Replacement idiom for "list_or_nil.to_a"?

ruby-talk
1

Many scripting languages, including Ruby, will "do nothing
gracefully" when asked to iterate through an empty list.
That is, they iterate NO times.

However, many Ruby methods do not return empty lists when
nothing is found. Instead, they return nil. This means
that idioms such as:

  list_or_nil.each do |value|
    ...
  end

will fail. As a workaround, I've been using this idiom:

  list_or_nil.to_a.each do |value|
    ...
  end

However, I gather that this will soon be deprecated:

  http://www.megasolutions.net/ruby/to_a-68350.aspx

This page suggests putting the list in brackets:

  [list_or_nil]

but this doesn't solve MY problem:

  >> list_or_nil=nil
  => nil
  >> [list_or_nil]
  => [nil]

because the loop will now run once with the value nil.
I really dislike having to wrap the iteration block in
an "if" block:

  if list_or_nil
    list_or_nil.each do |value|
      ...
    end
  end

and this is only marginally better:

  list_or_nil.each do |value|
    ...
  end if list_or_nil

I suppose I could define my own to_a method for nil,
but that seems a bit questionable, as well. So, is
there a Better Way To Do It?

-r

···

--
http://www.cfcl.com/rdm Rich Morin
http://www.cfcl.com/rdm/resume rdm@cfcl.com
http://www.cfcl.com/rdm/weblog +1 650-873-7841

Technical editing and writing, programming, and web development

2

  list_or_nil.each do |value|
    ...
  end

If you really want the extra loop in each just to find out that
nothing has to be done, you could also do:

    (list_or_nil || ).each do |value|
        ...
    end

3

Rich Morin wrote:

will fail. As a workaround, I've been using this idiom:

  list_or_nil.to_a.each do |value|
    ...
  end

However, I gather that this will soon be deprecated:

Object#to_a will be deprecated but not nil.to_a, so the
idiom you have above works just fine. It will break if
list_or_nil is a non-array non-nil value but I think
that's as it should be.

Daniel

4

Probably. How do you like

list_or_nil and list_or_nil.each do |x|
   ...
end

Or

irb(main):001:0> class NilClass
irb(main):002:1> def each; self; end
irb(main):003:1> include Enumerable
irb(main):004:1> end
=> NilClass
irb(main):005:0> nil.each {|x| puts x}
=> nil
irb(main):006:0>

?

Kind regards

  robert

···

On 25.05.2007 12:13, Rich Morin wrote:

Many scripting languages, including Ruby, will "do nothing
gracefully" when asked to iterate through an empty list.
That is, they iterate NO times.

However, many Ruby methods do not return empty lists when
nothing is found. Instead, they return nil. This means
that idioms such as:

  list_or_nil.each do |value|
    ...
  end

will fail. As a workaround, I've been using this idiom:

  list_or_nil.to_a.each do |value|
    ...
  end

However, I gather that this will soon be deprecated:

  Interview questions and answers

This page suggests putting the list in brackets:

  [list_or_nil]

but this doesn't solve MY problem:

  >> list_or_nil=nil
  => nil
  >> [list_or_nil]
  => [nil]

because the loop will now run once with the value nil.
I really dislike having to wrap the iteration block in
an "if" block:

  if list_or_nil
    list_or_nil.each do |value|
      ...
    end
  end

and this is only marginally better:

  list_or_nil.each do |value|
    ...
  end if list_or_nil

I suppose I could define my own to_a method for nil,
but that seems a bit questionable, as well. So, is
there a Better Way To Do It?

5

Thanks for the suggestions! Here's a summary, with comments:

foo.to_a.each {|value| ... }

  This is what I was using. It's nice to know that it's not
  going away, after all.

foo.to_a.each {|value| ... } if foo

  This moves the "if" clause away from the iteration, which
  might make it harder to see. Also, it evaluates foo twice
  (could duplicate side-effects if foo is a method or be a
  performance issue id evaluating foo is expensive).

foo and foo.each {|value| ... }

  A bit wordy and obscure. Also, it evaluates foo twice.

class NilClass
  def each; self; end
  include Enumerable
end

foo.each {|value| ... }

  This shortens the usage code, but I don't like playing with
  the definition of NilClass. Specifically, this might confuse
  a human reader or get in the way of code that already expects
  the current behavior.

(foo || ).each {|value| ... }

  This smells like line-noise, but is otherwise quite adequate.

(foo || VOID).each {|value| ... }
(foo || void).each {|value| ... }

  Aside from the line-noise issue, these seem similar to me.
  Both require the use of a supplementary definition, however,
  which the reader might have to loop up.

Array(foo).each {|value| ... }

  This seems like an improvement, from a clarity standpoint.

My preference, at this point, is for either:

foo.to_a.each {|value| ... }

Array(foo).each {|value| ... }

I think I prefer the latter, from a clarity standpoint. Also,
the fact that to_a is being deprecated in other contexts might
confuse some readers.

If foo is a BIG array, there might be performance differences
between these two, but I'm loathe to second-guess interpreter
implementation issues of this sort...

-r

···

--
http://www.cfcl.com/rdm Rich Morin
http://www.cfcl.com/rdm/resume rdm@cfcl.com
http://www.cfcl.com/rdm/weblog +1 650-873-7841

Technical editing and writing, programming, and web development

6

list_or_nil.each do |value| unless list_or_nil.nil?

···

--
Aníbal Rojas

http://www.hasmanydvelopers.com

On May 25, 6:13 am, Rich Morin <r...@cfcl.com> wrote:

Many scripting languages, including Ruby, will "do nothing
gracefully" when asked to iterate through an empty list.
That is, they iterate NO times.

However, many Ruby methods do not return empty lists when
nothing is found. Instead, they return nil. This means
that idioms such as:

  list_or_nil.each do |value|
    ...
  end

will fail. As a workaround, I've been using this idiom:

  list_or_nil.to_a.each do |value|
    ...
  end

However, I gather that this will soon be deprecated:

Interview questions and answers

This page suggests putting the list in brackets:

  [list_or_nil]

but this doesn't solve MY problem:

  >> list_or_nil=nil
  => nil
  >> [list_or_nil]
  => [nil]

because the loop will now run once with the value nil.
I really dislike having to wrap the iteration block in
an "if" block:

  if list_or_nil
    list_or_nil.each do |value|
      ...
    end
  end

and this is only marginally better:

  list_or_nil.each do |value|
    ...
  end if list_or_nil

I suppose I could define my own to_a method for nil,
but that seems a bit questionable, as well. So, is
there a Better Way To Do It?

-r
--http://www.cfcl.com/rdm Rich Morinhttp://www.cfcl.com/rdm/resume r...@cfcl.comhttp://www.cfcl.com/rdm/weblog +1 650-873-7841

Technical editing and writing, programming, and web development

7

That hack will also make nil.to_a => .

···

On 5/25/07, Robert Klemme <shortcutter@googlemail.com> wrote:

irb(main):001:0> class NilClass
irb(main):002:1> def each; self; end
irb(main):003:1> include Enumerable
irb(main):004:1> end
=> NilClass
irb(main):005:0> nil.each {|x| puts x}
=> nil
irb(main):006:0>

--
Chris Carter
concentrationstudios.com
brynmawrcs.com

8

Thanks for the suggestions! Here's a summary, with comments:

foo.to_a.each {|value| ... }

  This is what I was using. It's nice to know that it's not
  going away, after all.

foo.to_a.each {|value| ... } if foo

  This moves the "if" clause away from the iteration, which
  might make it harder to see. Also, it evaluates foo twice
  (could duplicate side-effects if foo is a method or be a
  performance issue id evaluating foo is expensive).

foo and foo.each {|value| ... }

  A bit wordy and obscure. Also, it evaluates foo twice.

This is an issue only if foo is a method that does a complex calculation.

class NilClass
  def each; self; end
  include Enumerable
end

foo.each {|value| ... }

  This shortens the usage code, but I don't like playing with
  the definition of NilClass. Specifically, this might confuse
  a human reader or get in the way of code that already expects
  the current behavior.

(foo || ).each {|value| ... }

  This smells like line-noise, but is otherwise quite adequate.

(foo || VOID).each {|value| ... }
(foo || void).each {|value| ... }

  Aside from the line-noise issue, these seem similar to me.
  Both require the use of a supplementary definition, however,
  which the reader might have to loop up.

Array(foo).each {|value| ... }

  This seems like an improvement, from a clarity standpoint.

My preference, at this point, is for either:

foo.to_a.each {|value| ... }

This might create another object which might impose a performance hit.

Array(foo).each {|value| ... }

I think I prefer the latter, from a clarity standpoint. Also,
the fact that to_a is being deprecated in other contexts might
confuse some readers.

If foo is a BIG array, there might be performance differences
between these two, but I'm loathe to second-guess interpreter
implementation issues of this sort...

There will likely be a performance hit if foo is something different than an array because then it will have to be converted to an array.

Kind regards

  robert

···

On 25.05.2007 18:23, Rich Morin wrote:

9

Thanks for the suggestions! Here's a summary, with comments:

> foo.to_a.each {|value| ... }

  This is what I was using. It's nice to know that it's not
  going away, after all.

> foo.to_a.each {|value| ... } if foo

  This moves the "if" clause away from the iteration, which
  might make it harder to see. Also, it evaluates foo twice
  (could duplicate side-effects if foo is a method or be a
  performance issue id evaluating foo is expensive).

> foo and foo.each {|value| ... }

  A bit wordy and obscure. Also, it evaluates foo twice.

> class NilClass
> def each; self; end
> include Enumerable
> end

> foo.each {|value| ... }

  This shortens the usage code, but I don't like playing with
  the definition of NilClass. Specifically, this might confuse
  a human reader or get in the way of code that already expects
  the current behavior.

> (foo || ).each {|value| ... }

A very common idiom in Ruby however.

  This smells like line-noise, but is otherwise quite adequate.

> (foo || VOID).each {|value| ... }
> (foo || void).each {|value| ... }

  Aside from the line-noise issue, these seem similar to me.
  Both require the use of a supplementary definition, however,
  which the reader might have to loop up.

This is a hypothetical. The significant advantage of "void" would come
from a core implementation which would be much more efficient than .
(A "void" object would also have some other good uses however. hint
hint matz :wink:

T.

···

On May 25, 12:23 pm, Rich Morin <r...@cfcl.com> wrote:

10

If the code will be reusable, that's a little too risky.

It would be interesting however for an object to exist, say Void.
Something like:

  class Void
    def method_missing(*args)
      void
    end
  end

  def void
    $void ||= Void.new
  end

  (list_or_nil || void).each { |x| p x }

Though, a core implementation of Void would be much more efficient b/c
it would not need to even bother with method_missing, and just know to
return void and move on.

Hmmm... actually this has been discussed in another context before
(Hash lookup) as Null. Looks like another good reason for it.

T.

···

On May 25, 8:05 am, "Chris Carter" <cdcar...@gmail.com> wrote:

On 5/25/07, Robert Klemme <shortcut...@googlemail.com> wrote:

> irb(main):001:0> class NilClass
> irb(main):002:1> def each; self; end
> irb(main):003:1> include Enumerable
> irb(main):004:1> end
> => NilClass
> irb(main):005:0> nil.each {|x| puts x}
> => nil
> irb(main):006:0>

That hack will also make nil.to_a => .