## Issue

In brief, I have the code below in Python

```
[n, m] = f()
```

and I want to convert it to Julia, which will not be more than one line (I hope).

Here is an example in Python:

```
from numpy import *
x = [0,1,2]
y = [3,4,5]
x = array(x)
y = array(y)
def f():
z = concatenate((x, y))
a = z*2
b = z*3
return [a, b]
def g():
[n,m] = f()
n = n/2
m = m/3
return [n, m]
print(g())
```

This is how I wanted it to be in Julia, but did not work:

```
x = [0,1,2]
y = [3,4,5]
function f()
z = vcat(x, y)
a = z*2
b = z*3
return [a, b]
end
function g()
[n, m] = f()
n = n/2
m = m/3
return [n, m]
end
print(g())
```

Here is how I made it work, but I do not want codes like this:

```
x = [0,1,2]
y = [3,4,5]
function f()
z = vcat(x, y)
a = z*2
b = z*3
global c = [a, b]
return c
end
function g()
n = c[1]
m = c[2]
n = n/2
m = m/3
return [n, m]
end
f()
print(g())
```

Thank you very much.

## Solution

Just drop square brackets like this:

```
x = [0,1,2]
y = [3,4,5]
function f()
z = vcat(x, y)
a = z*2
b = z*3
return a, b # here it is not needed, but typically in such cases it is dropped to return a tuple
end
function g()
n, m = f() # here it is needed
n = n/2
m = m/3
return n, m # here it is not needed, but typically in such cases it is dropped to return a tuple
end
print(g())
```

The reason why it is typically better to return a `Tuple`

rather than a vector is that `Tuple`

does not perform auto promotion of values (+ it is non-allocating):

```
julia> (1, 2.0) # here 1 stays an integer
(1, 2.0)
julia> [1, 2.0] # here 1 got implicitly converted to 1.0
2-element Vector{Float64}:
1.0
2.0
```

Of course if you want such implicit behavior use a vector.

Answered By – Bogumił Kamiński

Answer Checked By – Senaida (BugsFixing Volunteer)