uniquesa#
Purpose#
Removes duplicate elements from a string vector.
Note
uniquesa()
is deprecated. Instances of this function should be replaced by unique()
.
Format#
- y = uniquesa(sv)#
- Parameters:
sv (Nx1 or 1xN string vector) – data
- Returns:
y (sorted Mx1 string vector) – containing all unique elements found in sv.
Examples#
// Create a 8x1 string array
string comTrades = { "corn", "gold", "soybeans", "silver", "coffee",
"oil", "silver", "soybeans" };
// Return an alphabetized string array containing the
// unique elements from 'comTrades'
commodity = uniquesa(comTrades);
After the code above, the variables comTrades
and commodity
will be equal to:
corn
gold coffee
soybeans corn
comTrades = silver commodity = gold
coffee oil
oil silver
silver soybeans
soybeans
Remarks#
It is important to note that the return from uniquesa()
will always be a
column vector, even if the input string array is a row vector.
Source#
uniquesa.src
See also
Functions unique()
, uniqindxsa()
, uniqindx()