From 5705b6a5f885ea8d237f12b6b9e1def126f57435 Mon Sep 17 00:00:00 2001 From: Martin Reinecke Date: Wed, 19 Jun 2019 16:47:00 +0200 Subject: [PATCH] clarify documentation --- pypocketfft.cc | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/pypocketfft.cc b/pypocketfft.cc index 19428b4..9f03d51 100644 --- a/pypocketfft.cc +++ b/pypocketfft.cc @@ -347,7 +347,7 @@ inorm : int 2 : divide by N where N is the product of the lengths of the transformed axes. out : numpy.ndarray (same shape as `a`, complex type with same accuracy as `a`) - May be identical to `a`. + May be identical to `a`, but if it isn't, it must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed @@ -378,6 +378,7 @@ inorm : int where N is the product of the lengths of the transformed input axes. out : numpy.ndarray (complex type with same accuracy as `a`) For the required shape, see the `Returns` section. + Must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed @@ -412,6 +413,7 @@ inorm : int where N is the product of the lengths of the transformed output axes. out : numpy.ndarray (real type with same accuracy as `a`) For the required shape, see the `Returns` section. + Must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed @@ -446,7 +448,7 @@ inorm : int 2 : divide by N where N is the length of `axis`. out : numpy.ndarray (same shape and data type as `a`) - May be identical to `a`. + May be identical to `a`, but if it isn't, it must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed @@ -477,7 +479,7 @@ inorm : int 2 : divide by N where N is the product of the lengths of the transformed axes. out : numpy.ndarray (same shape and data type as `a`) - May be identical to `a`. + May be identical to `a`, but if it isn't, it must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed @@ -509,7 +511,7 @@ inorm : int 2 : divide by N where N is the product of the lengths of the transformed axes. out : numpy.ndarray (same shape and data type as `a`) - May be identical to `a`. + May be identical to `a`, but if it isn't, it must not overlap with `a`. If None, a new array is allocated to store the output. nthreads : int Number of threads to use. If 0, use the system default (typically governed -- GitLab