Hi all, I have a physical dataset based on a NFS directory containing csv files.
I would like to filter the rows according to the line number in the files, in order to get all the headers.
I tried something like :
with cte as (
select
to_timestamp(cast($mtime as double)/1000.) as file_timestamp_utc
,$file as file_path
,columns
from table( cbp.chrglperiod.rtjreco."rterecof_v1"( type=>'text' ))
qualify row_number() over(partition by $file) = 1
)
select * from cte
order by file_path
This does not work on a directory based dataset, but works on a file based dataset.
Is there a way to access the line number of a row (like a $rowid column), like we can access the file name ($file) ?
Thanks !