On Thu, Apr 18, 2013 at 06:44:51PM +0200, Nicolas Bertrand wrote:
> --- a/libavcodec/jpeg2000.c
> +++ b/libavcodec/jpeg2000.c
> @@ -308,9 +308,12 @@ int ff_jpeg2000_init_component(Jpeg2000Component *comp,
> + /* BITEXACT computing case --> convert to int */
> + if (avctx->flags & CODEC_FLAG_BITEXACT)
> + band->stepsize = (int32_t)(band->stepsize * (1<<16));
spaces around <<
> --- a/libavcodec/jpeg2000dec.c
> +++ b/libavcodec/jpeg2000dec.c
> @@ -266,7 +266,11 @@ static int get_cox(Jpeg2000DecoderContext *s,
> Jpeg2000CodingStyle *c)
> }
> - c->transform = bytestream_get_byte(&s->buf); // transformation
> + c->transform = bytestream_get_byte(&s->buf); // DWT transformation type
> + /* set integer 9/7 DWT in case of BITEXACT flag*/
space before */
> @@ -881,31 +885,96 @@ static int decode_cblk(Jpeg2000DecoderContext *s,
> Jpeg2000CodingStyle *codsty,
>
> +/* TODO: Verify dequantization in case of lossless case
The comment sounds kind of scary :)
> +static void dequantization_float(int x, int y, Jpeg2000Cblk *cblk,
> + Jpeg2000Component *comp,
> + Jpeg2000T1Context *t1, Jpeg2000Band *band) {
{ on the next line
> +static void dequantization_int(int x, int y, Jpeg2000Cblk *cblk,
> + Jpeg2000Component *comp,
> + Jpeg2000T1Context *t1, Jpeg2000Band *band) {
same
> @@ -1034,16 +1095,17 @@ static int
> jpeg2000_decode_tile(Jpeg2000DecoderContext *s, Jpeg2000Tile *tile,
> val = av_clip(val, 0, (1 << s->cbps[compno]) - 1);
> /* align 12 bit values in little-endian mode */
> - val = val << 4;
> - *dst = val;
> - //*datap = (float)val;
> + *dst = (val << 4);
pointless ()
> @@ -1231,7 +1293,6 @@ static int jpeg2000_decode_frame(AVCodecContext *avctx,
> void *data,
>
> return s->buf - s->buf_start;
> }
> -
> #define OFFSET(x) offsetof(Jpeg2000DecoderContext, x)
> #define VD AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_DECODING_PARAM
Is that on purpose? I think it was better before.
> --- a/libavcodec/jpeg2000dwt.c
> +++ b/libavcodec/jpeg2000dwt.c
> @@ -30,17 +30,26 @@
>
> +/* Defines for 9/7 DWT lifting parameters.
> + * Parameters are in float. */
> +#define F_LFTG_ALPHA (1.586134342059924f)
> +#define F_LFTG_BETA (0.052980118572961f)
> +#define F_LFTG_GAMMA (0.882911075530934f)
> +#define F_LFTG_DELTA (0.443506852043971f)
> +#define F_LFTG_K (1.230174104914001f)
> +#define F_LFTG_X (1.625732422f )
> +
> +/* Lifting parameters in integer format.
> + * Computed as param = (float param) * (1 << 16) */
> +#define I_LFTG_ALPHA (103949)
> +#define I_LFTG_BETA ( 3472)
> +#define I_LFTG_GAMMA ( 57862)
> +#define I_LFTG_DELTA ( 29066)
> +#define I_LFTG_K ( 80621)
> +#define I_LFTG_X (106544)
The parentheses are pointless; the defines are values, not expressions.
> +/* FIXME: Why use 1.625732422 instead of 1/F_LFTG_K?
> + * see (ISO/IEC 15444:1 (version 2002) F.3.8.2 */
Hmmmm...
> @@ -231,8 +327,11 @@ int ff_jpeg2000_dwt_init(DWTContext *s, uint16_t
> border[2][2],
> case FF_DWT97:
> s->linebuf = av_malloc((maxlen + 12) * sizeof(float));
> break;
> + case FF_DWT97_INT:
> + s->linebuf = av_malloc((maxlen + 12) * sizeof(int32_t));
> + break;
> case FF_DWT53:
> - s->linebuf = av_malloc((maxlen + 6) * sizeof(int));
> + s->linebuf = av_malloc((maxlen + 6) * sizeof(int32_t));
These sizeofs should better be sizeof(*var) than sizeof(type).
Diego
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
