On Fri, Mar 15, 2013 at 03:35:14PM +0100, Nicolas Bertrand wrote:
> --- a/libavcodec/avcodec.h
> +++ b/libavcodec/avcodec.h
> @@ -2561,6 +2561,12 @@ typedef struct AVCodecContext {
> #define FF_PROFILE_MPEG4_SIMPLE_STUDIO 14
> #define FF_PROFILE_MPEG4_ADVANCED_SIMPLE 15
>
> +#define FF_PROFILE_JPEG2000_CSTRM_RSTRCTN_0 0
> +#define FF_PROFILE_JPEG2000_CSTRM_RSTRCTN_1 1
> +#define FF_PROFILE_JPEG2000_CSTRM_NO_RSTRCTN 2
> +#define FF_PROFILE_JPEG2000_DCINEMA_2K 3
> +#define FF_PROFILE_JPEG2000_DCINEMA_4K 4
Don't we have a way to export these per-codec? Anton?
> --- /dev/null
> +++ b/libavcodec/jpeg2000.c
> @@ -0,0 +1,475 @@
> +
> +/* allocate the memory for tag tree */
> +static int32_t tag_tree_size(uint16_t w, uint16_t h)
> +{
> + uint32_t res = 0;
> + while (w > 1 || h > 1) {
> + res += w * h;
> + if ( (res + 1) >= INT32_MAX)
stray space inside (), pointless ()
> + static const int contribtab[3][3] = { { 0, -1, 1 }, { -1, -1, 0 }, { 1,
> 0, 1 } };
> + static const int ctxlbltab[3][3] = { { 13, 12, 11 }, { 10, 9, 10 }, {
> 11, 12, 13 } };
> + static const int xorbittab[3][3] = { { 1, 1, 1, }, { 1, 0, 0 }, { 0, 0,
> 0 } };
Vertical alignment could make these more readable.
> + log2_band_prec_width = reslevel->log2_prec_width - 1;
> + log2_band_prec_height = reslevel->log2_prec_height - 1;
align
> --- /dev/null
> +++ b/libavcodec/jpeg2000.h
> @@ -0,0 +1,256 @@
> +
> +/* TODO: data can be float of integer depending of reversible/irreversible
> + * transformation.
or?
> --- /dev/null
> +++ b/libavcodec/jpeg2000dec.c
> @@ -0,0 +1,1253 @@
> +static int get_siz(Jpeg2000DecoderContext *s)
> +
> + s->avctx->profile = bytestream_get_be16(&s->buf); // Rsiz
> + s->width = bytestream_get_be32(&s->buf); // width
> + s->height = bytestream_get_be32(&s->buf); // height
> + s->image_offset_x = bytestream_get_be32(&s->buf); // X0Siz
> + s->image_offset_y = bytestream_get_be32(&s->buf); // Y0Siz
align
> + // check if a precinct exist
exist_S_
> + prcx = ff_jpeg2000_ceildivpow2(x,reducedresno) >>
> rlevel->log2_prec_width;
> + prcy = ff_jpeg2000_ceildivpow2(y,reducedresno) >>
> rlevel->log2_prec_height;
> + precno = prcx + rlevel->num_precincts_x * prcy;
> + for (layno = 0; layno < tile->codsty[0].nlayers;
> layno++) {
> + if (jpeg2000_decode_packet( s,
> + codsty, rlevel,
> + precno, layno,
> + qntsty->expn +
> (reslevelno ? 3 * (reslevelno - 1) + 1 : 0),
> + qntsty->nguardbits)
> + )
Ugh, no, that's ugly formatting.
> +static void decode_clnpass(Jpeg2000DecoderContext *s, Jpeg2000T1Context *t1,
> + int width, int height, int bpno, int bandno,
> + int seg_symbols)
> +{
> + if (seg_symbols) {
> + int val;
> + val = ff_mqc_decode(&t1->mqc, t1->mqc.cx_states + MQC_CX_UNI);
> + val = (val << 1) + ff_mqc_decode(&t1->mqc, t1->mqc.cx_states +
> MQC_CX_UNI);
> + val = (val << 1) + ff_mqc_decode(&t1->mqc, t1->mqc.cx_states +
> MQC_CX_UNI);
> + val = (val << 1) + ff_mqc_decode(&t1->mqc, t1->mqc.cx_states +
> MQC_CX_UNI);
> + if (val != 0xa) {
> + av_log(s->avctx, AV_LOG_ERROR, "Segmentation symbol value
> incorrect\n");
> + }
nit: pointless {}
> +static int decode_cblk(Jpeg2000DecoderContext *s, Jpeg2000CodingStyle
> *codsty,
> + Jpeg2000T1Context *t1, Jpeg2000Cblk *cblk,
> + int width, int height, int bandpos)
> +{
> + int passno = cblk->npasses, pass_t = 2, bpno = cblk->nonzerobits - 1, y;
> +
> + for (y = 0; y < height + 2; y++)
> + memset(t1->flags[y], 0, (width + 2) * sizeof(int));
> +
> + for (y = 0; y < height; y++)
> + memset(t1->data[y], 0, width * sizeof(int));
Please replace sizeof(type) by sizeof(*var) - here and everywhere else.
> +static int jpeg2000_decode_tile(Jpeg2000DecoderContext *s, Jpeg2000Tile
> *tile, AVFrame *picture)
> +{
> + /* TODO: only the case or irreversible transform is
> implemented (float)
> + * comp->data can be float or int regarding the type
> of transformation
> + * see ISO/IEC 15444-1:2002 §A.6.1
> + *
> + */
nit: Skip the two empty lines.
> +static int jpeg2000_decode_frame(AVCodecContext *avctx, void *data,
> + int *got_frame, AVPacket *avpkt)
> +{
> + s->avctx = avctx;
> + // init
> + s->buf = s->buf_start = avpkt->data;
> + s->buf_end = s->buf_start + avpkt->size;
pointless comment
> + // FIXME: add verficiation accodeing number of resolution levels
> + s->reduction_factor = s->lowres;
multiple typos
> --- /dev/null
> +++ b/libavcodec/jpeg2000dwt.c
> @@ -0,0 +1,250 @@
> + if (type == FF_DWT97)
> + s->linebuf = av_malloc((maxlen + 12) * sizeof(float));
> + else if (type == FF_DWT53)
> + s->linebuf = av_malloc((maxlen + 6) * sizeof(int));
> + else
> + return -1;
switch/case?
> --- /dev/null
> +++ b/libavcodec/jpeg2000dwt.h
> @@ -0,0 +1,62 @@
> +
> +#include "avcodec.h"
Just stdint.h should be enough.
> +typedef struct {
> + /// line lengths { horizontal, vertical } in consecutive decomposition
> levels
> + uint16_t linelen[FF_DWT_MAX_DECLVLS][2];
> + uint8_t mod[FF_DWT_MAX_DECLVLS][2]; ///< coordinates (x0, y0) of
> decomp. levels mod 2
> + uint8_t ndeclevels; ///< number of decomposition levels
> + uint8_t type; ///< 0 for 9/7; 1 for 5/3
> + void *linebuf; ///< buffer used by transform (int
> or float)
> +} DWTContext;
no anonymous structs (in headers)
> +/**
> + * initialize DWT
Capitalize and end in a period.
> + * @param s DWT context
> + * @param border coordinates of transformed region {{x0, x1}, {y0, y1}}
> + * @param decomp_levels number of decomposition levels
> + * @param type 0 for DWT 9/7; 1 for DWT 5/3
Please vertically align the parameter descriptions.
> --- /dev/null
> +++ b/libavcodec/mqc.c
> @@ -0,0 +1,111 @@
> +
> +typedef struct {
> + uint16_t qe;
> + uint8_t nmps;
> + uint8_t nlps;
> + uint8_t sw;
> +} MqcCxState;
no anonymous structs please
> --- /dev/null
> +++ b/libavcodec/mqc.h
> @@ -0,0 +1,61 @@
> +
> +#include "avcodec.h"
stdint.h, rather
> +extern uint16_t ff_mqc_qe[2*47];
> +extern uint8_t ff_mqc_nlps[2*47];
> +extern uint8_t ff_mqc_nmps[2*47];
spaces around *
> +typedef struct {
> + uint8_t *bp, *bpstart;
> + unsigned int a;
> + unsigned int c;
> + unsigned int ct;
> + uint8_t cx_states[19];
> +} MqcState;
no anonymous structs (in headers)
> +/* decoder */
> +
> +/** initialize the decoder */
> +void ff_mqc_initdec(MqcState *mqc, uint8_t *bp);
> +
> +/* common */
> +
> +/** initialize the contexts */
> +void ff_mqc_init_contexts(MqcState *mqc);
These are pointless comments, Doxygen to boot.
> --- /dev/null
> +++ b/libavcodec/mqcdec.c
> @@ -0,0 +1,93 @@
> +
> +#include "mqc.h"
> +
> +static void bytein(MqcState *mqc)
> +{
> + if (*mqc->bp == 0xff){
> + if (*(mqc->bp+1) > 0x8f)
spaces around +
> +static int exchange(MqcState *mqc, uint8_t *cxstate, int lps)
> +{
> + } else{
space before {
> + // renormd:
What?
> + do{
space before {
> +void ff_mqc_initdec(MqcState *mqc, uint8_t *bp)
> +{
> + ff_mqc_init_contexts(mqc);
> + mqc->bp = bp;
> + mqc->c = (*mqc->bp ^ 0xff) << 16;
align
> --- a/libavformat/img2.c
> +++ b/libavformat/img2.c
> @@ -62,6 +62,7 @@ static const IdStrMap img_tags[] = {
> { AV_CODEC_ID_SUNRAST, "sunras" },
> { AV_CODEC_ID_JPEG2000, "jp2" },
> { AV_CODEC_ID_JPEG2000, "jpc" },
> + { AV_CODEC_ID_JPEG2000, "j2k" },
> { AV_CODEC_ID_DPX, "dpx" },
> { AV_CODEC_ID_PICTOR, "pic" },
> { AV_CODEC_ID_XBM, "xbm" },
This could probably go in separately, right away.
Diego
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
