[Scipy-tickets] [SciPy] #201: Bug in UnivariateSpline.integrate

SciPy scipy-tickets at scipy.net
Tue Dec 5 08:56:50 CST 2006


#201: Bug in UnivariateSpline.integrate
-------------------------------+--------------------------------------------
 Reporter:  dhuard             |        Owner:  pearu
     Type:  defect             |       Status:  new  
 Priority:  normal             |    Milestone:       
Component:  scipy.interpolate  |      Version:       
 Severity:  normal             |   Resolution:       
 Keywords:  splines            |  
-------------------------------+--------------------------------------------
Changes (by dhuard):

  * summary:  Bug in UnivariateSpline.integrate and .derivatives => Bug in
              UnivariateSpline.integrate

Comment:

 US.derivatives seems to run fine now.
 On the other hand, US.integral(2,5) still bugs with the following message:
 {{{
 /usr/local/lib/python2.4/site-packages/scipy/interpolate/fitpack2.py in
 integral (self, a, b)
     186         given points.
     187         """
 --> 188         return dfitpack.splint(*(self._eval_args+(a,b)))
     189
     190     def derivatives(self, x):

 The problem seems to be again one of length of passed arrays. In the
 example, both t and c have the same length, whereas splint assumes t is
 (n) and c is (n-k-1).
 error: failed in converting 2nd argument `c' of dfitpack.splint to
 C/Fortran array
 }}}
 Scipy: 0.5.2.dev2287

-- 
Ticket URL: <http://projects.scipy.org/scipy/scipy/ticket/201#comment:1>
SciPy <http://www.scipy.org/>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list