Plus as concatenate for numeric vectors?

I've been using Matlab for years and I've never seen this before, is it new? It was hard to debug this since it's pretty subtle

size([ones(1,10) +ones(1,10)])

ans = 1×2

1 20

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

Apparently if you don't put a space after the plus sign -- and there is a space before it -- then + concatenates instead of adding. There's nothing in the documentation for + about concatenating numeric arrays. I don't really have a question, this just seems really confusing as a feature.

[Stephen23]

You are using a unary plus: https://www.mathworks.com/help/matlab/ref/double.uplus.html

This behavior is explained here:

https://www.mathworks.com/help/matlab/matlab_prog/operator-precedence.html

https://www.mathworks.com/help/matlab/math/creating-and-concatenating-matrices.html

Answer 1

[dpb]

No it's not new. In that expression the plus sign is interpreted as the sign of the value, not as an arithmetic operator.

It's no different than

[1 +1] 

ans = 1×2

1 1

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

or

[1 -1]

ans = 1×2

1 -1

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

the latter of which wouldn't surprise you -- in fact, you'd be very disconcerted if it produced the same as

[1-1]

ans = 0

or

[1 - 1]

ans = 0