Re: [Pixman] [PATCH v14 08/22] pixman-filter: rename "scale" to "size" when it is 1/scale

[Bill Spitzak]

Okay that sounds pretty good. Also 'r' is better than 'i' since 'i' can be
mis-read as "input" or "image". 'r' I think is mostly going to be read as
"reverse" which actually makes sense.

Sorry to be a pain about this but I really find it confusing to use the
same term for different numbers.

On Sun, Apr 3, 2016 at 9:29 AM, Søren Sandmann 
wrote:

> I guess I can live with 'rscale' (for reciprocal scale).
>
>
> Søren
>
> On Thu, Mar 24, 2016 at 9:30 PM, Bill Spitzak  wrote:
>
>> Would using "iscale" (for inverse scale) work? Another is "tscale"
>> because it is the scale from the transform matrix.
>>
>> I really would like to use two different names for these variables as it
>> is really confusing using the same one.
>>
>>
>> On Fri, Mar 18, 2016 at 8:24 AM, Bill Spitzak  wrote:
>>
>>>
>>>
>>> On Thu, Mar 17, 2016 at 10:06 PM, Søren Sandmann <
>>> soren.sandm...@gmail.com> wrote:
>>>
 I suppose it's a little illogical that scale_x and scale_y really are
 the reciprocal values of how much the source image should be scaled. I
 don't remember exactly what I was thinking, but it might have something
 like "this allows you to just pass t[0][0] and t[1][1] if you have a pure
 scaling, which avoids a division". There is also kind of precedence in the
 Pixman API since the transformation given as in the dst->source direction.

 I don't really like "size" either though. It's not really the size of
 the filter that we are specifying; it just happens to be proportional to 
 it.

 If it comes down to "size" and "scale", I prefer "scale".

>>>
>>> I tried "width" but it is not actually the filter width. How about "dx"
>>> and "dy" since it is almost always the derivative of the x and y in the
>>> input (or du,dv or dtx dtx)
>>>


 Søren


 On Sun, Mar 6, 2016 at 8:06 PM,  wrote:

> From: Bill Spitzak 
>
> This is to remove some confusion when reading the code. "scale" gets
> larger
> as the picture gets larger, while "size" (ie the size of the filter)
> gets
> smaller.
>
> v14: Removed changes to integral function
>
> Signed-off-by: Bill Spitzak 
> Reviewed-by: Oded Gabbay 
> ---
>  pixman/pixman-filter.c | 18 +-
>  pixman/pixman.h|  6 +++---
>  2 files changed, 12 insertions(+), 12 deletions(-)
>
> diff --git a/pixman/pixman-filter.c b/pixman/pixman-filter.c
> index a29116a..c03a7f6 100644
> --- a/pixman/pixman-filter.c
> +++ b/pixman/pixman-filter.c
> @@ -221,7 +221,7 @@ static void
>  create_1d_filter (int  width,
>   pixman_kernel_t  reconstruct,
>   pixman_kernel_t  sample,
> - double   scale,
> + double   size,
>   int  n_phases,
>   pixman_fixed_t *p)
>  {
> @@ -251,8 +251,8 @@ create_1d_filter (int  width,
> double pos = x + 0.5 - frac;
> double rlow = - filters[reconstruct].width / 2.0;
> double rhigh = rlow + filters[reconstruct].width;
> -   double slow = pos - scale * filters[sample].width / 2.0;
> -   double shigh = slow + scale * filters[sample].width;
> +   double slow = pos - size * filters[sample].width / 2.0;
> +   double shigh = slow + size * filters[sample].width;
> double c = 0.0;
> double ilow, ihigh;
>
> @@ -262,7 +262,7 @@ create_1d_filter (int  width,
> ihigh = MIN (shigh, rhigh);
>
> c = integral (reconstruct, ilow,
> - sample, 1.0 / scale, ilow - pos,
> + sample, 1.0 / size, ilow - pos,
>   ihigh - ilow);
> }
>
> @@ -335,12 +335,12 @@ filter_width (pixman_kernel_t reconstruct,
> pixman_kernel_t sample, double size)
>  }
>
>  /* Create the parameter list for a SEPARABLE_CONVOLUTION filter
> - * with the given kernels and scale parameters
> + * with the given kernels and size parameters
>   */
>  PIXMAN_EXPORT pixman_fixed_t *
>  pixman_filter_create_separable_convolution (int *n_values,
> -   pixman_fixed_t   scale_x,
> -   pixman_fixed_t   scale_y,
> +   pixman_fixed_t   size_x,
> +   pixman_fixed_t   size_y,
> pixman_kernel_t
> reconstruct_x,
> 

Re: [Pixman] [PATCH v14 08/22] pixman-filter: rename "scale" to "size" when it is 1/scale

[Søren Sandmann]

I guess I can live with 'rscale' (for reciprocal scale).


Søren

On Thu, Mar 24, 2016 at 9:30 PM, Bill Spitzak  wrote:

> Would using "iscale" (for inverse scale) work? Another is "tscale" because
> it is the scale from the transform matrix.
>
> I really would like to use two different names for these variables as it
> is really confusing using the same one.
>
>
> On Fri, Mar 18, 2016 at 8:24 AM, Bill Spitzak  wrote:
>
>>
>>
>> On Thu, Mar 17, 2016 at 10:06 PM, Søren Sandmann <
>> soren.sandm...@gmail.com> wrote:
>>
>>> I suppose it's a little illogical that scale_x and scale_y really are
>>> the reciprocal values of how much the source image should be scaled. I
>>> don't remember exactly what I was thinking, but it might have something
>>> like "this allows you to just pass t[0][0] and t[1][1] if you have a pure
>>> scaling, which avoids a division". There is also kind of precedence in the
>>> Pixman API since the transformation given as in the dst->source direction.
>>>
>>> I don't really like "size" either though. It's not really the size of
>>> the filter that we are specifying; it just happens to be proportional to it.
>>>
>>> If it comes down to "size" and "scale", I prefer "scale".
>>>
>>
>> I tried "width" but it is not actually the filter width. How about "dx"
>> and "dy" since it is almost always the derivative of the x and y in the
>> input (or du,dv or dtx dtx)
>>
>>>
>>>
>>> Søren
>>>
>>>
>>> On Sun, Mar 6, 2016 at 8:06 PM,  wrote:
>>>
 From: Bill Spitzak 

 This is to remove some confusion when reading the code. "scale" gets
 larger
 as the picture gets larger, while "size" (ie the size of the filter)
 gets
 smaller.

 v14: Removed changes to integral function

 Signed-off-by: Bill Spitzak 
 Reviewed-by: Oded Gabbay 
 ---
  pixman/pixman-filter.c | 18 +-
  pixman/pixman.h|  6 +++---
  2 files changed, 12 insertions(+), 12 deletions(-)

 diff --git a/pixman/pixman-filter.c b/pixman/pixman-filter.c
 index a29116a..c03a7f6 100644
 --- a/pixman/pixman-filter.c
 +++ b/pixman/pixman-filter.c
 @@ -221,7 +221,7 @@ static void
  create_1d_filter (int  width,
   pixman_kernel_t  reconstruct,
   pixman_kernel_t  sample,
 - double   scale,
 + double   size,
   int  n_phases,
   pixman_fixed_t *p)
  {
 @@ -251,8 +251,8 @@ create_1d_filter (int  width,
 double pos = x + 0.5 - frac;
 double rlow = - filters[reconstruct].width / 2.0;
 double rhigh = rlow + filters[reconstruct].width;
 -   double slow = pos - scale * filters[sample].width / 2.0;
 -   double shigh = slow + scale * filters[sample].width;
 +   double slow = pos - size * filters[sample].width / 2.0;
 +   double shigh = slow + size * filters[sample].width;
 double c = 0.0;
 double ilow, ihigh;

 @@ -262,7 +262,7 @@ create_1d_filter (int  width,
 ihigh = MIN (shigh, rhigh);

 c = integral (reconstruct, ilow,
 - sample, 1.0 / scale, ilow - pos,
 + sample, 1.0 / size, ilow - pos,
   ihigh - ilow);
 }

 @@ -335,12 +335,12 @@ filter_width (pixman_kernel_t reconstruct,
 pixman_kernel_t sample, double size)
  }

  /* Create the parameter list for a SEPARABLE_CONVOLUTION filter
 - * with the given kernels and scale parameters
 + * with the given kernels and size parameters
   */
  PIXMAN_EXPORT pixman_fixed_t *
  pixman_filter_create_separable_convolution (int *n_values,
 -   pixman_fixed_t   scale_x,
 -   pixman_fixed_t   scale_y,
 +   pixman_fixed_t   size_x,
 +   pixman_fixed_t   size_y,
 pixman_kernel_t
 reconstruct_x,
 pixman_kernel_t
 reconstruct_y,
 pixman_kernel_t  sample_x,
 @@ -348,8 +348,8 @@ pixman_filter_create_separable_convolution (int
  *n_values,
 int
 subsample_bits_x,
 int
 subsample_bits_y)
  {
 -double sx = fabs (pixman_fixed_to_double (scale_x));
 -double sy = fabs (pixman_fixed_to_double (scale_y));
 +

Re: [Pixman] [PATCH v14 08/22] pixman-filter: rename "scale" to "size" when it is 1/scale

[Bill Spitzak]

Would using "iscale" (for inverse scale) work? Another is "tscale" because
it is the scale from the transform matrix.

I really would like to use two different names for these variables as it is
really confusing using the same one.


On Fri, Mar 18, 2016 at 8:24 AM, Bill Spitzak  wrote:

>
>
> On Thu, Mar 17, 2016 at 10:06 PM, Søren Sandmann  > wrote:
>
>> I suppose it's a little illogical that scale_x and scale_y really are the
>> reciprocal values of how much the source image should be scaled. I don't
>> remember exactly what I was thinking, but it might have something like
>> "this allows you to just pass t[0][0] and t[1][1] if you have a pure
>> scaling, which avoids a division". There is also kind of precedence in the
>> Pixman API since the transformation given as in the dst->source direction.
>>
>> I don't really like "size" either though. It's not really the size of the
>> filter that we are specifying; it just happens to be proportional to it.
>>
>> If it comes down to "size" and "scale", I prefer "scale".
>>
>
> I tried "width" but it is not actually the filter width. How about "dx"
> and "dy" since it is almost always the derivative of the x and y in the
> input (or du,dv or dtx dtx)
>
>>
>>
>> Søren
>>
>>
>> On Sun, Mar 6, 2016 at 8:06 PM,  wrote:
>>
>>> From: Bill Spitzak 
>>>
>>> This is to remove some confusion when reading the code. "scale" gets
>>> larger
>>> as the picture gets larger, while "size" (ie the size of the filter) gets
>>> smaller.
>>>
>>> v14: Removed changes to integral function
>>>
>>> Signed-off-by: Bill Spitzak 
>>> Reviewed-by: Oded Gabbay 
>>> ---
>>>  pixman/pixman-filter.c | 18 +-
>>>  pixman/pixman.h|  6 +++---
>>>  2 files changed, 12 insertions(+), 12 deletions(-)
>>>
>>> diff --git a/pixman/pixman-filter.c b/pixman/pixman-filter.c
>>> index a29116a..c03a7f6 100644
>>> --- a/pixman/pixman-filter.c
>>> +++ b/pixman/pixman-filter.c
>>> @@ -221,7 +221,7 @@ static void
>>>  create_1d_filter (int  width,
>>>   pixman_kernel_t  reconstruct,
>>>   pixman_kernel_t  sample,
>>> - double   scale,
>>> + double   size,
>>>   int  n_phases,
>>>   pixman_fixed_t *p)
>>>  {
>>> @@ -251,8 +251,8 @@ create_1d_filter (int  width,
>>> double pos = x + 0.5 - frac;
>>> double rlow = - filters[reconstruct].width / 2.0;
>>> double rhigh = rlow + filters[reconstruct].width;
>>> -   double slow = pos - scale * filters[sample].width / 2.0;
>>> -   double shigh = slow + scale * filters[sample].width;
>>> +   double slow = pos - size * filters[sample].width / 2.0;
>>> +   double shigh = slow + size * filters[sample].width;
>>> double c = 0.0;
>>> double ilow, ihigh;
>>>
>>> @@ -262,7 +262,7 @@ create_1d_filter (int  width,
>>> ihigh = MIN (shigh, rhigh);
>>>
>>> c = integral (reconstruct, ilow,
>>> - sample, 1.0 / scale, ilow - pos,
>>> + sample, 1.0 / size, ilow - pos,
>>>   ihigh - ilow);
>>> }
>>>
>>> @@ -335,12 +335,12 @@ filter_width (pixman_kernel_t reconstruct,
>>> pixman_kernel_t sample, double size)
>>>  }
>>>
>>>  /* Create the parameter list for a SEPARABLE_CONVOLUTION filter
>>> - * with the given kernels and scale parameters
>>> + * with the given kernels and size parameters
>>>   */
>>>  PIXMAN_EXPORT pixman_fixed_t *
>>>  pixman_filter_create_separable_convolution (int *n_values,
>>> -   pixman_fixed_t   scale_x,
>>> -   pixman_fixed_t   scale_y,
>>> +   pixman_fixed_t   size_x,
>>> +   pixman_fixed_t   size_y,
>>> pixman_kernel_t
>>> reconstruct_x,
>>> pixman_kernel_t
>>> reconstruct_y,
>>> pixman_kernel_t  sample_x,
>>> @@ -348,8 +348,8 @@ pixman_filter_create_separable_convolution (int
>>>*n_values,
>>> int
>>> subsample_bits_x,
>>> int
>>> subsample_bits_y)
>>>  {
>>> -double sx = fabs (pixman_fixed_to_double (scale_x));
>>> -double sy = fabs (pixman_fixed_to_double (scale_y));
>>> +double sx = fabs (pixman_fixed_to_double (size_x));
>>> +double sy = fabs (pixman_fixed_to_double (size_y));
>>>  pixman_fixed_t *params;
>>>  int subsample_x, subsample_y;
>>>  int width, height;
>>> diff --git a/pixman/pixman.h b/pixman/pixman.h
>>> 

Re: [Pixman] [PATCH v14 08/22] pixman-filter: rename "scale" to "size" when it is 1/scale

[Bill Spitzak]

On Thu, Mar 17, 2016 at 10:06 PM, Søren Sandmann 
wrote:

> I suppose it's a little illogical that scale_x and scale_y really are the
> reciprocal values of how much the source image should be scaled. I don't
> remember exactly what I was thinking, but it might have something like
> "this allows you to just pass t[0][0] and t[1][1] if you have a pure
> scaling, which avoids a division". There is also kind of precedence in the
> Pixman API since the transformation given as in the dst->source direction.
>
> I don't really like "size" either though. It's not really the size of the
> filter that we are specifying; it just happens to be proportional to it.
>
> If it comes down to "size" and "scale", I prefer "scale".
>

I tried "width" but it is not actually the filter width. How about "dx" and
"dy" since it is almost always the derivative of the x and y in the input
(or du,dv or dtx dtx)

>
>
> Søren
>
>
> On Sun, Mar 6, 2016 at 8:06 PM,  wrote:
>
>> From: Bill Spitzak 
>>
>> This is to remove some confusion when reading the code. "scale" gets
>> larger
>> as the picture gets larger, while "size" (ie the size of the filter) gets
>> smaller.
>>
>> v14: Removed changes to integral function
>>
>> Signed-off-by: Bill Spitzak 
>> Reviewed-by: Oded Gabbay 
>> ---
>>  pixman/pixman-filter.c | 18 +-
>>  pixman/pixman.h|  6 +++---
>>  2 files changed, 12 insertions(+), 12 deletions(-)
>>
>> diff --git a/pixman/pixman-filter.c b/pixman/pixman-filter.c
>> index a29116a..c03a7f6 100644
>> --- a/pixman/pixman-filter.c
>> +++ b/pixman/pixman-filter.c
>> @@ -221,7 +221,7 @@ static void
>>  create_1d_filter (int  width,
>>   pixman_kernel_t  reconstruct,
>>   pixman_kernel_t  sample,
>> - double   scale,
>> + double   size,
>>   int  n_phases,
>>   pixman_fixed_t *p)
>>  {
>> @@ -251,8 +251,8 @@ create_1d_filter (int  width,
>> double pos = x + 0.5 - frac;
>> double rlow = - filters[reconstruct].width / 2.0;
>> double rhigh = rlow + filters[reconstruct].width;
>> -   double slow = pos - scale * filters[sample].width / 2.0;
>> -   double shigh = slow + scale * filters[sample].width;
>> +   double slow = pos - size * filters[sample].width / 2.0;
>> +   double shigh = slow + size * filters[sample].width;
>> double c = 0.0;
>> double ilow, ihigh;
>>
>> @@ -262,7 +262,7 @@ create_1d_filter (int  width,
>> ihigh = MIN (shigh, rhigh);
>>
>> c = integral (reconstruct, ilow,
>> - sample, 1.0 / scale, ilow - pos,
>> + sample, 1.0 / size, ilow - pos,
>>   ihigh - ilow);
>> }
>>
>> @@ -335,12 +335,12 @@ filter_width (pixman_kernel_t reconstruct,
>> pixman_kernel_t sample, double size)
>>  }
>>
>>  /* Create the parameter list for a SEPARABLE_CONVOLUTION filter
>> - * with the given kernels and scale parameters
>> + * with the given kernels and size parameters
>>   */
>>  PIXMAN_EXPORT pixman_fixed_t *
>>  pixman_filter_create_separable_convolution (int *n_values,
>> -   pixman_fixed_t   scale_x,
>> -   pixman_fixed_t   scale_y,
>> +   pixman_fixed_t   size_x,
>> +   pixman_fixed_t   size_y,
>> pixman_kernel_t
>> reconstruct_x,
>> pixman_kernel_t
>> reconstruct_y,
>> pixman_kernel_t  sample_x,
>> @@ -348,8 +348,8 @@ pixman_filter_create_separable_convolution (int
>>*n_values,
>> int
>> subsample_bits_x,
>> int
>> subsample_bits_y)
>>  {
>> -double sx = fabs (pixman_fixed_to_double (scale_x));
>> -double sy = fabs (pixman_fixed_to_double (scale_y));
>> +double sx = fabs (pixman_fixed_to_double (size_x));
>> +double sy = fabs (pixman_fixed_to_double (size_y));
>>  pixman_fixed_t *params;
>>  int subsample_x, subsample_y;
>>  int width, height;
>> diff --git a/pixman/pixman.h b/pixman/pixman.h
>> index 509ba5e..b012a33 100644
>> --- a/pixman/pixman.h
>> +++ b/pixman/pixman.h
>> @@ -845,12 +845,12 @@ typedef enum
>>  } pixman_kernel_t;
>>
>>  /* Create the parameter list for a SEPARABLE_CONVOLUTION filter
>> - * with the given kernels and scale parameters.
>> + * with the given kernels and size parameters.
>>   */
>>  pixman_fixed_t *
>>  pixman_filter_create_separable_convolution (int *n_values,
>> -