Address some review comments.

asingh · asingh · commit 903f8e323601 · 2015-06-29T12:38:51.000-07:00
* Ignore stats for writer's version &lt; 1.8.0
* Refactor shoudlIgnoreStatistics method a bit
* Assume implementations other than parquet-mr were writing binary
  statistics correctly
* Add toParquetStatistics method's original method signature to maintain
  backwards compatibility and mark it as deprecated
diff --git a/parquet-avro/src/main/java/org/apache/parquet/avro/AvroWriteSupport.java b/parquet-avro/src/main/java/org/apache/parquet/avro/AvroWriteSupport.java
@@ -276,7 +276,7 @@ private void writeValue(Type type, Schema avroSchema, Object value) {
   private Binary fromAvroString(Object value) {
     if (value instanceof Utf8) {
       Utf8 utf8 = (Utf8) value;
-      return Binary.fromConstantByteArray(utf8.getBytes(), 0, utf8.getByteLength());
+      return Binary.fromReusedByteArray(utf8.getBytes(), 0, utf8.getByteLength());
     }
     return Binary.fromString(value.toString());
   }
diff --git a/parquet-hadoop/src/main/java/org/apache/parquet/format/converter/ParquetMetadataConverter.java b/parquet-hadoop/src/main/java/org/apache/parquet/format/converter/ParquetMetadataConverter.java
@@ -77,7 +77,7 @@ private ParquetMetadataConverter() { }
   public static final MetadataFilter SKIP_ROW_GROUPS = new SkipMetadataFilter();
 
   private static final Log LOG = Log.getLog(ParquetMetadataConverter.class);
-  private static final int STATISTICS_FIXED_VERSION = 161; // 1.6.1
+  private static final int STATISTICS_FIXED_VERSION = 180; // 1.8.0
 
   // NOTE: this cache is for memory savings, not cpu savings, and is used to de-duplicate
   // sets of encodings. It is important that all collections inserted to this cache be
diff --git a/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetFileWriter.java b/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetFileWriter.java
@@ -418,7 +418,7 @@ public void testConvertToThriftStatistics() throws Exception {
       parquetMRstats.updateStats(l);
     }
     final String createdBy =
-        "parquet-mr version 1.7.0rc3 (build d4d5a07ec9bd262ca1e93c309f1d7d4a74ebda4c)";
+        "parquet-mr version 1.8.0rc3 (build d4d5a07ec9bd262ca1e93c309f1d7d4a74ebda4c)";
     Statistics thriftStats =
         org.apache.parquet.format.converter.ParquetMetadataConverter.toParquetStatistics(
             createdBy, parquetMRstats);

Original file line number	Diff line number	Diff line change
`@@ -276,7 +276,7 @@ private void writeValue(Type type, Schema avroSchema, Object value) {`
`276`	`276`	`private Binary fromAvroString(Object value) {`
`277`	`277`	`if (value instanceof Utf8) {`
`278`	`278`	`Utf8 utf8 = (Utf8) value;`
`279`		`- return Binary.fromConstantByteArray(utf8.getBytes(), 0, utf8.getByteLength());`
	`279`	`+ return Binary.fromReusedByteArray(utf8.getBytes(), 0, utf8.getByteLength());`
`280`	`280`	`}`
`281`	`281`	`return Binary.fromString(value.toString());`
`282`	`282`	`}`
Original file line number	Diff line number	Diff line change
`@@ -418,7 +418,7 @@ public void testConvertToThriftStatistics() throws Exception {`
`418`	`418`	`parquetMRstats.updateStats(l);`
`419`	`419`	`}`
`420`	`420`	`final String createdBy =`
`421`		`- "parquet-mr version 1.7.0rc3 (build d4d5a07ec9bd262ca1e93c309f1d7d4a74ebda4c)";`
	`421`	`+ "parquet-mr version 1.8.0rc3 (build d4d5a07ec9bd262ca1e93c309f1d7d4a74ebda4c)";`
`422`	`422`	`Statistics thriftStats =`
`423`	`423`	`org.apache.parquet.format.converter.ParquetMetadataConverter.toParquetStatistics(`
`424`	`424`	`createdBy, parquetMRstats);`