wide tables data overlap previously written row data in the CLI

when querying a parquet or csv file with 18 columns the columns from the end of the row overwrite the data at the beginning of the row. You can use the below data and query it and notice that the 01001 will not show up in the output results.

repeat with storage format:
"nndb":

{ "type": "text", "extensions": [ "txt" ], "quote": "~", "escape": "~", "delimiter": "^" }

,

Data records in file NUT_DATA.txt:
01001203^0.85^16^0.0741~^^^^^~^11/1976
01001204^81.11^580^0.0651~^^^^^~^11/1976
01001205^0.06^04NC~^^^^^^~^11/1976
01001207^2.11^35^0.0541~^^^^^~^11/1976
01001208^717^04NC~^^^^^^~^08/2010
01001221^0.0^07~^^^^^^~^04/1985
01001255^15.87^522^0.0611~^^^^^~^11/1976
01001262^0^07Z~^^^^^^~^02/2001
01001263^0^07Z~^^^^^^~^02/2001
01001268^3000^04~^^^^^^~^06/2013

select columns[0], columns[1], columns[2], columns[3], columns[4], columns[5], columns[6], columns[7], columns[8], columns[9], columns[10], columns[11], columns[12], columns[13], columns[14], columns[15], columns[16], columns[17] from dfs.tmp.`sr27asc/NUT_DATA.txt`