Re: [Boost-users] [GIL] copy_and_convert_pixels vs transform_pixels

Subject: Re: [Boost-users] [GIL] copy_and_convert_pixels vs transform_pixels
From: Mateusz Loskot (mateusz_at_[hidden])
Date: 2012-09-27 13:38:47

On 27 September 2012 18:31, Christian Henning <chhenning_at_[hidden]> wrote:
>
> transform_pixels is gil's attempt to provide the user with STL like
> transform algorithm. You can use any sort of operation you want  which
> would include pixel conversion.

Yes, that's clear to me.

> The scope of copy_and_convert_pixels is suppose to be limited to
> converting pixels from one format to another. Sure, you could use this
> function for something else but that's not the intended purpose. Also,
> copy_and_convert_pixel will use a bunch of conversion algorithms ( see
> color_convert.hpp ) automatically. transform_pixels is unaware of such
> operations.

OK, but conceptually the copy_and_convert_pixel is a transform function too.
I presume it would be possible to implement it using the transform_pixel.
So, as you said, the transform_pixel has much wider application scope, indeed.

I simply was missing reference between these two from the docs,
but I'm fairly sure many users may consider these algorithms as
somewhat similar.

Best regards,
-- 
Mateusz Loskot, http://mateusz.loskot.net