[SPARK-2195] Parquet extraMetadata can contain key information

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 1.0.0
• Fix Version/s: 1.0.1, 1.1.0
• Component/s: SQL
• Target Version/s: 1.1.0
• Assignee: Unassigned
• Reporter: Michael Armbrust
• Created: 19/Jun/14 09:02
• Updated: 21/Jun/14 18:28
• Resolved: 21/Jun/14 18:28

Description

14/06/19 01:52:05 INFO NewHadoopRDD: Input split: ParquetInputSplit{part: file:/Users/pat/Projects/spark-summit-training-2014/usb/data/wiki-parquet/part-r-1.parquet start: 0 length: 24971040 hosts: [localhost] blocks: 1 requestedSchema: same as file fileSchema: message root {

  optional int32 id;

  optional binary title;

  optional int64 modified;

  optional binary text;

  optional binary username;

}

 extraMetadata: {org.apache.spark.sql.parquet.row.metadata=StructType(List(StructField(id,IntegerType,true), StructField(title,StringType,true), StructField(modified,LongType,true), StructField(text,StringType,true), StructField(username,StringType,true))), path=************************ MY AWS KEYS!!! ************************} readSupportMetadata: {org.apache.spark.sql.parquet.row.metadata=StructType(List(StructField(id,IntegerType,true), StructField(title,StringType,true), StructField(modified,LongType,true), StructField(text,StringType,true), StructField(username,StringType,true))), path=**************************** MY AWS KEYS!!!! ***********************************}}