Ayende @ Rahien

My name is Oren Eini
Founder of Hibernating Rhinos LTD and RavenDB.
You can reach me by phone or email:


+972 52-548-6969

, @ Q c

Posts: 6,124 | Comments: 45,475

filter by tags archive

Checking For Empty Enumerations

time to read 3 min | 442 words

Phil Haack has an interesting post about this topic, where he presents the following solution:

public static bool IsNullOrEmpty<T>(this IEnumerable<T> items) {
    return items == null || !items.Any();

This solution, unfortunately, suffers from a common problem related to handling IEnumerables. The assumption that you can iterate over enumerable more than once. This hold true for things like collections, but in many cases, this sort of code will silently hide data:

var files = Directory.EnumerateFiles(".","*.cs");
    Cosnole.WriteLine("No files");
   foreach(var file in files)

The first file will never appear here.

A better solution is:

public static bool IsNullOrEmpty<T>(this IEnumerable<T> items, out IEnumerable<T> newItems) 
    newItems = items;
    if(items == null)
        return false;
    var enumerator = items.GetEnumerator();
    if(enumerator.MoveNext() == false)
        return false;
    newItems = new[]{enumerator.Current}.Concat(enumerator);
    return true;

That will not lose data.



Good! But if we use the better solution, the client code will be like this:

IEnumerable <t items = ...;

if (items.IsNullOrEmpty(items)) {

  // ...


I think the parameter of IsNullOrEmpty() method would be confusing.


But your solution will fail on infinite ranges.

Ayende Rahien

Dennis, why would it fail?


I agree it can be confusing, but that is about the only thing that can work in this situation


Actually it doesn't compile... There isn't a Concat operator for IEnumerators

Tom de Koning

You might want to check up on that again


public class EnumerableTests



    public void will_enumerate_all_items()


        var counter = 0;

        var ints = new[] { 1, 2, 3, 4 }.AsEnumerable();

        if (ints.Any())


            foreach (var i in ints)





        Assert.AreEqual(3, counter);



This will work, but it is vastly more code. (sorry I didnt dump i somewhere)

public class IsNullOrEmptyChecker <t : IEnumerable <t

  private IEnumerator

<t _enumerator;

  private readonly IEnumerable

<t _orig;

  private bool _isEmpty;

  public IsNullOrEmptyChecker(IEnumerable

<t orig)


     _orig = orig;


  private void Init()


     if (_enumerator != null)


     if (_orig != null)


        _enumerator = _orig.GetEnumerator();

        _isEmpty = !_enumerator.MoveNext();




        _isEmpty = true;



  public bool IsNullOrEmpty





        return _isEmpty;



  public IEnumerator

<t GetEnumerator()



     if (_isEmpty)

        yield break;

     yield return _enumerator.Current;

     while (_enumerator.MoveNext())

        yield return _enumerator.Current;

     _enumerator = null;


  IEnumerator IEnumerable.GetEnumerator()


     return GetEnumerator();



Use: var files = new IsNullOrEmptyChecker <int(Directory.EnumerateFiles(".","*.cs"));


Tom, that will fail with this:

  private static bool hest = false;

  public static IEnumerable

<int GetEnumerator()


     if (hest)

        throw new Exception();

     hest = true;

     yield return 1;

     yield return 2;


An enumerable that only allows 1 iteration


"The first file will never appear here."

I'm afraid you didn't try to run your code.

Because it will appear.

So your point is valid only for some really strange implementation of IEnumerable, that change content based on number of iteration of enumerator that it return through call to GetEnumerator.


Ayende, enumerators are IDisposable and should be disposed.

Patrik H&#228;gne

That would be a totally wrong implemented IEnumerable. I think you confusing IEnumerable(T) with IEnumerator(T).


Patrik, below is the structure of IEnumerator <t, courtesy of Reflector:

public interface IEnumerator <t : IDisposable, IEnumerator


// Properties

T Current { get; }



Your example disproves your assertion.

Would be nice to see an actual example where Phil's IsNullOrEmpty() check will lose the first item.

Your assertion seems to be based on the assumption that when a foreach contstruct is called on a IEnumerable collection, it will use an existing enumerator for the colleciton if it is available instead of getting a new one. This assumption doesn't seem right to me.

Patrik H&#228;gne

LL, I wasn't commenting on your comment. I was commenting on Ayendes blog post.

An that IEnumerable would return an enumerator that skips the first element the second time GetEnumerator is called would be totally wrong implemented IEnumerable.

Patrik H&#228;gne

My last comment should read:

An IEnumerable that would return an enumerator that skips the first element the second time GetEnumerator is called would be totally wrong implemented IEnumerable.

Diego Mijelshon


As Patrick noted, separate enumerations of an enumerable should start at the first element again.

In fact, the example you gave (Directory.EnumerateFiles) is wrong: it works as expected when doing .Any() first.


Ayende, did you actually run the code. I think that Directory.EnumerateFiles(".","*.cs"); returns an IEnumerable. The IsNullOrEmpty method will create ea fresh enumerator from it as well as the foreach loop.


But you are right that this will do side-effects twice although only for one element.


I don't like the semantics of this method. Calling a method in an instance fashion to check for null should throw a null pointer exception if it's null.

We don't say "somestring".IsNullOrEmpty, we use String.IsNullOrEmpty(someString)

Seems counter-intuitive.

Marcel Popescu

[OT] Jason, you mean I'm the only one who wrote an IsNullOrEmpty extension method? "string".IsNullOrEmpty() is part of all my projects :)



Oh I've done it. But I decided against it b/c it just seemed so wrong to have a NULL instance return something from a method call rather than throwing and exception.

Patrik H&#228;gne

I totally agree with Jason that the semantics are off, but I guess that's a comment to Phils original post, not to this post.


As many commenters mentioned, the implementation of the Any method gets a new Enumerator which starts at the beginning. So it should be safe to call. I wasn't able to produce your results.


As a side note your implementation is inversed from the name. Or I might be missing something obvious? I. e. the method will return false when the IEnumerable is null or empty which is the opposite of you might expect when calling IsNullOrEmpty().

Bryan Avery

It works fine for me, sounds like brain thinking before testing a solution.

The other issue with this is that you are making a copy of the object, lets just hope the object is not too large...!

Royston Shufflebotham

Oren, if you were talking about IEnumerator <t, you'd be bang on, but with IEnumerable <t you get a brand new IEnumerator <t each time. As long as the underlying data doesn't change, repeated calls to IEnumerable <t.First() would give you the same value each time, as each call gets a new IEnumerator <t.

(With Phil H's original post, we really shouldn't have to be checking for null IEnumerable <t objects - APIs that return null IEnumerables are really broken; you just don't want to have two representations of 'no value', and you can always return Enumerable.Empty <t() if performance is crazily critical.)

Paul Irwin

In your example Ayende, what about IEnumerator.Reset()? Although I agree with previous post about it being IDisposable and should be disposed, which usually would call Reset. Enumerable.Any() in System.Linq does the using pattern to reset the enumerator after it is used.

Luis Abreu

Guys, can't believe this.

X x = null;


what's the expected result? null reference exception. Why are these extension methods behaving differently from instance methods? Why aren't they simply helper methods? There's simply no reasoning for keeping this madness...

Erlis Vidal

Hi Luis,

I totally agree, is damn confussing. It should be a Helper.


Wow, that is some tricky code

Dhananjay Goyani

+1 for Phil. Any() doesn't iterate next onto passed enumerator. Also Oren's solution doesn't seem to be good one.

Lucas O.

Enumerables can be iterated just once.

It is not the same thing use the same enumerator instance that use 2 different instances. Remember that we cannot modify the iterated collection when we use enumerator, in fact if we do that, the next call to MoveNext() method throws an exception.

We can get so many as we want to enumerators but between

its creations, lot of things can happen. In the Oren's example, he use two differents enumerators but between them can be a line deleting all the files. That is the problem.


In following sample booth implementation with self.Any()==false and self.Count() = 0 loses data. Any lose only first element, and Count lose all of it.

    static void Main(string[] args)


        var enumerable = Enumerable();

        if (enumerable.IsNullOrEmpty())


            Console.WriteLine("enumerable is empty");




            foreach (var i in enumerable)





    private static int item;

    private static IEnumerable

<int Enumerable()


        while (item < 10)

            yield return item++;


@hazzik That's just one more way in which the c# implementation of IEnumerable and IEnumerator with the yield-keyword is broken.

Eduard Tom&#224;s


I disagree... the @hazzik problem is about using a static variable, not yield...

Yes you can build non-valid IEnumerables with yield, but that does not mean that yield was broken :)



1) You don't need to check for null - if program fails with that, it means there is an error that requires your immediate attention - so it SHOULD stop execution . Otherwise, you can always catch the resulting exception.

2) Why are there no .IsEmpty member in IEnumerable? Why must we try to re-invent the wheel?



You're right, I didn't read the code in detail. I didn't catch the static variable. This is clearly a case where the implementation of the method that returns the IEnumerable is faulty, not the implementation of yield. It's still broken though in other ways.

Comment preview

Comments have been closed on this topic.


  1. RavenDB 3.5 whirl wind tour: You want all the data, you can’t handle all the data - 10 hours from now
  2. The design of RavenDB 4.0: Making Lucene reliable - about one day from now
  3. RavenDB 3.5 whirl wind tour: I’ll find who is taking my I/O bandwidth and they SHALL pay - 2 days from now
  4. The design of RavenDB 4.0: Physically segregating collections - 3 days from now
  5. RavenDB 3.5 Whirlwind tour: I need to be free to explore my data - 4 days from now

And 14 more posts are pending...

There are posts all the way to May 30, 2016


  1. RavenDB 3.5 whirl wind tour (14):
    29 Apr 2016 - A large cluster goes into a bar and order N^2 drinks
  2. The design of RavenDB 4.0 (13):
    28 Apr 2016 - The implications of the blittable format
  3. Tasks for the new comer (2):
    15 Apr 2016 - Quartz.NET with RavenDB
  4. Code through the looking glass (5):
    18 Mar 2016 - And a linear search to rule them
  5. Find the bug (8):
    29 Feb 2016 - When you can't rely on your own identity
View all series



Main feed Feed Stats
Comments feed   Comments Feed Stats