# Indices of non nil elements in an array **URL:** \ **Category:** ruby-talk\ **Created:** [25 August 2008 01:20 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584 "2008-08-25T01:20:41Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![Patrick\_Doyle](https://avatars.discourse-cdn.com/v4/letter/p/22d042/32.png) [@Patrick\_Doyle](https://rubytalk.org/u/Patrick_Doyle)\ **Post date:** [25 August 2008 01:20 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/1 "2008-08-25T01:20:41Z")
a = [] a[10] = "foo" a.nitems returns a count of the non-nil, items. What is the simplest way to return the indices of "a" that are not nil? I'm sure there is some simple elegant way to do this, but I can't figure it out. --wpd ---
**Author:** ![Christopher\_Dicely](https://avatars.discourse-cdn.com/v4/letter/c/7c8e57/32.png) [@Christopher\_Dicely](https://rubytalk.org/u/Christopher_Dicely)\ **Post date:** [25 August 2008 02:14 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/2 "2008-08-25T02:14:16Z")
If you need this frequently, you can monkeypatch Array to offer this functionality; like Array#detect but gathers the indexes instead of the values, and gathers indexes of non-nil values where no block is given: class Array   def detect\_indices     result =     each\_with\_index { |val, idx| (block\_given? ? yield(val) : (not val.nil?)) && result \<\< idx }     result   end end > **···** > > On Sun, Aug 24, 2008 at 6:20 PM, Patrick Doyle \ wrote: > > > a = > > a[10] = "foo" > > > > a.nitems returns a count of the non-nil, items. What is the simplest > > way to return the indices of "a" that are not nil? > > > > I'm sure there is some simple elegant way to do this, but I can't figure it out. ---
**Author:** ![Xavier\_Noria](https://yyz1.discourse-cdn.com/flex029/user_avatar/rubytalk.org/xavier_noria/32/1867_2.png) [@Xavier\_Noria](https://rubytalk.org/u/Xavier_Noria)\ **Post date:** [25 August 2008 11:38 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/3 "2008-08-25T11:38:01Z")
Not exactly what you ask, but in case it's actually what you are after have a look at Array#compact. > **···** > > On Mon, Aug 25, 2008 at 3:20 AM, Patrick Doyle \ wrote: > > > a = > > a[10] = "foo" > > > > a.nitems returns a count of the non-nil, items. What is the simplest > > way to return the indices of "a" that are not nil? ---
**Author:** ![David\_A\_Black1](https://avatars.discourse-cdn.com/v4/letter/d/bcef8e/32.png) [@David\_A\_Black1](https://rubytalk.org/u/David_A_Black1)\ **Post date:** [25 August 2008 11:54 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/4 "2008-08-25T11:54:56Z")
Hi -- > **···** > > On Mon, 25 Aug 2008, Patrick Doyle wrote: > > > a = > > a[10] = "foo" > > > > a.nitems returns a count of the non-nil, items. What is the simplest > > way to return the indices of "a" that are not nil? > > > > I'm sure there is some simple elegant way to do this, but I can't figure it out. > > Try this: > > (0...a.size).reject {|i| a[i].nil? } > > or select for a[i] if you don't mind losing false as well as nil. > > David > > -- > Rails training from David A. Black and Ruby Power and Light: >    Intro to Ruby on Rails January 12-15 Fort Lauderdale, FL >    Advancing with Rails January 19-22 Fort Lauderdale, FL > See [http://www.rubypal.com](http://www.rubypal.com) for details and updates! ---
**Author:** ![Patrick\_Doyle](https://avatars.discourse-cdn.com/v4/letter/p/22d042/32.png) [@Patrick\_Doyle](https://rubytalk.org/u/Patrick_Doyle)\ **Post date:** [25 August 2008 11:34 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/5 "2008-08-25T11:34:31Z")
Thanks. That looks workable. What I could also do (I decided in the mean time) is to change my array into a hash whose keys happen to be integers. That way I could look at things like a.size and a.keys.sort. --wpd ---
**Author:** ![Patrick\_Doyle](https://avatars.discourse-cdn.com/v4/letter/p/22d042/32.png) [@Patrick\_Doyle](https://rubytalk.org/u/Patrick_Doyle)\ **Post date:** [25 August 2008 12:02 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/6 "2008-08-25T12:02:00Z")
Thanks. I tried that and realized that I lost all of the information about the indices in so doing. As I've thought about this some more and have realized that the index of the non-nil elements is just as important to me as the non-nil elements themselves, I've decided that using a Hash is probably a better solution to my needs than using an Array. --wpd ---
**Author:** ![Patrick\_Doyle](https://avatars.discourse-cdn.com/v4/letter/p/22d042/32.png) [@Patrick\_Doyle](https://rubytalk.org/u/Patrick_Doyle)\ **Post date:** [25 August 2008 12:04 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/7 "2008-08-25T12:04:12Z")
Ahhh -- that's the simple elegant way for which I was searching. I kept going down the path of a.each\_index and didn't like where that was going. --wpd > **···** > > On Mon, Aug 25, 2008 at 7:54 AM, David A. Black \ wrote: > > > On Mon, 25 Aug 2008, Patrick Doyle wrote: > > > > > I'm sure there is some simple elegant way to do this, but I can't figure > > > it out. > > > > Try this: > > > > (0...a.size).reject {|i| a[i].nil? } ---
**Author:** ![Dave\_Goodchild](https://avatars.discourse-cdn.com/v4/letter/d/d9b06d/32.png) [@Dave\_Goodchild](https://rubytalk.org/u/Dave_Goodchild)\ **Post date:** [25 August 2008 12:22 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/8 "2008-08-25T12:22:09Z")
That can be made even more elegant: a.reject (/e/ e.nil?) Apologies I am on an iPod so the brackets should be braces and the slashes pipes! > **···** > > Sent from my iPod > > On Aug 25, 2008, at 1:04 PM, "Patrick Doyle" \ wrote: > > > On Mon, Aug 25, 2008 at 7:54 AM, David A. Black \ \> wrote: > > > > > On Mon, 25 Aug 2008, Patrick Doyle wrote: > > > > > > > I'm sure there is some simple elegant way to do this, but I can't figure > > > > it out. > > > > > > Try this: > > > > > > (0...a.size).reject {|i| a[i].nil? } > > > > Ahhh -- that's the simple elegant way for which I was searching. I > > kept going down the path of a.each\_index and didn't like where that > > was going. > > > > --wpd ---
**Author:** ![Patrick\_Doyle](https://avatars.discourse-cdn.com/v4/letter/p/22d042/32.png) [@Patrick\_Doyle](https://rubytalk.org/u/Patrick_Doyle)\ **Post date:** [25 August 2008 12:38 UTC](https://rubytalk.org/t/indices-of-non-nil-elements-in-an-array/48584/9 "2008-08-25T12:38:25Z")
But that won't return the indices of the the non-nil elements, instead it extracts the non-nil elements from the array (which is the same thing that Array#compact) does. --wpd > **···** > > On Mon, Aug 25, 2008 at 8:22 AM, \ wrote: > > > That can be made even more elegant: > > > > a.reject (/e/ e.nil?)