[Numpy-tickets] [NumPy] #425: Faster clip for scalar clipping values

[NumPy]

#425: Faster clip for scalar clipping values
------------------------------+---------------------------------------------
 Reporter:  david cournapeau  |        Owner:  somebody
     Type:  enhancement       |       Status:  new     
 Priority:  normal            |    Milestone:          
Component:  Other             |      Version:  devel   
 Severity:  normal            |   Resolution:          
 Keywords:                    |  
------------------------------+---------------------------------------------
Comment (by david cournapeau):

 Ok for the magic list. Something like int fastclip(void* in, npy_intp
 size, void* min, void* max, void* out) ?

 For the coercion, I was not precise enough: my question is about the
 actual C implementation for coercion. The patch is now using
 PyArray_ConvertToCommonType for coercion which follows the rule with
 respect to scalar; but when I am looking to existing numpy code using f
 functions, PyArray_ObjectType is used, and if I remember correctly, this
 function does not change its behaviour if the object is a scalar or not ?
 If PyArrayConvertToCommonType is ok, I can modify my patch easily and
 commit the change today

-- 
Ticket URL: <http://projects.scipy.org/scipy/numpy/ticket/425#comment:4>
NumPy <http://projects.scipy.org/scipy/numpy>
The fundamental package needed for scientific computing with Python.