At times when using Magic ETL it is useful to be able to show a distinct count for a column in a DataSet; however, this action is not currently available in ETL's pool of tiles. Luckily you can accomplish this by doing a small amount of data manipulation. All you need to do is pull out the column to be counted, remove duplicates, and do a GROUP BY, and a distinct count of the column will be shown.
To show the distinct count for a column in ETL,
Use the Select Columns tile to select any columns that need a distinct count or can be used to group the other columns.
Use the Remove Duplicates tile to reduce the data down to unique rows of data.
Use the Group By tile. Identify which column you will be grouping by (i.e. whatever would be in the GROUP BY statement within a MySQL DataFlow) and what columns will be aggregated, then select Number of Values for the aggregation.
Join this data back onto the final transform of the original DataFlow.
This will result in a unique count for each row of data. You can then use a MIN() or MAX() aggregation to get the actual count.