[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Odd behavior of time-domain convolutions
From: |
John W. Eaton |
Subject: |
Re: Odd behavior of time-domain convolutions |
Date: |
Wed, 05 Sep 2007 13:33:48 -0400 |
On 5-Sep-2007, David Bateman wrote:
| Note that xr and yr are defined as "const" to avoid a copy being made
| with the fortran_vec method.
In cases like this, I typically use data() instead of fortran_vec(),
though the results are the same. But as you say, the important thing
is to declare the Array object const so that the const version of the
fortran_vec() method is used. Note that currently there is only a
const version of the data() method, so using it means that the code
will only compile without warning if the Array object is actually
const.
jwe