what's the difference between | and || in matlab??

what's the difference between | and || in matlab??

Answer 1

[matt dash]

Clarification: for short circuit "and" no tests are evaluated after the first "false". For short circuit "or" no tests are evaluated after the first true.

Answer 2

[Sean de Wolski]

| applies to each element in the array, || applies to a scalar condition:

[1 0 1] | [ 0 0 1]

v.

 [1 0 1] || [ 0 0 1]

[Stephen23]

It is unclear to me what those code examples are supposed to clarify:

[1 0 1] | [0 0 1]

ans = 1×3 logical array
   1   0   1

[1 0 1] || [0 0 1]

Operands to the logical AND (&&) and OR (||) operators must be convertible to logical scalar values. Use the ANY or ALL functions to reduce operands to logical scalar values.

[Steve Van Hooser]

The | operator does an element-by-element comparison on its inputs and produces an output the same size as the input (as in the first example).

The || operator can only operate on scalars and produces a logical scalar as an output.

The second statement fails because the inputs (operands) are vectors.

[DGM]

I hate to nitpick, but I'm bored and have a minor tweak to add:

The || is a scalar operator, and the | operator is elementwise; that's true. The problem is describing what's expected of an elementwise operator's output size (at least if we're going to explicitly describe it in the context of an ancient question).

Now if we're living in 2014, then the statement that the output is the same size as the input would be true. In the present day, the size of the output of an elementwise operator isn't necessarily the same size as the input. That's only the case when the two inputs are the same size.

Assuming that the inputs have compatible size, the output size is the maximum of input sizes (assuming we're using R2016b or newer (or are using bsxfun() in older versions)).

[1 0 1] | [0 0 1] % output is 1x3

ans = 1x3 logical array
   1   0   1

[1 0 1] | [0 0 1].' % output is 3x3

ans = 3x3 logical array
   1   0   1
   1   0   1
   1   1   1

Even without this addendum, I think that @Steve Van Hooser's comment is a more meaningful answer than the original. The short-circuit functionality is really secondary to the fact that || is scalar.

[Steve Van Hooser]

Nice addition. I don't see how to give a thumbs up so I will do it in words.