Question

Scenario

I have a database consisting of a table called Problems which in turn consists of 2 columns:

• ID
• Problem

I populated the tables with millions of records.

In my application, I want to retrive a set of N distinct records taken at random to make a problem sheet. A problem sheet consist of N problems. The randomness should be good enough to make sure I don’t produce an problem sheet that is similar to the previous examination.

What is the simplest way to do so using Linq?

Answer 1

Something like:

IQueryable<Problem> problems = db.Problems;
int count = problems.Count();
List<Problem> sheet = new List<Problem>(n);
Random rand = new Random();
while(sheet.Count < n) {
    var next = problems.OrderBy(p => p.ID).Skip(rand.Next(count))
        .FirstOrDefault();
    if(next != null && !sheet.Any(p => p.ID == next.ID) {
        sheet.Add(next);
    }
}

?