Clone Tools
  • last updated a few minutes ago
Constraints: committers
Constraints: files
Constraints: dates
DRILL-1700: Removed incorrect assertion in memory allocation code

DRILL-1404: Queries with empty results are throwing "HTTP ERROR 500" from Web UI

DRILL-1664: Count(*) returns incorrect result for a parquet file with repeated data.

DRILL-1701: Fix for nullable dictionary columns in optimized parquet reader. Re-enable the optimized reader for dictionary encoded files.

DRILL-1705: Fix bug when loading nested repeated vectors

DRILL-1702: Fix issue where 4 byte DrillBuf is sent even though there are no values when using varchar or repeated vector.

DRILL-1703: Fix situation where repeated map vector was re-adding column name on second record batch.

DRILL-1610: Don't apply partition pruning optimization if one of the disjuncts don't meet pruning criteria

DRILL-1704: Use complex reader for dictionary encoded files, as original reader seems to be broken

DRILL-1376 ORDER BY clause ignored in query run through web UI - Bug fixed

DRILL-1660: return accurate buffers size when repeated map vector has no underlying vectors

DRILL-1698: QueryWrapper creates a logger for FlattenRecordBatch class

DRILL-1535:Logical plan deserialization error in Join

DRILL-1695: Fix code generator comparisons

Drill-1688: Complex parquet reader fails to read wide records.

DRILL-1660: adding a test case to cover an array of empty and null valued objects. i) remove redundant guard statement from JsonRecordReader2 ii) rename unit test suite

DRILL-1559: increment counts even if we break out of the read loop

Add handling when a null exception is logged.

DRILL-1659: Increase default maxWidth for sqlline to 10k

DRILL-1671, DRILL-1653, DRILL-1652: Fixes for flatten bugs

    • -0
    • +7
    • -0
    • +203
    • -0
    • +11
DRILL-1638: display list of files in the logical plan

DRILL-1568: C++ Client - Handle Query Cancel

DRILL-1656: Fix wildcard queries to correctly determine selection root.

DRILL-1774: Update JSON Reader to do single pass reading and better use Jackson's interning. Also improve projection pushdown support.

  1. … 15 more files in changeset.
Remove extraneous System.out.print statements.

DRILL-1434: In ParquetGroupScan compute the non-null value count of a column if stats are available for each chunk. Don't apply ConvertCountToDirectScan rule if stats are not available.

DRILL-1591, DRILL-1676: Move javascript resources to local serving and update dagre-d3 to older version (2.9). Update profile page. Remove references to invalid servlet api.

    • binary
    • -0
    • +9215
    • -0
    • +313
  1. … 9 more files in changeset.
DRILL-1684, DRILL-1517, DRILL-1350: Profile and cancellation updates - Remove any storage of persisted profiles. - Store a separate query info object for active queries. - Update cancellation and running profile loading to query foreman server. - Make file store support HDFS APIs - Update PStoreProvider to use configuration to decide if you want PERSISTENT, EPHEMERAL, or BLOB storage rather than separate interfaces. - Update ZkPStore's persistent mode to leverage a cache and respond to changes rather than actively probing values. - Update ZkPStore's cache to be effectively write-through. - Automatically delete deprecated or default value options from PStore.

  1. … 28 more files in changeset.
DRILL-1455: enable projection pushdown past join

i) refactor project past filter to remove redundant code

ii) adding unit tests for projection past join & hybrid cases like projection past filter & join

iii) disable one TestExampleQueries#testSelectStartSubQueryJoinWithWhereClause until DRILL-1680 is fixed.

DRILL-1663: Fix casting to a variable width type implicitly in a join condition

Factor out logic in ExpressionTreeMaterializer so the same logic can be reused in ChainedHashTable.