On Thu, Aug 31, 2023 at 07:16:20PM +0200, Stefano Sabatini wrote:
> On date Thursday 2023-08-31 18:51:52 +0200, Andreas Rheinhardt wrote:
> > Stefano Sabatini:
> > > +int sws_get_gaussian_vec(SwsVector **vecp,
> > > +                         AVClass *log_ctx,
> > > +                         double standard_deviation, double quality);
> > >  
> > 
> > Seriously? A pointer to an AVClass as log_ctx? It is actually AVClass**
> > (the logcontext must have a pointer to an AVClass as its first member),
> > but we always use NULL.
> 
> Sorry, sloppy editing on my side.
> 
> > Apart from that: I am not really convinced that the improvement is worth
> > the hassle.
> 
> This is not a high-profile function (probably it's never used outside
> lavfi) and provides an opportunity to move the API to the correct
> direction.

>  doc/APIchanges             |    3 +++
>  libswscale/swscale.h       |   27 ++++++++++++++++++++++++++-
>  libswscale/utils.c         |   42 +++++++++++++++++++++++++++++++++---------
>  libswscale/version.h       |    2 +-
>  libswscale/version_major.h |    4 ++++
>  5 files changed, 67 insertions(+), 11 deletions(-)
> b91b721cea2752b28a51aaeab2a464b2699dfb49  0001-lsws-swscale.h-introduce-sws_get_gaussian_vec.patch
> From 69c33f62e15de3d199d54187d38c0856418f0981 Mon Sep 17 00:00:00 2001
> From: Stefano Sabatini <stefasab@gmail.com>
> Date: Sat, 26 Aug 2023 14:20:35 +0200
> Subject: [PATCH 1/2] lsws/swscale.h: introduce sws_get_gaussian_vec
> 
> Use in place of sws_getGaussianVec.
> 
> The new function enable better error handling, and provide better naming
> for the variance variable, now named standard_deviation to reflect the
> meaning of the parameter.
> ---
>  doc/APIchanges             |  3 +++
>  libswscale/swscale.h       | 27 +++++++++++++++++++++++-
>  libswscale/utils.c         | 42 ++++++++++++++++++++++++++++++--------
>  libswscale/version.h       |  2 +-
>  libswscale/version_major.h |  4 ++++
>  5 files changed, 67 insertions(+), 11 deletions(-)
> 
> diff --git a/doc/APIchanges b/doc/APIchanges
> index ad1efe708d..bad2d61027 100644
> --- a/doc/APIchanges
> +++ b/doc/APIchanges
> @@ -2,6 +2,9 @@ The last version increases of all libraries were on 2023-02-09
>  
>  API changes, most recent first:
>  
> +2023-08-26 - xxxxxxxxxx - lsws 7.4.100 - swscale.h
> +  Introduce sws_get_gaussian_vec, use in place of sws_getGaussianVec.
> +
>  2023-08-18 - xxxxxxxxxx - lavu 58.17.100 - channel_layout.h
>    All AV_CHANNEL_LAYOUT_* macros are now compatible with C++ 17 and older.
>  
> diff --git a/libswscale/swscale.h b/libswscale/swscale.h
> index 9d4612aaf3..55f2fc4a48 100644
> --- a/libswscale/swscale.h
> +++ b/libswscale/swscale.h
> @@ -355,11 +355,36 @@ int sws_getColorspaceDetails(struct SwsContext *c, int **inv_table,
>   */
>  SwsVector *sws_allocVec(int length);
>  
> +#if FF_API_SWS_GET_GAUSSIAN_VEC
>  /**
> - * Return a normalized Gaussian curve used to filter stuff
> + * Return a normalized Gaussian curve used to filter stuff.
> + *
>   * quality = 3 is high quality, lower is lower quality.
> + * @deprecated use sws_get_gaussian_vector()
>   */
> +attribute_deprecated
>  SwsVector *sws_getGaussianVec(double variance, double quality);
> +#endif
> +
> +/**
> + * Compute and return a normalized Gaussian vector.
> + *
> + * @param vecp: pointer where the computed vector is put in case of
> + *        success
> + * @param standard_deviation the standard deviation used to generate
> + *        the Gaussian vector, must be a non-negative value
> + * @param quality the quality of the generated Gaussian vector, must
> + *        be a non-negative value. It affects the lenght of the generated
> + *        vector. A value equal to 3 corresponds to high quality.
> + * @param log_ctx a pointer to an arbitrary struct of which the first
> + *        field is a pointer to an AVClass struct (used for av_log)
> + *        used for logging, can be NULL
> + *
> + * @return a negative error code on error, non negative otherwise
> + */
> +int sws_get_gaussian_vec(SwsVector **vecp,
> +                         double standard_deviation, double quality,
> +                         void *log_ctx);

which of the two do you consider better?

First, here the central part we return is the vector

SwsVector *gaus_vec = sws_getGaussianVec(NULL, 1, 2);
SwsVector *temp_vec = sws_ConvolveVec(NULL, in_vec, gaus_vec);
sws_averageVec(temp_vec, temp_vec, in_vec);

av_free(gaus_vec);
return temp_vec; // Error checking here happens by temp_vec being NULL in all cases of error

vs.

Second, here the central part we return is the error code

SwsVector *gaus_vec = NULL;
SwsVector *temp_vec = NULL;
int err = sws_getGaussianVec(&gaus_vec, 1, 2);
if (err<0)
    goto fail;

err = sws_ConvolveVec(&temp_vec, in_vec, gaus_vec);
if (err<0)
    goto fail;

err = sws_averageVec(&temp_vec, temp_vec, in_vec);
if (err<0)
    goto fail;

*ret_argument = temp_vec
return 0;
fail:
    av_free(gaus_vec)
    av_free(temp_vec)
return ret;

thx

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact