diff --git ql/src/java/org/apache/hadoop/hive/ql/hooks/PostExecOrcFileDump.java ql/src/java/org/apache/hadoop/hive/ql/hooks/PostExecOrcFileDump.java index 87c3db2..ecda606 100644 --- ql/src/java/org/apache/hadoop/hive/ql/hooks/PostExecOrcFileDump.java +++ ql/src/java/org/apache/hadoop/hive/ql/hooks/PostExecOrcFileDump.java @@ -19,6 +19,7 @@ import java.io.IOException; import java.io.PrintStream; +import java.util.Collections; import java.util.List; import org.slf4j.Logger; @@ -101,6 +102,8 @@ List fileList = HdfsUtils.listLocatedStatus(fs, dir, hiddenFileFilter); + Collections.sort(fileList); + for (FileStatus fileStatus : fileList) { if (fileStatus.isDirectory()) { diff --git ql/src/test/results/clientpositive/llap/acid_bloom_filter_orc_file_dump.q.out ql/src/test/results/clientpositive/llap/acid_bloom_filter_orc_file_dump.q.out index da805b0..28fccd6 100644 --- ql/src/test/results/clientpositive/llap/acid_bloom_filter_orc_file_dump.q.out +++ ql/src/test/results/clientpositive/llap/acid_bloom_filter_orc_file_dump.q.out @@ -87,123 +87,6 @@ Stripe 1: Column 0: count: 1 hasNull: false Column 1: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 - Column 2: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 - Column 3: count: 1 hasNull: false bytesOnDisk: 9 min: 536870912 max: 536870912 sum: 536870912 - Column 4: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 - Column 5: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 - Column 6: count: 1 hasNull: false - Column 7: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 - Column 8: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 - Column 9: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 - Column 10: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 - -File Statistics: - Column 0: count: 1 hasNull: false - Column 1: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 - Column 2: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 - Column 3: count: 1 hasNull: false bytesOnDisk: 9 min: 536870912 max: 536870912 sum: 536870912 - Column 4: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 - Column 5: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 - Column 6: count: 1 hasNull: false - Column 7: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 - Column 8: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 - Column 9: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 - Column 10: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 - -Stripes: - Stripe: offset: 3 data: 73 rows: 1 tail: 103 index: 595 - Stream: column 0 section ROW_INDEX start: 3 length 11 - Stream: column 1 section ROW_INDEX start: 14 length 24 - Stream: column 2 section ROW_INDEX start: 38 length 24 - Stream: column 3 section ROW_INDEX start: 62 length 29 - Stream: column 4 section ROW_INDEX start: 91 length 24 - Stream: column 5 section ROW_INDEX start: 115 length 24 - Stream: column 6 section ROW_INDEX start: 139 length 11 - Stream: column 7 section ROW_INDEX start: 150 length 30 - Stream: column 7 section BLOOM_FILTER_UTF8 start: 180 length 112 - Stream: column 8 section ROW_INDEX start: 292 length 30 - Stream: column 8 section BLOOM_FILTER_UTF8 start: 322 length 112 - Stream: column 9 section ROW_INDEX start: 434 length 27 - Stream: column 10 section ROW_INDEX start: 461 length 27 - Stream: column 10 section BLOOM_FILTER_UTF8 start: 488 length 110 - Stream: column 1 section DATA start: 598 length 6 - Stream: column 2 section DATA start: 604 length 6 - Stream: column 3 section DATA start: 610 length 9 - Stream: column 4 section DATA start: 619 length 6 - Stream: column 5 section DATA start: 625 length 6 - Stream: column 7 section DATA start: 631 length 7 - Stream: column 7 section LENGTH start: 638 length 6 - Stream: column 8 section DATA start: 644 length 7 - Stream: column 8 section LENGTH start: 651 length 6 - Stream: column 9 section DATA start: 657 length 7 - Stream: column 10 section DATA start: 664 length 7 - Encoding column 0: DIRECT - Encoding column 1: DIRECT_V2 - Encoding column 2: DIRECT_V2 - Encoding column 3: DIRECT_V2 - Encoding column 4: DIRECT_V2 - Encoding column 5: DIRECT_V2 - Encoding column 6: DIRECT - Encoding column 7: DIRECT_V2 - Encoding column 8: DIRECT_V2 - Encoding column 9: DIRECT_V2 - Encoding column 10: DIRECT_V2 - Row group indices for column 0: - Entry 0: count: 1 hasNull: false positions: - Row group indices for column 1: - Entry 0: count: 1 hasNull: false min: 0 max: 0 sum: 0 positions: 0,0,0 - Row group indices for column 2: - Entry 0: count: 1 hasNull: false min: 2 max: 2 sum: 2 positions: 0,0,0 - Row group indices for column 3: - Entry 0: count: 1 hasNull: false min: 536870912 max: 536870912 sum: 536870912 positions: 0,0,0 - Row group indices for column 4: - Entry 0: count: 1 hasNull: false min: 0 max: 0 sum: 0 positions: 0,0,0 - Row group indices for column 5: - Entry 0: count: 1 hasNull: false min: 2 max: 2 sum: 2 positions: 0,0,0 - Row group indices for column 6: - Entry 0: count: 1 hasNull: false positions: - Row group indices for column 7: - Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 4 positions: 0,0,0,0,0 - Bloom filters for column 7: - Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - Row group indices for column 8: - Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 4 positions: 0,0,0,0,0 - Bloom filters for column 8: - Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - Row group indices for column 9: - Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 2345 positions: 0,0,0 - Row group indices for column 10: - Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 2345 positions: 0,0,0 - Bloom filters for column 10: - Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 - -File length: 1205 bytes -Padding length: 0 bytes -Padding ratio: 0% - -User Metadata: - hive.acid.key.index=2,536870912,0; - hive.acid.stats=1,0,0 - hive.acid.version=2 -________________________________________________________________________________________________________________________ - --- END ORC FILE DUMP -- --- BEGIN ORC FILE DUMP -- -#### A masked pattern was here #### -File Version: 0.12 with ORC_517 -Rows: 1 -Compression: ZLIB -Compression size: 32768 -Calendar: Julian/Gregorian -Type: struct> - -Stripe Statistics: - Stripe 1: - Column 0: count: 1 hasNull: false - Column 1: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 Column 2: count: 1 hasNull: false bytesOnDisk: 6 min: 1 max: 1 sum: 1 Column 3: count: 1 hasNull: false bytesOnDisk: 9 min: 536870912 max: 536870912 sum: 536870912 Column 4: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 @@ -308,4 +191,121 @@ ________________________________________________________________________________________________________________________ -- END ORC FILE DUMP -- +-- BEGIN ORC FILE DUMP -- +#### A masked pattern was here #### +File Version: 0.12 with ORC_517 +Rows: 1 +Compression: ZLIB +Compression size: 32768 +Calendar: Julian/Gregorian +Type: struct> + +Stripe Statistics: + Stripe 1: + Column 0: count: 1 hasNull: false + Column 1: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 + Column 2: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 + Column 3: count: 1 hasNull: false bytesOnDisk: 9 min: 536870912 max: 536870912 sum: 536870912 + Column 4: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 + Column 5: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 + Column 6: count: 1 hasNull: false + Column 7: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 + Column 8: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 + Column 9: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 + Column 10: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 + +File Statistics: + Column 0: count: 1 hasNull: false + Column 1: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 + Column 2: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 + Column 3: count: 1 hasNull: false bytesOnDisk: 9 min: 536870912 max: 536870912 sum: 536870912 + Column 4: count: 1 hasNull: false bytesOnDisk: 6 min: 0 max: 0 sum: 0 + Column 5: count: 1 hasNull: false bytesOnDisk: 6 min: 2 max: 2 sum: 2 + Column 6: count: 1 hasNull: false + Column 7: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 + Column 8: count: 1 hasNull: false bytesOnDisk: 13 min: 2345 max: 2345 sum: 4 + Column 9: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 + Column 10: count: 1 hasNull: false bytesOnDisk: 7 min: 2345 max: 2345 sum: 2345 + +Stripes: + Stripe: offset: 3 data: 73 rows: 1 tail: 103 index: 595 + Stream: column 0 section ROW_INDEX start: 3 length 11 + Stream: column 1 section ROW_INDEX start: 14 length 24 + Stream: column 2 section ROW_INDEX start: 38 length 24 + Stream: column 3 section ROW_INDEX start: 62 length 29 + Stream: column 4 section ROW_INDEX start: 91 length 24 + Stream: column 5 section ROW_INDEX start: 115 length 24 + Stream: column 6 section ROW_INDEX start: 139 length 11 + Stream: column 7 section ROW_INDEX start: 150 length 30 + Stream: column 7 section BLOOM_FILTER_UTF8 start: 180 length 112 + Stream: column 8 section ROW_INDEX start: 292 length 30 + Stream: column 8 section BLOOM_FILTER_UTF8 start: 322 length 112 + Stream: column 9 section ROW_INDEX start: 434 length 27 + Stream: column 10 section ROW_INDEX start: 461 length 27 + Stream: column 10 section BLOOM_FILTER_UTF8 start: 488 length 110 + Stream: column 1 section DATA start: 598 length 6 + Stream: column 2 section DATA start: 604 length 6 + Stream: column 3 section DATA start: 610 length 9 + Stream: column 4 section DATA start: 619 length 6 + Stream: column 5 section DATA start: 625 length 6 + Stream: column 7 section DATA start: 631 length 7 + Stream: column 7 section LENGTH start: 638 length 6 + Stream: column 8 section DATA start: 644 length 7 + Stream: column 8 section LENGTH start: 651 length 6 + Stream: column 9 section DATA start: 657 length 7 + Stream: column 10 section DATA start: 664 length 7 + Encoding column 0: DIRECT + Encoding column 1: DIRECT_V2 + Encoding column 2: DIRECT_V2 + Encoding column 3: DIRECT_V2 + Encoding column 4: DIRECT_V2 + Encoding column 5: DIRECT_V2 + Encoding column 6: DIRECT + Encoding column 7: DIRECT_V2 + Encoding column 8: DIRECT_V2 + Encoding column 9: DIRECT_V2 + Encoding column 10: DIRECT_V2 + Row group indices for column 0: + Entry 0: count: 1 hasNull: false positions: + Row group indices for column 1: + Entry 0: count: 1 hasNull: false min: 0 max: 0 sum: 0 positions: 0,0,0 + Row group indices for column 2: + Entry 0: count: 1 hasNull: false min: 2 max: 2 sum: 2 positions: 0,0,0 + Row group indices for column 3: + Entry 0: count: 1 hasNull: false min: 536870912 max: 536870912 sum: 536870912 positions: 0,0,0 + Row group indices for column 4: + Entry 0: count: 1 hasNull: false min: 0 max: 0 sum: 0 positions: 0,0,0 + Row group indices for column 5: + Entry 0: count: 1 hasNull: false min: 2 max: 2 sum: 2 positions: 0,0,0 + Row group indices for column 6: + Entry 0: count: 1 hasNull: false positions: + Row group indices for column 7: + Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 4 positions: 0,0,0,0,0 + Bloom filters for column 7: + Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + Row group indices for column 8: + Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 4 positions: 0,0,0,0,0 + Bloom filters for column 8: + Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + Row group indices for column 9: + Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 2345 positions: 0,0,0 + Row group indices for column 10: + Entry 0: count: 1 hasNull: false min: 2345 max: 2345 sum: 2345 positions: 0,0,0 + Bloom filters for column 10: + Entry 0: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + Stripe level merge: numHashFunctions: 6 bitCount: 81472 popCount: 6 loadFactor: 0.0001 expectedFpp: 1.5953551E-25 + +File length: 1205 bytes +Padding length: 0 bytes +Padding ratio: 0% + +User Metadata: + hive.acid.key.index=2,536870912,0; + hive.acid.stats=1,0,0 + hive.acid.version=2 +________________________________________________________________________________________________________________________ + +-- END ORC FILE DUMP -- 12345 12345 12345 12345