If I have a list like
myList = {"John", "Dan", "Rob", "Rob", "Matt", "Shane", "Matt", "Shane", "John"};
How do I get a list with duplicate entries removed ?
myList_new= {"John", "Dan", "Rob", "Matt", "Shane"};
If it doesn't matter that order changes you can use associative array
Names Default To Here(1); myList = {"John", "Dan", "Rob", "Rob", "Matt", "Shane", "Matt", "Shane", "John"}; my_list = Associative Array(myList) << get keys;
Unfortunately, Associative Arrays get slow if the list is long.
There are other approaches like Summarize and Tables/Summary which are way faster.
So, sometimes it could make sense to create a data table, fill a new column with the list values and use the fast table functions.
Best:go to Provide fast way to get unique values from JMP objects (column, lists), possibly similar to Python's Set and support @jthi 's wish with a Kudo