add AirportsByLatitudeSolution and AirportsByLatitudeProblem

James Lee · James Lee · commit e6b73d4565ea · 2017-03-10T15:24:44.000Z
diff --git a/src/main/java/com/sparkTutorial/rdd/airports/AirportsByLatitudeProblem.java b/src/main/java/com/sparkTutorial/rdd/airports/AirportsByLatitudeProblem.java
@@ -0,0 +1,20 @@
+package com.sparkTutorial.rdd.airports;
+
+public class AirportsByLatitudeProblem {
+
+    public static void main(String[] args) throws Exception {
+
+        /* Create a Spark program to read the airport data from in/airports.text,  find all the airports whose latitude are bigger than 40.
+           Then output the airport's name and the airport's latitude to out/airports_by_latitude.text.
+
+           Each row of the input file contains the following columns:
+           Airport ID, Name of airport, Main city served by airport, Country where airport is located, IATA/FAA code,
+           ICAO Code, Latitude, Longitude, Altitude, Timezone, DST, Timezone in Olson format
+
+           Sample output:
+           "St Anthony", 51.391944
+           "Tofino", 49.082222
+           ...
+         */
+    }
+}
diff --git a/src/main/java/com/sparkTutorial/rdd/airports/AirportsByLatitudeSolution.java b/src/main/java/com/sparkTutorial/rdd/airports/AirportsByLatitudeSolution.java
@@ -0,0 +1,28 @@
+package com.sparkTutorial.rdd.airports;
+
+import com.sparkTutorial.rdd.commons.Utils;
+import org.apache.commons.lang.StringUtils;
+import org.apache.spark.SparkConf;
+import org.apache.spark.api.java.JavaRDD;
+import org.apache.spark.api.java.JavaSparkContext;
+
+public class AirportsByLatitudeSolution {
+
+    public static void main(String[] args) throws Exception {
+
+        SparkConf conf = new SparkConf().setAppName("airports").setMaster("local[2]");
+
+        JavaSparkContext sc = new JavaSparkContext(conf);
+
+        JavaRDD<String> airports = sc.textFile("in/airports.text");
+
+        JavaRDD<String> airportsInUSA = airports.filter(line -> Float.valueOf(line.split(Utils.COMMA_DELIMITER)[6]) > 40);
+
+        JavaRDD<String> airportsNameAndCityNames = airportsInUSA.map(line -> {
+                    String[] splits = line.split(Utils.COMMA_DELIMITER);
+                    return StringUtils.join(new String[]{splits[1], splits[6]}, ",");
+                }
+        );
+        airportsNameAndCityNames.saveAsTextFile("out/airports_by_latitude.text");
+    }
+}
diff --git a/src/main/java/com/sparkTutorial/rdd/airports/AirportsInUsaProblem.java b/src/main/java/com/sparkTutorial/rdd/airports/AirportsInUsaProblem.java
@@ -0,0 +1,20 @@
+package com.sparkTutorial.rdd.airports;
+
+public class AirportsInUsaProblem {
+
+    public static void main(String[] args) throws Exception {
+
+        /* Create a Spark program to read the airport data from in/airports.text, find all the airports which are located in United States
+           and output the airport's name and the city's name to out/airports_in_usa.text.
+
+           Each row of the input file contains the following columns:
+           Airport ID, Name of airport, Main city served by airport, Country where airport is located, IATA/FAA code,
+           ICAO Code, Latitude, Longitude, Altitude, Timezone, DST, Timezone in Olson format
+
+           Sample output:
+           "Putnam County Airport", "Greencastle"
+           "Dowagiac Municipal Airport", "Dowagiac"
+           ...
+         */
+    }
+}
diff --git a/src/main/java/com/sparkTutorial/rdd/airports/AirportsInUsaSolution.java b/src/main/java/com/sparkTutorial/rdd/airports/AirportsInUsaSolution.java
@@ -1,27 +1,28 @@
 package com.sparkTutorial.rdd.airports;
 
+import com.sparkTutorial.rdd.commons.Utils;
 import org.apache.commons.lang.StringUtils;
 import org.apache.spark.SparkConf;
 import org.apache.spark.api.java.JavaRDD;
 import org.apache.spark.api.java.JavaSparkContext;
 
-public class AirportsSolution {
+public class AirportsInUsaSolution {
 
     public static void main(String[] args) throws Exception {
 
-        SparkConf conf = new SparkConf().setAppName("airports").setMaster("local[*]");
+        SparkConf conf = new SparkConf().setAppName("airports").setMaster("local[2]");
 
         JavaSparkContext sc = new JavaSparkContext(conf);
 
         JavaRDD<String> airports = sc.textFile("in/airports.text");
 
-        JavaRDD<String> airportsInUSA = airports.filter(line -> line.split(",")[3].equals("\"United States\""));
+        JavaRDD<String> airportsInUSA = airports.filter(line -> line.split(Utils.COMMA_DELIMITER)[3].equals("\"United States\""));
 
         JavaRDD<String> airportsNameAndCityNames = airportsInUSA.map(line -> {
-                    String[] splits = line.split(",");
+                    String[] splits = line.split(Utils.COMMA_DELIMITER);
                     return StringUtils.join(new String[]{splits[1], splits[2]}, ",");
                 }
         );
-        airportsNameAndCityNames.saveAsTextFile("out/airports.text");
+        airportsNameAndCityNames.saveAsTextFile("out/airports_in_usa.text");
     }
 }
diff --git a/src/main/java/com/sparkTutorial/rdd/airports/AirportsProblem.java b/src/main/java/com/sparkTutorial/rdd/airports/AirportsProblem.java
diff --git a/src/main/java/com/sparkTutorial/rdd/commons/Utils.java b/src/main/java/com/sparkTutorial/rdd/commons/Utils.java
@@ -0,0 +1,10 @@
+package com.sparkTutorial.rdd.commons;
+
+public class Utils {
+
+    private Utils(){
+    };
+
+    // a regular expression which matches commas but not commas within double quotations
+    public static final String COMMA_DELIMITER = ",(?=([^\"]*\"[^\"]*\")*[^\"]*$)";
+}
