Lahir Marni

[BUG] Loads a Hive Map columns as list when read with read_orc.

10

When the hive stores a columns as a map {key:value}. For example: ``` create table test (id int, mapcol map) ROW FORMAT SERDE 'org.apache.hadoop.hive.ql.io.orc.OrcSerde' STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.orc.OrcInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.orc.OrcOutputFormat'...

[BUG] Create table with Partitioned Columns returns empty Dataframe

4

I am running into a following issue reading partitioned data from local disk. it reads am empty data frame. it is able to resolve the file paths very fine. but...

[BUG] Read ORC fails to read with metadata

2

Read ORC fails to read with metadata. Also, with the addition of read_metadata argument, if i set this parameter to false. create_table works fine. ``` >>> bc.create_table('table','/root/data/*',file_format='orc') /conda/envs/bsql/lib/python3.7/site-packages/distributed/utils.py in sync(loop,...

have come across Grouping Sets in various scenarios where spark code is being converted to blazing. as blazing doesn't support groupby on sets now. the queries cannot be converted.

gcsfs put copies files out of order

2

i am trying to copy multiple files using gcsfs put API. The syntax provided to copy multiple files goes like this. lpath1 is supposed to put the data into rapth1...