# minimize_constrained for a one dimensional function

Hello all,

I have a bit of a problem with the minimize_constrained function: When I call

```
minimize_constrained(lambda x: x^2, [lambda x: 1], 99.6)
```

I get a `TypeError: iteration over a 0-d array`

(why?). If I call

```
minimize_constrained(lambda x: x^2, \
[lambda x: vector({0:1,1:x}).inner_product(vector((0,0)))], 99.6)
```

I get a `TypeError: unable to find a common ring for all elements`

.
This seems to be the case because the the constraint function is sometimes called
with a `numpy.ndarray`

and at other times with a `sage.modules.vector_real_double_dense.Vector_real_double_dense`

. In the first case
I need to call `float(x[0])`

, in the latter I can use `x[0]`

directly.

Is it be possible to convert the data to sage vectors throughout the iterations or do I have to set up a handler to sanitize the input myself?