Hello,
I have looked around for an answer to this, but havent found exactly what I
need. I want to be able to create a structured dtype representing a C
struct with non-default alignment. An example struct:
struct *__attribute__((aligned(8)))* float2
{
float x;
float y;
};
I can create dtype with two floats easily enough:
float2_dtype = np.dtype( [ ( 'x', 'f4' ), ( 'y', 'f4' ) ], align=True )
but the alignment for this dtype (float2_dtype.alignment) will be 4. This
means that if I pack this dtype into another structured dtype I will get
alignment errors. What I would really like to do is
float2_dtype.alignment = 8 # gives AttributeError: readonly attribute
or
float2_dtype = np.dtype( [ ( 'x', 'f4' ), ( 'y', 'f4' ) ], align=True,
*alignment=8* ) # Invalid keyword argument for dtype()
Is there a way to to this? I apologize if I have missed an obvious
solution to this issue -- I have grepped around the internet and this
mailing list with no success.
Thanks
keith
--
*Selfishness is not living as one wishes to live, it is asking others to
live as one wishes to live.- Oscar Wilde*
_______________________________________________
NumPy-Discussion mailing list -- numpy-discussion@python.org
To unsubscribe send an email to numpy-discussion-le...@python.org
https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
Member address: arch...@mail-archive.com
