Limitations:

• If you generate a new sample after a sort transform has been applied, the sort order is not retained. You can re-apply the sort step, although the following limitations still apply.
• Sort order is not preserved on output when the output is a multi-part file.

Output: Dataset is sorted in alphabetically ascending order based on the values in the LastName column, assuming that the values are strings.

order

Data is sorted based on the data type of the source:   

Example - sort methods

Source:

The column without a name identifies the original row numbers. In the data grid, this information is available when you hover over the black dot to the left of a row of data. 

Transformation:

First, you might want to clean up the number formatting in the lastOrder column. The following formats the values to always include two digits after the decimal point:

Now, you're interested in the highest value for your customers' most recent orders. You can apply the following sort:

Rows are sorted by the LastOrder column in descending order (largest to smallest):

The above row numbers represent the original order of the rows. Now, you want to get your data geographically organized by sorting by city and state. You can perform multi-column sorts such as the following, which sorts first by State and then by City columns:

In the generated output, the data is first sorted by the State value. Each set of rows within the same State value is also sorted by the City value.

Example - Sort by original row numbers