[FFmpeg-devel] [PATCH 2/3] Create a public API for FFT family of functions

[Måns Rullgård]

Alex Converse <alex.converse at gmail.com> writes:

> On Sat, Mar 6, 2010 at 9:26 AM, Michael Niedermayer <michaelni at gmx.at> wrote:
>> On Sat, Mar 06, 2010 at 07:33:16AM -0500, Alex Converse wrote:
>>> On Fri, Mar 5, 2010 at 10:38 PM, Mans Rullgard <mans at mansr.com> wrote:
>>> > ---
>>> > ?libavcodec/Makefile | ? ?4 +-
>>> > ?libavcodec/avfft.c ?| ?142 +++++++++++++++++++++++++++++++++++++++++++++++++++
>>> > ?libavcodec/avfft.h ?| ? 89 ++++++++++++++++++++++++++++++++
>>> > ?libavcodec/fft.h ? ?| ? 28 +++--------
>>> > ?4 files changed, 240 insertions(+), 23 deletions(-)
>>>
>>> > +
>>> > +/* Real Discrete Fourier Transform */
>>> > +
>>> > +enum RDFTransformType {
>>> > + ? ?RDFT,
>>> > + ? ?IRDFT,
>>> > + ? ?RIDFT,
>>> > + ? ?IRIDFT,
>>> > +};
>>>
>>> I never really liked the name IRIDFT it doesn't make a lot of sense to
>>> me. I just needed something at the time. Maybe we should change it
>>> before making it public.
>>>
>>> RDFT: real time -> complex freq
>>> IRDFT: complex freq -> real time
>>> RIDFT: real freq -> complex time (or alternatively real time to
>>> complex frequency with alternate sine convention)
>>> IRIDFT: complex time -> real frequency (or alternatively complex
>>> frequency to real time with alternate sine converntion)
>>
>> what about:
>> R2C_DFT
>> C2R_IDFT
>> R2C_IDFT
>> C2R_DFT
>> ?
>>
>
> that's better. I like DFT_C2R over C2R_DFT but I don't really care
> about it enough to make a big deal over it.

I was about to suggest the same change.

-- 
M?ns Rullg?rd
mans at mansr.com