Hi,
I got it. Maybe a better solution exists, but this one works fine.
1st, add a column (rank for example) representing the row number in the tFileExcelInput; result buffered into tHashOutput_1
2nd, use the tMap to join tHashOutput_1 with itself, filtering on rank
3rd, eliminate duplicates using tUniqRow
Must be validated with more complex data sample.
That's all.
Ragards,
TRF