# | versus || Any reason for this issue? **URL:** \ **Category:** Ruby\ **Created:** [May 2, 2008, 4:02pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485 "2008-05-02T16:02:18Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:02pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/1 "2008-05-02T16:02:18Z")
Hi, why in the second case the non declared “my\_var” variable is processed but not in the first case? irb\> nil || 2 || my\_var =\> 2 irb\> nil | 2 | my\_var NameError: undefined local variable or method `my\_var’ Of course, I prefer the first case: 2 is already true so there is no reason to read the next “my\_var”. Also, it’s not good for performance, is it? ---
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:14pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/2 "2008-05-02T16:14:42Z")
Hi – On Fri, 2 May 2008, Iñaki Baz C. wrote: > Of course, I prefer the first case: 2 is already true so there is no > reason to read the next “my\_var”. Also, it’s not good for performance, > is it? The single bar and the double bar are not related. | is a method that any class or object can define. For example, Fixnums use it as a bitwise operator: 4 | 1 # 5 (i.e., 100 | 001 == 101) The single bar doesn’t short circuit, whereas the || operator does. nil || 2 || 1/0 # 2 (it never reaches 1/0 so doesn’t blow up) If you do nil || my\_var, my\_var will be reached and will give you an error if it isn’t defined. David ---
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:15pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/3 "2008-05-02T16:15:53Z")
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On May 2, 2008, at 4:01 PM, Iñaki Baz C. wrote: > Of course, I prefer the first case: 2 is already true so there is no > reason to read the next “my\_var”. Also, it’s not good for performance, > is it? | is the bitwise OR and something entirely different from ||. | needs to evaluate both Operands while the value of a logical OR (||) is known at the moment that one of the Operands is true. Regards Florian G. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.8 (Darwin) iEYEARECAAYFAkgbIfUACgkQJA/zY0IIRZakmwCgwtL5xfsC/Z8FRNYN6QaJQn0I FUQAn39NRwifvYwfJcVurDBzJ1/mzH0i =SLZS -----END PGP SIGNATURE----- ---
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:26pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/4 "2008-05-02T16:26:32Z")
2008/5/2, Robert K. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid): > > =\> 2 > > “|” on the other hand is an operator that can be overloaded and it does not > > FalseClass > > TrueClass > > NilClass Thanks to all for those excellent explanations 🙂 ---
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:35pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/5 "2008-05-02T16:35:26Z")
On 02.05.2008 16:15, Florian G. wrote: > > =\> 2 > > > > irb\> nil | 2 | my\_var > > NameError: undefined local variable or method `my\_var’ > > > > Of course, I prefer the first case: 2 is already true so there is no > > reason to read the next “my\_var”. Also, it’s not good for performance, > > is it? > > | is the bitwise OR and something entirely different from ||. This is only mostly correct. Semantics of | depend on the class that implements it. irb(main):001:0\> %w{a b c} | %w{c d e} =\> [“a”, “b”, “c”, “d”, “e”] irb(main):002:0\> %w{a b c} & %w{c d e} =\> [“c”] Cheers robert ---
**Author:** ![ISSSSaki\_Baz\_C](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/51bf81/32.png) [@ISSSSaki\_Baz\_C](https://www.ruby-forum.com/u/ISSSSaki_Baz_C)\ **Post date:** [May 2, 2008, 4:21pm UTC](https://www.ruby-forum.com/t/versus-any-reason-for-this-issue/136485/6 "2008-05-02T16:21:11Z")
On 02.05.2008 16:01, Iñaki Baz C. wrote: > Hi, why in the second case the non declared “my\_var” variable is > processed but not in the first case? Because || does short circuit while | does not. > irb\> nil || 2 || my\_var > =\> 2 > > irb\> nil | 2 | my\_var > NameError: undefined local variable or method `my\_var’ > > Of course, I prefer the first case: 2 is already true so there is no > reason to read the next “my\_var”. Also, it’s not good for performance, > is it? “||” is the short circuit boolean “or” operator. You cannot override it. “|” on the other hand is an operator that can be overloaded and it does not short circuit. It would not make any sense to make “|” short circuit because it can contain arbitrary functionality. Here are some classes that implement “|”: $ ruby -e ‘ObjectSpace.each\_object(Module) {|cl| p cl if cl.instance\_methods.include? “|”}’ Array Bignum Fixnum FalseClass TrueClass NilClass Kind regards robert