Distinct

Summary#

.Distinct() removes duplicates from a result. It takes no arguments — "distinct" means the whole row, exactly as in SQL.

Signature#

<query>.Distinct().ToList()

Description#

On whole rows it does nothing#

Entity rows are already unique — each has its own Id — so Order.Where(…).Distinct() cannot remove anything. It is harmless, and it is also pointless, and writing it usually means someone expected it to do something it does not:

entity Order {
  [Required] string Code;
  [MaxLength(100)] string Region;
  decimal Total;
}

Order[] All() {
  return Order.Where(o => o.Total > 0).Distinct().ToList();   // same rows, either way
}

If what you meant was "one order per region", that is not Distinct — it is a grouping, or a projection of the region alone.

When does Distinct actually remove something?#

Duplicates are real the moment you stop selecting whole rows. Two orders from the same region project to the same region string — and that is where Distinct does the work you wanted:

int RegionCount() {
  return Order.Where(o => o.Total > 0).Distinct().ToList().Count;
}

What does Distinct cost?#

De-duplicating means the database must compare rows, which usually means sorting them. On a large result that is real work. If you are reaching for Distinct to paper over a join that is producing duplicates, fix the join — the duplicates are a symptom, and Distinct only hides it.

See also#

• Where / Single / Count — the query being de-duplicated
• ToList — materialising the result
• Union / Concat / Intersect / Except — union / intersect / except, which also concern duplicates