1. add array_max_count_element() and max_count_element() functions; 2…

…. build with dependencies instead of using teradata's dependencies

.gitignore

@@ -1,4 +1,5 @@
-/target/
+.idea
+*.iml
 
 # Compiled class file
 *.class
@@ -15,11 +16,14 @@
 # Package Files #
 *.jar
 *.war
-*.nar
 *.ear
 *.zip
 *.tar.gz
 *.rar
 
 # virtual machine crash logs, see http://www.java.com/en/download/help/error_hotspot.xml
 hs_err_pid*
+
+# idea
+target/
+out/

README.md

@@ -1,17 +1,23 @@
 # Installation
-1. `mvn package`
-2. Copy `presto-udf-*.jar` to `PRESTO_HOME/plugin/teradata-functions/` in all presto nodes
-(copy to this directory because it has all jars we need)
-3. Restart presto
+1. `mvn clean assembly:assembly`
+2. Copy `presto-udf-*-jar-with-dependencies.jar` to `${PRESTO_HOME}/plugin/custom-functions/` in all presto nodes.
+(create directory if not exists)
+3. Restart presto cluster
 
 
 # Functions
-| Function              | Return Type | Argument Types | Description              | Usage                                 |
-|-----------------------|-------------|----------------|--------------------------|---------------------------------------|
-| first_day             | date        | date           | first day of month       | first_day(current_date)               |
-| last_day              | date        | date           | last day of month        | last_day(current_date)                |
-| to_datetime           | timestamp   | date, varchar  | combine the two args     | to_datetime(current_date, '23:59:59') |
-| last_second           | timestamp   | date           | last second of the date  | last_second(current_date)             |
-| yesterday_last_second | timestamp   |                | last second of yesterday | yesterday_last_second()               |
-| yesterday             | date        |                | yesterday                | yesterday()                           |
+## Scalar Functions
+| Function                | Return Type | Argument Types | Description                                                                          | Usage                                 |
+| ----------------------- | ----------- | -------------- | ------------------------------------------------------------------------------------ | ------------------------------------- |
+| first_day               | date        | date           | first day of month                                                                   | first_day(current_date)               |
+| last_day                | date        | date           | last day of month                                                                    | last_day(current_date)                |
+| to_datetime             | timestamp   | date, varchar  | combine the two args                                                                 | to_datetime(current_date, '23:59:59') |
+| last_second             | timestamp   | date           | last second of the date                                                              | last_second(current_date)             |
+| yesterday_last_second   | timestamp   |                | last second of yesterday                                                             | yesterday_last_second()               |
+| yesterday               | date        |                | yesterday                                                                            | yesterday()                           |
+| array_max_count_element | T           | array(T)       | Get maximum count element (null is not counting; if has multiple return one of them) | array_max_count_element(array['1'])   |
 
+## Aggregate Functions
+| Function          | Return Type | Argument Types | Description                                                                          | Usage                   |
+| ----------------- | ----------- | -------------- | ------------------------------------------------------------------------------------ | ----------------------- |
+| max_count_element | VARCHAR     | array(VARCHAR) | Get maximum count element (null is not counting; if has multiple return one of them) | max_count_element(name) |

pom.xml

@@ -4,24 +4,25 @@
          xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
     <modelVersion>4.0.0</modelVersion>
 
-    <groupId>com.archon</groupId>
+    <groupId>com.github.archongum</groupId>
     <artifactId>presto-udf</artifactId>
-    <version>2</version>
+    <version>3</version>
 
     <properties>
-        <presto.version>305</presto.version>
+        <presto.version>318</presto.version>
         <maven.compiler.source>1.8</maven.compiler.source>
         <maven.compiler.target>1.8</maven.compiler.target>
         <maven.compiler.version>3.8.0</maven.compiler.version>
-        <guava.verson>26.0-jre</guava.verson>
+        <jackson.version>2.9.9.3</jackson.version>
+        <junit.version>5.4.2</junit.version>
     </properties>
 
 
     <dependencies>
         <dependency>
-            <groupId>com.google.guava</groupId>
-            <artifactId>guava</artifactId>
-            <version>${guava.verson}</version>
+            <groupId>com.fasterxml.jackson.core</groupId>
+            <artifactId>jackson-databind</artifactId>
+            <version>${jackson.version}</version>
         </dependency>
 
         <dependency>
@@ -32,21 +33,16 @@
         </dependency>
 
         <dependency>
-            <groupId>org.junit.platform</groupId>
-            <artifactId>junit-platform-launcher</artifactId>
-            <version>1.3.1</version>
-            <scope>test</scope>
+            <groupId>io.prestosql</groupId>
+            <artifactId>presto-array</artifactId>
+            <version>${presto.version}</version>
+            <scope>provided</scope>
         </dependency>
+
         <dependency>
             <groupId>org.junit.jupiter</groupId>
-            <artifactId>junit-jupiter-engine</artifactId>
-            <version>5.3.1</version>
-            <scope>test</scope>
-        </dependency>
-        <dependency>
-            <groupId>org.junit.vintage</groupId>
-            <artifactId>junit-vintage-engine</artifactId>
-            <version>5.3.1</version>
+            <artifactId>junit-jupiter-api</artifactId>
+            <version>${junit.version}</version>
             <scope>test</scope>
         </dependency>
     </dependencies>
@@ -61,7 +57,15 @@
                     <encoding>utf-8</encoding>
                 </configuration>
             </plugin>
+
+            <plugin>
+                <artifactId>maven-assembly-plugin</artifactId>
+                <configuration>
+                    <descriptorRefs>
+                        <descriptorRef>jar-with-dependencies</descriptorRef>
+                    </descriptorRefs>
+                </configuration>
+            </plugin>
         </plugins>
     </build>
-
 </project>

src/main/java/com/archon/presto/UdfPlugin.java → src/main/java/com/github/archongum/presto/UdfPlugin.java

@@ -13,13 +13,15 @@
  * See the License for the specific language governing permissions and
  * limitations under the License.
  */
-package com.archon.presto;
+package com.github.archongum.presto;
 
-import com.archon.presto.udf.scalar.CommonFunctions;
-import com.archon.presto.udf.scalar.DateTimeFunctions;
-import com.google.common.collect.ImmutableSet;
+import com.github.archongum.presto.udf.aggregate.MaxCountElementAggregation;
+import com.github.archongum.presto.udf.scalar.ArrayMaxCountElementFunction;
+import com.github.archongum.presto.udf.scalar.CommonFunctions;
+import com.github.archongum.presto.udf.scalar.DateTimeFunctions;
 import io.prestosql.spi.Plugin;
 
+import java.util.HashSet;
 import java.util.Set;
 
 
@@ -28,9 +30,11 @@ public class UdfPlugin implements Plugin
     @Override
     public Set<Class<?>> getFunctions()
     {
-        return ImmutableSet.<Class<?>>builder()
-                .add(DateTimeFunctions.class)
-                .add(CommonFunctions.class)
-                .build();
+        Set<Class<?>> set = new HashSet<>();
+        set.add(ArrayMaxCountElementFunction.class);
+        set.add(CommonFunctions.class);
+        set.add(DateTimeFunctions.class);
+        set.add(MaxCountElementAggregation.class);
+        return set;
     }
 }

src/main/java/com/github/archongum/presto/udf/aggregate/MaxCountElementAggregation.java

@@ -0,0 +1,54 @@
+package com.github.archongum.presto.udf.aggregate;
+
+import com.github.archongum.presto.udf.aggregate.state.MapState;
+import io.airlift.slice.Slice;
+import io.airlift.slice.Slices;
+import io.prestosql.spi.block.BlockBuilder;
+import io.prestosql.spi.function.AggregationFunction;
+import io.prestosql.spi.function.CombineFunction;
+import io.prestosql.spi.function.Description;
+import io.prestosql.spi.function.InputFunction;
+import io.prestosql.spi.function.OutputFunction;
+import io.prestosql.spi.function.SqlType;
+import io.prestosql.spi.type.StandardTypes;
+
+import java.util.Map;
+import java.util.Map.Entry;
+
+import static io.prestosql.spi.type.VarcharType.VARCHAR;
+
+
+@AggregationFunction("max_count_element")
+@Description("Get maximum count element (null is not counting; if has multiple return one of them)")
+public class MaxCountElementAggregation
+{
+    @InputFunction
+    public static void input(MapState state, @SqlType(StandardTypes.VARCHAR) Slice value)
+    {
+        String v = value.toStringUtf8();
+        Map<String, Long> map = state.getMap();
+        Long cnt = map.get(v);
+        if (cnt == null) {
+            map.put(v, 1L);
+        } else {
+            map.put(v, cnt+1);
+        }
+    }
+
+    @CombineFunction
+    public static void combine(MapState state, MapState otherState)
+    {
+        otherState.getMap().forEach((k, v) -> state.getMap().merge(k, v, Long::sum));
+    }
+
+    @OutputFunction(StandardTypes.VARCHAR)
+    public static void output(MapState state, BlockBuilder out)
+    {
+        if (state.getMap().isEmpty()) {
+            out.appendNull();
+        } else {
+            VARCHAR.writeSlice(out,
+                Slices.utf8Slice(state.getMap().entrySet().stream().max(Entry.comparingByValue()).get().getKey()));
+        }
+    }
+}

src/main/java/com/github/archongum/presto/udf/aggregate/state/MapState.java

@@ -0,0 +1,14 @@
+package com.github.archongum.presto.udf.aggregate.state;
+
+import io.prestosql.spi.function.AccumulatorState;
+import io.prestosql.spi.function.AccumulatorStateMetadata;
+
+import java.util.Map;
+
+
+@AccumulatorStateMetadata(stateSerializerClass = MapStateSerializer.class, stateFactoryClass = MapStateFactory.class)
+public interface MapState extends AccumulatorState {
+    Map<String, Long> getMap();
+
+    void setMap(Map<String, Long> value);
+}

src/main/java/com/github/archongum/presto/udf/aggregate/state/MapStateFactory.java

@@ -0,0 +1,88 @@
+package com.github.archongum.presto.udf.aggregate.state;
+
+import io.prestosql.array.ObjectBigArray;
+import io.prestosql.spi.function.AccumulatorStateFactory;
+import io.prestosql.spi.function.GroupedAccumulatorState;
+
+import java.util.HashMap;
+import java.util.Map;
+
+
+/**
+ * @author Archon  8/30/19
+ * @since
+ */
+public class MapStateFactory implements AccumulatorStateFactory<MapState> {
+
+    public static final class SingleMapState implements MapState {
+
+        private Map<String, Long> map = new HashMap<>();
+
+        @Override
+        public Map<String, Long> getMap() {
+            return map;
+        }
+
+        @Override
+        public void setMap(Map<String, Long> value) {
+            this.map = value;
+        }
+
+        @Override
+        public long getEstimatedSize() {
+            return map.size();
+        }
+    }
+
+    public static class GroupedMapState implements GroupedAccumulatorState, MapState {
+
+        private final ObjectBigArray<Map<String, Long>> maps = new ObjectBigArray<>();
+        private long groupId;
+
+        @Override
+        public Map<String, Long> getMap() {
+            return maps.get(groupId);
+        }
+
+        @Override
+        public void setMap(Map<String, Long> value) {
+            maps.set(groupId, value);
+        }
+
+        @Override
+        public void setGroupId(long groupId) {
+            this.groupId = groupId;
+        }
+
+        @Override
+        public void ensureCapacity(long size) {
+            maps.ensureCapacity(size);
+        }
+
+        @Override
+        public long getEstimatedSize() {
+            return maps.sizeOf();
+        }
+    }
+
+    @Override
+    public MapState createSingleState() {
+        return new SingleMapState();
+    }
+
+    @Override
+    public Class<? extends MapState> getSingleStateClass() {
+        return SingleMapState.class;
+    }
+
+
+    @Override
+    public MapState createGroupedState() {
+        return new GroupedMapState();
+    }
+
+    @Override
+    public Class<? extends MapState> getGroupedStateClass() {
+        return GroupedMapState.class;
+    }
+}