajo.storage.fragment.kind.jdbc</name>
     <value>org.apache.tajo.storage.jdbc.JdbcFragment</value>
   </property>
+  <property>
+    <name>tajo.storage.fragment.kind.mongodb</name>
+    <value>org.apache.tajo.storage.mongodb.MongoDBFragment</value>
+  </property>
   <property>
     <name>tajo.storage.fragment.kind.example-http</name>
     <value>org.apache.tajo.storage.http.ExampleHttpFileFragment</value>
@@ -75,6 +79,10 @@
     <name>tajo.storage.fragment.serde.jdbc</name>
     <value>org.apache.tajo.storage.jdbc.JdbcFragmentSerde</value>
   </property>
+  <property>
+    <name>tajo.storage.fragment.serde.mongodb</name>
+    <value>org.apache.tajo.storage.mongodb.MongoDBFragmentSerde</value>
+  </property>
   <property>
     <name>tajo.storage.fragment.serde.example-http</name>
     <value>org.apache.tajo.storage.http.ExampleHttpFileFragmentSerde</value>
@@ -144,7 +152,10 @@
     <name>tajo.storage.scanner-handler.hbase.class</name>
     <value>org.apache.tajo.storage.hbase.HBaseScanner</value>
   </property>
-
+  <property>
+    <name>tajo.storage.scanner-handler.mongodb.class</name>
+    <value>org.apache.tajo.storage.mongodb.MongoDBScanner</value>
+  </property>
   <property>
     <name>tajo.storage.scanner-handler.ex_http_json.class</name>
     <value>org.apache.tajo.storage.http.ExampleHttpJsonScanner</value>