HUDI-243 Rename HoodieInputFormat and HoodieRealtimeInputFormat to HoodieParquetInputFormat and HoodieParquetRealtimeInputFormat

hudi-client/src/test/java/org/apache/hudi/common/HoodieMergeOnReadTestUtils.java

@@ -35,7 +35,7 @@ import org.apache.hadoop.mapred.JobConf;
 import org.apache.hadoop.mapred.RecordReader;
 import org.apache.hudi.common.model.HoodieTestUtils;
 import org.apache.hudi.common.util.HoodieAvroUtils;
-import org.apache.hudi.hadoop.realtime.HoodieRealtimeInputFormat;
+import org.apache.hudi.hadoop.realtime.HoodieParquetRealtimeInputFormat;
 
 /**
  * Utility methods to aid in testing MergeOnRead (workaround for HoodieReadClient for MOR)
@@ -46,7 +46,7 @@ public class HoodieMergeOnReadTestUtils {
       throws IOException {
     JobConf jobConf = new JobConf();
     Schema schema = HoodieAvroUtils.addMetadataFields(Schema.parse(HoodieTestDataGenerator.TRIP_EXAMPLE_SCHEMA));
-    HoodieRealtimeInputFormat inputFormat = new HoodieRealtimeInputFormat();
+    HoodieParquetRealtimeInputFormat inputFormat = new HoodieParquetRealtimeInputFormat();
     setPropsForInputFormat(inputFormat, jobConf, schema, basePath);
     return inputPaths.stream().map(path -> {
       setInputPath(jobConf, path);
@@ -76,8 +76,8 @@ public class HoodieMergeOnReadTestUtils {
     }).get();
   }
 
-  private static void setPropsForInputFormat(HoodieRealtimeInputFormat inputFormat, JobConf jobConf, Schema schema,
+  private static void setPropsForInputFormat(HoodieParquetRealtimeInputFormat inputFormat, JobConf jobConf,
-      String basePath) {
+      Schema schema, String basePath) {
     List<Schema.Field> fields = schema.getFields();
     String names = fields.stream().map(f -> f.name().toString()).collect(Collectors.joining(","));
     String postions = fields.stream().map(f -> String.valueOf(f.pos())).collect(Collectors.joining(","));

hudi-hadoop-mr/src/main/java/com/uber/hoodie/hadoop/HoodieInputFormat.java

@@ -18,9 +18,11 @@
 
 package com.uber.hoodie.hadoop;
 
+import org.apache.hudi.hadoop.HoodieParquetInputFormat;
+
 /**
  * Temporary class to allow seamless migration of com.uber.hoodie to org.apache.hudi
  */
-public class HoodieInputFormat extends org.apache.hudi.hadoop.HoodieInputFormat {
+public class HoodieInputFormat extends HoodieParquetInputFormat {
 
 }

hudi-hadoop-mr/src/main/java/com/uber/hoodie/hadoop/realtime/HoodieRealtimeInputFormat.java

@@ -18,9 +18,11 @@
 
 package com.uber.hoodie.hadoop.realtime;
 
+import org.apache.hudi.hadoop.realtime.HoodieParquetRealtimeInputFormat;
+
 /**
  * Temporary class to allow seamless migration of com.uber.hoodie to org.apache.hudi
  */
-public class HoodieRealtimeInputFormat extends org.apache.hudi.hadoop.realtime.HoodieRealtimeInputFormat {
+public class HoodieRealtimeInputFormat extends HoodieParquetRealtimeInputFormat {
 
 }

hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/HoodieParquetInputFormat.java

@@ -57,9 +57,9 @@ import org.apache.log4j.Logger;
  * Hoodie/Non-Hoodie datasets
  */
 @UseFileSplitsFromInputFormat
-public class HoodieInputFormat extends MapredParquetInputFormat implements Configurable {
+public class HoodieParquetInputFormat extends MapredParquetInputFormat implements Configurable {
 
-  private static final transient Logger LOG = LogManager.getLogger(HoodieInputFormat.class);
+  private static final transient Logger LOG = LogManager.getLogger(HoodieParquetInputFormat.class);
 
   protected Configuration conf;
 

hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/hive/HoodieCombineHiveInputFormat.java

@@ -70,8 +70,8 @@ import org.apache.hadoop.mapred.Reporter;
 import org.apache.hadoop.mapred.lib.CombineFileInputFormat;
 import org.apache.hadoop.mapred.lib.CombineFileSplit;
 import org.apache.hadoop.mapreduce.JobContext;
-import org.apache.hudi.hadoop.HoodieInputFormat;
+import org.apache.hudi.hadoop.HoodieParquetInputFormat;
-import org.apache.hudi.hadoop.realtime.HoodieRealtimeInputFormat;
+import org.apache.hudi.hadoop.realtime.HoodieParquetRealtimeInputFormat;
 import org.apache.log4j.LogManager;
 import org.apache.log4j.Logger;
 
@@ -403,9 +403,9 @@ public class HoodieCombineHiveInputFormat<K extends WritableComparable, V extend
       InputFormat inputFormat = getInputFormatFromCache(inputFormatClass, job);
       LOG.info("Input Format => " + inputFormatClass.getName());
       // **MOD** Set the hoodie filter in the combine
-      if (inputFormatClass.getName().equals(HoodieInputFormat.class.getName())) {
+      if (inputFormatClass.getName().equals(HoodieParquetInputFormat.class.getName())) {
         combine.setHoodieFilter(true);
-      } else if (inputFormatClass.getName().equals(HoodieRealtimeInputFormat.class.getName())) {
+      } else if (inputFormatClass.getName().equals(HoodieParquetRealtimeInputFormat.class.getName())) {
         LOG.info("Setting hoodie filter and realtime input format");
         combine.setHoodieFilter(true);
         combine.setRealTime(true);
@@ -857,13 +857,13 @@ public class HoodieCombineHiveInputFormat<K extends WritableComparable, V extend
       LOG.info("Listing status in HoodieCombineHiveInputFormat.HoodieCombineFileInputFormatShim");
       List<FileStatus> result;
       if (hoodieFilter) {
-        HoodieInputFormat input;
+        HoodieParquetInputFormat input;
         if (isRealTime) {
           LOG.info("Using HoodieRealtimeInputFormat");
-          input = new HoodieRealtimeInputFormat();
+          input = new HoodieParquetRealtimeInputFormat();
         } else {
           LOG.info("Using HoodieInputFormat");
-          input = new HoodieInputFormat();
+          input = new HoodieParquetInputFormat();
         }
         input.setConf(job.getConfiguration());
         result = new ArrayList<FileStatus>(

hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/realtime/HoodieParquetRealtimeInputFormat.java

@@ -52,7 +52,7 @@ import org.apache.hudi.common.util.FSUtils;
 import org.apache.hudi.common.util.Option;
 import org.apache.hudi.exception.HoodieException;
 import org.apache.hudi.exception.HoodieIOException;
-import org.apache.hudi.hadoop.HoodieInputFormat;
+import org.apache.hudi.hadoop.HoodieParquetInputFormat;
 import org.apache.hudi.hadoop.UseFileSplitsFromInputFormat;
 import org.apache.log4j.LogManager;
 import org.apache.log4j.Logger;
@@ -61,9 +61,9 @@ import org.apache.log4j.Logger;
  * Input Format, that provides a real-time view of data in a Hoodie dataset
  */
 @UseFileSplitsFromInputFormat
-public class HoodieRealtimeInputFormat extends HoodieInputFormat implements Configurable {
+public class HoodieParquetRealtimeInputFormat extends HoodieParquetInputFormat implements Configurable {
 
-  private static final transient Logger LOG = LogManager.getLogger(HoodieRealtimeInputFormat.class);
+  private static final transient Logger LOG = LogManager.getLogger(HoodieParquetRealtimeInputFormat.class);
 
   // These positions have to be deterministic across all tables
   public static final int HOODIE_COMMIT_TIME_COL_POS = 0;

hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/realtime/RealtimeCompactedRecordReader.java

@@ -80,7 +80,7 @@ class RealtimeCompactedRecordReader extends AbstractRealtimeRecordReader impleme
       // TODO(VC): Right now, we assume all records in log, have a matching base record. (which
       // would be true until we have a way to index logs too)
       // return from delta records map if we have some match.
-      String key = arrayWritable.get()[HoodieRealtimeInputFormat.HOODIE_RECORD_KEY_COL_POS]
+      String key = arrayWritable.get()[HoodieParquetRealtimeInputFormat.HOODIE_RECORD_KEY_COL_POS]
           .toString();
       if (deltaRecordMap.containsKey(key)) {
         // TODO(NA): Invoke preCombine here by converting arrayWritable to Avro. This is required since the

hudi-hadoop-mr/src/test/java/org/apache/hudi/hadoop/AnnotationTest.java

@@ -27,8 +27,8 @@ public class AnnotationTest {
 
   @Test
   public void testAnnotation() {
-    assertTrue(HoodieInputFormat.class.isAnnotationPresent(UseFileSplitsFromInputFormat.class));
+    assertTrue(HoodieParquetInputFormat.class.isAnnotationPresent(UseFileSplitsFromInputFormat.class));
-    Annotation[] annotations = HoodieInputFormat.class.getAnnotations();
+    Annotation[] annotations = HoodieParquetInputFormat.class.getAnnotations();
     boolean found = false;
     for (Annotation annotation : annotations) {
       if ("UseFileSplitsFromInputFormat".equals(annotation.annotationType().getSimpleName())) {

hudi-hadoop-mr/src/test/java/org/apache/hudi/hadoop/HoodieInputFormatTest.java

@@ -38,12 +38,12 @@ import org.junit.rules.TemporaryFolder;
 
 public class HoodieInputFormatTest {
 
-  private HoodieInputFormat inputFormat;
+  private HoodieParquetInputFormat inputFormat;
   private JobConf jobConf;
 
   @Before
   public void setUp() {
-    inputFormat = new HoodieInputFormat();
+    inputFormat = new HoodieParquetInputFormat();
     jobConf = new JobConf();
     inputFormat.setConf(jobConf);
   }

hudi-hive/src/main/java/org/apache/hudi/hive/HiveSyncTool.java

@@ -31,8 +31,8 @@ import org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe;
 import org.apache.hudi.common.util.FSUtils;
 import org.apache.hudi.common.util.Option;
 import org.apache.hudi.exception.InvalidDatasetException;
-import org.apache.hudi.hadoop.HoodieInputFormat;
+import org.apache.hudi.hadoop.HoodieParquetInputFormat;
-import org.apache.hudi.hadoop.realtime.HoodieRealtimeInputFormat;
+import org.apache.hudi.hadoop.realtime.HoodieParquetRealtimeInputFormat;
 import org.apache.hudi.hive.HoodieHiveClient.PartitionEvent;
 import org.apache.hudi.hive.HoodieHiveClient.PartitionEvent.PartitionEventType;
 import org.apache.hudi.hive.util.SchemaUtil;
@@ -128,7 +128,7 @@ public class HiveSyncTool {
         // for now)
         String inputFormatClassName =
             cfg.usePreApacheInputFormat ? com.uber.hoodie.hadoop.HoodieInputFormat.class.getName()
-                : HoodieInputFormat.class.getName();
+                : HoodieParquetInputFormat.class.getName();
         hoodieHiveClient.createTable(schema, inputFormatClassName,
             MapredParquetOutputFormat.class.getName(), ParquetHiveSerDe.class.getName());
       } else {
@@ -137,7 +137,7 @@ public class HiveSyncTool {
         // /ql/exec/DDLTask.java#L3488
         String inputFormatClassName =
             cfg.usePreApacheInputFormat ? com.uber.hoodie.hadoop.realtime.HoodieRealtimeInputFormat.class.getName()
-                : HoodieRealtimeInputFormat.class.getName();
+                : HoodieParquetRealtimeInputFormat.class.getName();
         hoodieHiveClient.createTable(schema, inputFormatClassName,
             MapredParquetOutputFormat.class.getName(), ParquetHiveSerDe.class.getName());
       }