2012/11/5 Diego Biurrun <[email protected]>:
> On Sat, Nov 03, 2012 at 10:14:05PM +0100, Jordi Ortiz wrote:
>> +static void spatial_compose97i_init2(DiracDWTCompose *cs, IDWTELEM *buffer,
>> +                                     int height, int stride)
>> +
>> +static void spatial_compose53i_init2(DiracDWTCompose *cs, IDWTELEM *buffer,
>> +                                     int height, int stride)
>> +
>> +int ff_spatial_idwt_init2(DiracDWTContext *d, IDWTELEM *buffer, int width,
>> +                          int height, int stride, enum dwt_type type,
>> +                          int decomposition_count, IDWTELEM *temp)
>
> Why are all these functions called "_init2()"?  Where's the corresponding
> "_init()"?

These functions were located in dwt.c where already exist the init()
version. I'm going to change init2->init in my next patch.

>
>> +typedef struct DiracDWTContext {
>> +    void (*vertical_compose_h1)(void);
>> +    void (*vertical_compose)(void);
>> +     ///< vertical_compose -> one set of lowpass and highpass combined
>
>    void (*vertical_compose)(void); ///< vertical_compose -> one set of
>                                         lowpass and highpass combined
>

     void (*vertical_compose)(void); // vertical_compose -> one set of
                                                        // lowpass and
highpass combined

ok?


>> +// -1 if an error occurred, e.g. the dwt_type isn't recognized
>> +int ff_spatial_idwt_init3(DiracDWTContext *d, IDWTELEM *buffer, int width,
>> +                          int height, int stride, enum dwt_type type,
>> +                          int decomposition_count, IDWTELEM *temp);
>
> -1 what?  The comment is rather mysterious...

That comment has no sense any more. Now AV_ERROR code is returned on
error and 0 if SUCCESS, I'll just remove the comment if you agree

>> --- /dev/null
>> +++ b/libavcodec/diracdsp.c
>> @@ -0,0 +1,301 @@
>> +
>> +#include "dsputil.h"
>> +#include "diracdsp.h"
>> +#include "libavutil/common.h"
>
> nit: Move the libavutil #include before the others.
>
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to