split mjpeg.c into an encoder and a decoder file

Originally committed as revision 9048 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavcodec/Makefile

@@ -99,13 +99,13 @@ OBJS-$(CONFIG_INTERPLAY_VIDEO_DECODER) += interplayvideo.o
 OBJS-$(CONFIG_INTERPLAY_DPCM_DECODER)  += dpcm.o
 OBJS-$(CONFIG_JPEGLS_ENCODER)          += jpeglsenc.o jpegls.o
 OBJS-$(CONFIG_KMVC_DECODER)            += kmvc.o
-OBJS-$(CONFIG_LJPEG_ENCODER)           += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_LJPEG_ENCODER)           += mjpegenc.o mjpeg.o mpegvideo.o
 OBJS-$(CONFIG_LOCO_DECODER)            += loco.o
 OBJS-$(CONFIG_MACE3_DECODER)           += mace.o
 OBJS-$(CONFIG_MACE6_DECODER)           += mace.o
-OBJS-$(CONFIG_MJPEG_DECODER)           += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_MJPEG_DECODER)           += mjpegdec.o mjpeg.o jpeglsdec.o jpegls.o
-OBJS-$(CONFIG_MJPEG_ENCODER)           += mjpeg.o jpeglsdec.o mpegvideo.o jpegls.o
+OBJS-$(CONFIG_MJPEG_ENCODER)           += mjpegenc.o mjpeg.o mpegvideo.o
-OBJS-$(CONFIG_MJPEGB_DECODER)          += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_MJPEGB_DECODER)          += mjpegdec.o mjpeg.o jpeglsdec.o jpegls.o
 OBJS-$(CONFIG_MMVIDEO_DECODER)         += mmvideo.o
 OBJS-$(CONFIG_MP2_ENCODER)             += mpegaudio.o
 OBJS-$(CONFIG_MPC7_DECODER)            += mpc.o
@@ -153,14 +153,14 @@ OBJS-$(CONFIG_SOL_DPCM_DECODER)        += dpcm.o
 OBJS-$(CONFIG_SONIC_DECODER)           += sonic.o
 OBJS-$(CONFIG_SONIC_ENCODER)           += sonic.o
 OBJS-$(CONFIG_SONIC_LS_DECODER)        += sonic.o
-OBJS-$(CONFIG_SP5X_DECODER)            += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_SP5X_DECODER)            += mjpegdec.o mjpeg.o jpeglsdec.o jpegls.o
 OBJS-$(CONFIG_SVQ1_DECODER)            += svq1.o
 OBJS-$(CONFIG_SVQ1_ENCODER)            += svq1.o
 OBJS-$(CONFIG_SVQ3_DECODER)            += h264.o
 OBJS-$(CONFIG_TARGA_DECODER)           += targa.o
 OBJS-$(CONFIG_TARGA_ENCODER)           += targaenc.o rle.o
 OBJS-$(CONFIG_THEORA_DECODER)          += vp3.o xiph.o vp3dsp.o
-OBJS-$(CONFIG_THP_DECODER)             += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_THP_DECODER)             += mjpegdec.o mjpeg.o jpeglsdec.o jpegls.o
 OBJS-$(CONFIG_TIERTEXSEQVIDEO_DECODER) += tiertexseqv.o
 OBJS-$(CONFIG_TIFF_DECODER)            += tiff.o lzw.o
 OBJS-$(CONFIG_TIFF_ENCODER)            += tiffenc.o rle.o lzwenc.o
@@ -310,7 +310,7 @@ OBJS-$(CONFIG_REMOVE_EXTRADATA_BSF)    += bitstream_filter.o
 OBJS-$(CONFIG_NOISE_BSF)               += bitstream_filter.o
 OBJS-$(CONFIG_MP3_HEADER_COMPRESS_BSF) += bitstream_filter.o
 OBJS-$(CONFIG_MP3_HEADER_DECOMPRESS_BSF) += bitstream_filter.o
-OBJS-$(CONFIG_MJPEGA_DUMP_HEADER_BSF)  += mjpeg.o jpeglsdec.o jpegls.o
+OBJS-$(CONFIG_MJPEGA_DUMP_HEADER_BSF)  += mjpegdec.o mjpeg.o jpeglsdec.o jpegls.o
 OBJS-$(CONFIG_IMX_DUMP_HEADER_BSF)     += mpeg12.o
 
 OBJS-$(HAVE_PTHREADS)                  += pthread.o

libavcodec/mjpeg.h

@@ -192,4 +192,31 @@ typedef struct MJpegDecodeContext {
     int cur_scan; /* current scan, used by JPEG-LS */
 } MJpegDecodeContext;
 
+#define PREDICT(ret, topleft, top, left, predictor)\
+    switch(predictor){\
+        case 1: ret= left; break;\
+        case 2: ret= top; break;\
+        case 3: ret= topleft; break;\
+        case 4: ret= left   +   top - topleft; break;\
+        case 5: ret= left   + ((top - topleft)>>1); break;\
+        case 6: ret= top + ((left   - topleft)>>1); break;\
+        default:\
+        case 7: ret= (left + top)>>1; break;\
+    }
+
+extern const uint8_t bits_dc_luminance[];
+extern const uint8_t val_dc_luminance[];
+
+extern const uint8_t bits_dc_chrominance[];
+extern const uint8_t val_dc_chrominance[];
+
+extern const uint8_t bits_ac_luminance[];
+extern const uint8_t val_ac_luminance[];
+
+extern const uint8_t bits_ac_chrominance[];
+extern const uint8_t val_ac_chrominance[];
+
+void build_huffman_codes(uint8_t *huff_size, uint16_t *huff_code,
+                         const uint8_t *bits_table, const uint8_t *val_table);
+
 #endif /* MJPEG_H */

libavcodec/mjpegenc.c

@@ -0,0 +1,617 @@
+/*
+ * MJPEG encoder
+ * Copyright (c) 2000, 2001 Fabrice Bellard.
+ * Copyright (c) 2003 Alex Beregszaszi
+ * Copyright (c) 2003-2004 Michael Niedermayer
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ *
+ * Support for external huffman table, various fixes (AVID workaround),
+ * aspecting, new decode_frame mechanism and apple mjpeg-b support
+ *                                  by Alex Beregszaszi
+ */
+
+/**
+ * @file mjpegenc.c
+ * MJPEG encoder.
+ */
+
+//#define DEBUG
+#include <assert.h>
+
+#include "avcodec.h"
+#include "dsputil.h"
+#include "mpegvideo.h"
+#include "mjpeg.h"
+
+/* use two quantizer tables (one for luminance and one for chrominance) */
+/* not yet working */
+#undef TWOMATRIXES
+
+typedef struct MJpegContext {
+    uint8_t huff_size_dc_luminance[12]; //FIXME use array [3] instead of lumi / chrom, for easier addressing
+    uint16_t huff_code_dc_luminance[12];
+    uint8_t huff_size_dc_chrominance[12];
+    uint16_t huff_code_dc_chrominance[12];
+
+    uint8_t huff_size_ac_luminance[256];
+    uint16_t huff_code_ac_luminance[256];
+    uint8_t huff_size_ac_chrominance[256];
+    uint16_t huff_code_ac_chrominance[256];
+} MJpegContext;
+
+
+int mjpeg_init(MpegEncContext *s)
+{
+    MJpegContext *m;
+
+    m = av_malloc(sizeof(MJpegContext));
+    if (!m)
+        return -1;
+
+    s->min_qcoeff=-1023;
+    s->max_qcoeff= 1023;
+
+    /* build all the huffman tables */
+    build_huffman_codes(m->huff_size_dc_luminance,
+                        m->huff_code_dc_luminance,
+                        bits_dc_luminance,
+                        val_dc_luminance);
+    build_huffman_codes(m->huff_size_dc_chrominance,
+                        m->huff_code_dc_chrominance,
+                        bits_dc_chrominance,
+                        val_dc_chrominance);
+    build_huffman_codes(m->huff_size_ac_luminance,
+                        m->huff_code_ac_luminance,
+                        bits_ac_luminance,
+                        val_ac_luminance);
+    build_huffman_codes(m->huff_size_ac_chrominance,
+                        m->huff_code_ac_chrominance,
+                        bits_ac_chrominance,
+                        val_ac_chrominance);
+
+    s->mjpeg_ctx = m;
+    return 0;
+}
+
+void mjpeg_close(MpegEncContext *s)
+{
+    av_free(s->mjpeg_ctx);
+}
+
+/* table_class: 0 = DC coef, 1 = AC coefs */
+static int put_huffman_table(MpegEncContext *s, int table_class, int table_id,
+                             const uint8_t *bits_table, const uint8_t *value_table)
+{
+    PutBitContext *p = &s->pb;
+    int n, i;
+
+    put_bits(p, 4, table_class);
+    put_bits(p, 4, table_id);
+
+    n = 0;
+    for(i=1;i<=16;i++) {
+        n += bits_table[i];
+        put_bits(p, 8, bits_table[i]);
+    }
+
+    for(i=0;i<n;i++)
+        put_bits(p, 8, value_table[i]);
+
+    return n + 17;
+}
+
+static void jpeg_table_header(MpegEncContext *s)
+{
+    PutBitContext *p = &s->pb;
+    int i, j, size;
+    uint8_t *ptr;
+
+    /* quant matrixes */
+    put_marker(p, DQT);
+#ifdef TWOMATRIXES
+    put_bits(p, 16, 2 + 2 * (1 + 64));
+#else
+    put_bits(p, 16, 2 + 1 * (1 + 64));
+#endif
+    put_bits(p, 4, 0); /* 8 bit precision */
+    put_bits(p, 4, 0); /* table 0 */
+    for(i=0;i<64;i++) {
+        j = s->intra_scantable.permutated[i];
+        put_bits(p, 8, s->intra_matrix[j]);
+    }
+#ifdef TWOMATRIXES
+    put_bits(p, 4, 0); /* 8 bit precision */
+    put_bits(p, 4, 1); /* table 1 */
+    for(i=0;i<64;i++) {
+        j = s->intra_scantable.permutated[i];
+        put_bits(p, 8, s->chroma_intra_matrix[j]);
+    }
+#endif
+
+    /* huffman table */
+    put_marker(p, DHT);
+    flush_put_bits(p);
+    ptr = pbBufPtr(p);
+    put_bits(p, 16, 0); /* patched later */
+    size = 2;
+    size += put_huffman_table(s, 0, 0, bits_dc_luminance, val_dc_luminance);
+    size += put_huffman_table(s, 0, 1, bits_dc_chrominance, val_dc_chrominance);
+
+    size += put_huffman_table(s, 1, 0, bits_ac_luminance, val_ac_luminance);
+    size += put_huffman_table(s, 1, 1, bits_ac_chrominance, val_ac_chrominance);
+    ptr[0] = size >> 8;
+    ptr[1] = size;
+}
+
+static void jpeg_put_comments(MpegEncContext *s)
+{
+    PutBitContext *p = &s->pb;
+    int size;
+    uint8_t *ptr;
+
+    if (s->aspect_ratio_info /* && !lossless */)
+    {
+    /* JFIF header */
+    put_marker(p, APP0);
+    put_bits(p, 16, 16);
+    ff_put_string(p, "JFIF", 1); /* this puts the trailing zero-byte too */
+    put_bits(p, 16, 0x0201); /* v 1.02 */
+    put_bits(p, 8, 0); /* units type: 0 - aspect ratio */
+    put_bits(p, 16, s->avctx->sample_aspect_ratio.num);
+    put_bits(p, 16, s->avctx->sample_aspect_ratio.den);
+    put_bits(p, 8, 0); /* thumbnail width */
+    put_bits(p, 8, 0); /* thumbnail height */
+    }
+
+    /* comment */
+    if(!(s->flags & CODEC_FLAG_BITEXACT)){
+        put_marker(p, COM);
+        flush_put_bits(p);
+        ptr = pbBufPtr(p);
+        put_bits(p, 16, 0); /* patched later */
+        ff_put_string(p, LIBAVCODEC_IDENT, 1);
+        size = strlen(LIBAVCODEC_IDENT)+3;
+        ptr[0] = size >> 8;
+        ptr[1] = size;
+    }
+
+    if(  s->avctx->pix_fmt == PIX_FMT_YUV420P
+       ||s->avctx->pix_fmt == PIX_FMT_YUV422P
+       ||s->avctx->pix_fmt == PIX_FMT_YUV444P){
+        put_marker(p, COM);
+        flush_put_bits(p);
+        ptr = pbBufPtr(p);
+        put_bits(p, 16, 0); /* patched later */
+        ff_put_string(p, "CS=ITU601", 1);
+        size = strlen("CS=ITU601")+3;
+        ptr[0] = size >> 8;
+        ptr[1] = size;
+    }
+}
+
+void mjpeg_picture_header(MpegEncContext *s)
+{
+    const int lossless= s->avctx->codec_id != CODEC_ID_MJPEG;
+
+    put_marker(&s->pb, SOI);
+
+    jpeg_put_comments(s);
+
+    jpeg_table_header(s);
+
+    switch(s->avctx->codec_id){
+    case CODEC_ID_MJPEG:  put_marker(&s->pb, SOF0 ); break;
+    case CODEC_ID_LJPEG:  put_marker(&s->pb, SOF3 ); break;
+    default: assert(0);
+    }
+
+    put_bits(&s->pb, 16, 17);
+    if(lossless && s->avctx->pix_fmt == PIX_FMT_RGB32)
+        put_bits(&s->pb, 8, 9); /* 9 bits/component RCT */
+    else
+        put_bits(&s->pb, 8, 8); /* 8 bits/component */
+    put_bits(&s->pb, 16, s->height);
+    put_bits(&s->pb, 16, s->width);
+    put_bits(&s->pb, 8, 3); /* 3 components */
+
+    /* Y component */
+    put_bits(&s->pb, 8, 1); /* component number */
+    put_bits(&s->pb, 4, s->mjpeg_hsample[0]); /* H factor */
+    put_bits(&s->pb, 4, s->mjpeg_vsample[0]); /* V factor */
+    put_bits(&s->pb, 8, 0); /* select matrix */
+
+    /* Cb component */
+    put_bits(&s->pb, 8, 2); /* component number */
+    put_bits(&s->pb, 4, s->mjpeg_hsample[1]); /* H factor */
+    put_bits(&s->pb, 4, s->mjpeg_vsample[1]); /* V factor */
+#ifdef TWOMATRIXES
+    put_bits(&s->pb, 8, lossless ? 0 : 1); /* select matrix */
+#else
+    put_bits(&s->pb, 8, 0); /* select matrix */
+#endif
+
+    /* Cr component */
+    put_bits(&s->pb, 8, 3); /* component number */
+    put_bits(&s->pb, 4, s->mjpeg_hsample[2]); /* H factor */
+    put_bits(&s->pb, 4, s->mjpeg_vsample[2]); /* V factor */
+#ifdef TWOMATRIXES
+    put_bits(&s->pb, 8, lossless ? 0 : 1); /* select matrix */
+#else
+    put_bits(&s->pb, 8, 0); /* select matrix */
+#endif
+
+    /* scan header */
+    put_marker(&s->pb, SOS);
+    put_bits(&s->pb, 16, 12); /* length */
+    put_bits(&s->pb, 8, 3); /* 3 components */
+
+    /* Y component */
+    put_bits(&s->pb, 8, 1); /* index */
+    put_bits(&s->pb, 4, 0); /* DC huffman table index */
+    put_bits(&s->pb, 4, 0); /* AC huffman table index */
+
+    /* Cb component */
+    put_bits(&s->pb, 8, 2); /* index */
+    put_bits(&s->pb, 4, 1); /* DC huffman table index */
+    put_bits(&s->pb, 4, lossless ? 0 : 1); /* AC huffman table index */
+
+    /* Cr component */
+    put_bits(&s->pb, 8, 3); /* index */
+    put_bits(&s->pb, 4, 1); /* DC huffman table index */
+    put_bits(&s->pb, 4, lossless ? 0 : 1); /* AC huffman table index */
+
+    put_bits(&s->pb, 8, lossless ? s->avctx->prediction_method+1 : 0); /* Ss (not used) */
+
+    switch(s->avctx->codec_id){
+    case CODEC_ID_MJPEG:  put_bits(&s->pb, 8, 63); break; /* Se (not used) */
+    case CODEC_ID_LJPEG:  put_bits(&s->pb, 8,  0); break; /* not used */
+    default: assert(0);
+    }
+
+    put_bits(&s->pb, 8, 0); /* Ah/Al (not used) */
+}
+
+static void escape_FF(MpegEncContext *s, int start)
+{
+    int size= put_bits_count(&s->pb) - start*8;
+    int i, ff_count;
+    uint8_t *buf= s->pb.buf + start;
+    int align= (-(size_t)(buf))&3;
+
+    assert((size&7) == 0);
+    size >>= 3;
+
+    ff_count=0;
+    for(i=0; i<size && i<align; i++){
+        if(buf[i]==0xFF) ff_count++;
+    }
+    for(; i<size-15; i+=16){
+        int acc, v;
+
+        v= *(uint32_t*)(&buf[i]);
+        acc= (((v & (v>>4))&0x0F0F0F0F)+0x01010101)&0x10101010;
+        v= *(uint32_t*)(&buf[i+4]);
+        acc+=(((v & (v>>4))&0x0F0F0F0F)+0x01010101)&0x10101010;
+        v= *(uint32_t*)(&buf[i+8]);
+        acc+=(((v & (v>>4))&0x0F0F0F0F)+0x01010101)&0x10101010;
+        v= *(uint32_t*)(&buf[i+12]);
+        acc+=(((v & (v>>4))&0x0F0F0F0F)+0x01010101)&0x10101010;
+
+        acc>>=4;
+        acc+= (acc>>16);
+        acc+= (acc>>8);
+        ff_count+= acc&0xFF;
+    }
+    for(; i<size; i++){
+        if(buf[i]==0xFF) ff_count++;
+    }
+
+    if(ff_count==0) return;
+
+    /* skip put bits */
+    for(i=0; i<ff_count-3; i+=4)
+        put_bits(&s->pb, 32, 0);
+    put_bits(&s->pb, (ff_count-i)*8, 0);
+    flush_put_bits(&s->pb);
+
+    for(i=size-1; ff_count; i--){
+        int v= buf[i];
+
+        if(v==0xFF){
+//printf("%d %d\n", i, ff_count);
+            buf[i+ff_count]= 0;
+            ff_count--;
+        }
+
+        buf[i+ff_count]= v;
+    }
+}
+
+void ff_mjpeg_stuffing(PutBitContext * pbc)
+{
+    int length;
+    length= (-put_bits_count(pbc))&7;
+    if(length) put_bits(pbc, length, (1<<length)-1);
+}
+
+void mjpeg_picture_trailer(MpegEncContext *s)
+{
+    ff_mjpeg_stuffing(&s->pb);
+    flush_put_bits(&s->pb);
+
+    assert((s->header_bits&7)==0);
+
+    escape_FF(s, s->header_bits>>3);
+
+    put_marker(&s->pb, EOI);
+}
+
+static inline void mjpeg_encode_dc(MpegEncContext *s, int val,
+                                   uint8_t *huff_size, uint16_t *huff_code)
+{
+    int mant, nbits;
+
+    if (val == 0) {
+        put_bits(&s->pb, huff_size[0], huff_code[0]);
+    } else {
+        mant = val;
+        if (val < 0) {
+            val = -val;
+            mant--;
+        }
+
+        nbits= av_log2_16bit(val) + 1;
+
+        put_bits(&s->pb, huff_size[nbits], huff_code[nbits]);
+
+        put_bits(&s->pb, nbits, mant & ((1 << nbits) - 1));
+    }
+}
+
+static void encode_block(MpegEncContext *s, DCTELEM *block, int n)
+{
+    int mant, nbits, code, i, j;
+    int component, dc, run, last_index, val;
+    MJpegContext *m = s->mjpeg_ctx;
+    uint8_t *huff_size_ac;
+    uint16_t *huff_code_ac;
+
+    /* DC coef */
+    component = (n <= 3 ? 0 : (n&1) + 1);
+    dc = block[0]; /* overflow is impossible */
+    val = dc - s->last_dc[component];
+    if (n < 4) {
+        mjpeg_encode_dc(s, val, m->huff_size_dc_luminance, m->huff_code_dc_luminance);
+        huff_size_ac = m->huff_size_ac_luminance;
+        huff_code_ac = m->huff_code_ac_luminance;
+    } else {
+        mjpeg_encode_dc(s, val, m->huff_size_dc_chrominance, m->huff_code_dc_chrominance);
+        huff_size_ac = m->huff_size_ac_chrominance;
+        huff_code_ac = m->huff_code_ac_chrominance;
+    }
+    s->last_dc[component] = dc;
+
+    /* AC coefs */
+
+    run = 0;
+    last_index = s->block_last_index[n];
+    for(i=1;i<=last_index;i++) {
+        j = s->intra_scantable.permutated[i];
+        val = block[j];
+        if (val == 0) {
+            run++;
+        } else {
+            while (run >= 16) {
+                put_bits(&s->pb, huff_size_ac[0xf0], huff_code_ac[0xf0]);
+                run -= 16;
+            }
+            mant = val;
+            if (val < 0) {
+                val = -val;
+                mant--;
+            }
+
+            nbits= av_log2(val) + 1;
+            code = (run << 4) | nbits;
+
+            put_bits(&s->pb, huff_size_ac[code], huff_code_ac[code]);
+
+            put_bits(&s->pb, nbits, mant & ((1 << nbits) - 1));
+            run = 0;
+        }
+    }
+
+    /* output EOB only if not already 64 values */
+    if (last_index < 63 || run != 0)
+        put_bits(&s->pb, huff_size_ac[0], huff_code_ac[0]);
+}
+
+void mjpeg_encode_mb(MpegEncContext *s,
+                     DCTELEM block[6][64])
+{
+    int i;
+    for(i=0;i<5;i++) {
+        encode_block(s, block[i], i);
+    }
+    if (s->chroma_format == CHROMA_420) {
+        encode_block(s, block[5], 5);
+    } else {
+        encode_block(s, block[6], 6);
+        encode_block(s, block[5], 5);
+        encode_block(s, block[7], 7);
+    }
+}
+
+static int encode_picture_lossless(AVCodecContext *avctx, unsigned char *buf, int buf_size, void *data){
+    MpegEncContext * const s = avctx->priv_data;
+    MJpegContext * const m = s->mjpeg_ctx;
+    AVFrame *pict = data;
+    const int width= s->width;
+    const int height= s->height;
+    AVFrame * const p= (AVFrame*)&s->current_picture;
+    const int predictor= avctx->prediction_method+1;
+
+    init_put_bits(&s->pb, buf, buf_size);
+
+    *p = *pict;
+    p->pict_type= FF_I_TYPE;
+    p->key_frame= 1;
+
+    mjpeg_picture_header(s);
+
+    s->header_bits= put_bits_count(&s->pb);
+
+    if(avctx->pix_fmt == PIX_FMT_RGB32){
+        int x, y, i;
+        const int linesize= p->linesize[0];
+        uint16_t (*buffer)[4]= (void *) s->rd_scratchpad;
+        int left[3], top[3], topleft[3];
+
+        for(i=0; i<3; i++){
+            buffer[0][i]= 1 << (9 - 1);
+        }
+
+        for(y = 0; y < height; y++) {
+            const int modified_predictor= y ? predictor : 1;
+            uint8_t *ptr = p->data[0] + (linesize * y);
+
+            if(s->pb.buf_end - s->pb.buf - (put_bits_count(&s->pb)>>3) < width*3*4){
+                av_log(s->avctx, AV_LOG_ERROR, "encoded frame too large\n");
+                return -1;
+            }
+
+            for(i=0; i<3; i++){
+                top[i]= left[i]= topleft[i]= buffer[0][i];
+            }
+            for(x = 0; x < width; x++) {
+                buffer[x][1] = ptr[4*x+0] - ptr[4*x+1] + 0x100;
+                buffer[x][2] = ptr[4*x+2] - ptr[4*x+1] + 0x100;
+                buffer[x][0] = (ptr[4*x+0] + 2*ptr[4*x+1] + ptr[4*x+2])>>2;
+
+                for(i=0;i<3;i++) {
+                    int pred, diff;
+
+                    PREDICT(pred, topleft[i], top[i], left[i], modified_predictor);
+
+                    topleft[i]= top[i];
+                    top[i]= buffer[x+1][i];
+
+                    left[i]= buffer[x][i];
+
+                    diff= ((left[i] - pred + 0x100)&0x1FF) - 0x100;
+
+                    if(i==0)
+                        mjpeg_encode_dc(s, diff, m->huff_size_dc_luminance, m->huff_code_dc_luminance); //FIXME ugly
+                    else
+                        mjpeg_encode_dc(s, diff, m->huff_size_dc_chrominance, m->huff_code_dc_chrominance);
+                }
+            }
+        }
+    }else{
+        int mb_x, mb_y, i;
+        const int mb_width  = (width  + s->mjpeg_hsample[0] - 1) / s->mjpeg_hsample[0];
+        const int mb_height = (height + s->mjpeg_vsample[0] - 1) / s->mjpeg_vsample[0];
+
+        for(mb_y = 0; mb_y < mb_height; mb_y++) {
+            if(s->pb.buf_end - s->pb.buf - (put_bits_count(&s->pb)>>3) < mb_width * 4 * 3 * s->mjpeg_hsample[0] * s->mjpeg_vsample[0]){
+                av_log(s->avctx, AV_LOG_ERROR, "encoded frame too large\n");
+                return -1;
+            }
+            for(mb_x = 0; mb_x < mb_width; mb_x++) {
+                if(mb_x==0 || mb_y==0){
+                    for(i=0;i<3;i++) {
+                        uint8_t *ptr;
+                        int x, y, h, v, linesize;
+                        h = s->mjpeg_hsample[i];
+                        v = s->mjpeg_vsample[i];
+                        linesize= p->linesize[i];
+
+                        for(y=0; y<v; y++){
+                            for(x=0; x<h; x++){
+                                int pred;
+
+                                ptr = p->data[i] + (linesize * (v * mb_y + y)) + (h * mb_x + x); //FIXME optimize this crap
+                                if(y==0 && mb_y==0){
+                                    if(x==0 && mb_x==0){
+                                        pred= 128;
+                                    }else{
+                                        pred= ptr[-1];
+                                    }
+                                }else{
+                                    if(x==0 && mb_x==0){
+                                        pred= ptr[-linesize];
+                                    }else{
+                                        PREDICT(pred, ptr[-linesize-1], ptr[-linesize], ptr[-1], predictor);
+                                    }
+                                }
+
+                                if(i==0)
+                                    mjpeg_encode_dc(s, (int8_t)(*ptr - pred), m->huff_size_dc_luminance, m->huff_code_dc_luminance); //FIXME ugly
+                                else
+                                    mjpeg_encode_dc(s, (int8_t)(*ptr - pred), m->huff_size_dc_chrominance, m->huff_code_dc_chrominance);
+                            }
+                        }
+                    }
+                }else{
+                    for(i=0;i<3;i++) {
+                        uint8_t *ptr;
+                        int x, y, h, v, linesize;
+                        h = s->mjpeg_hsample[i];
+                        v = s->mjpeg_vsample[i];
+                        linesize= p->linesize[i];
+
+                        for(y=0; y<v; y++){
+                            for(x=0; x<h; x++){
+                                int pred;
+
+                                ptr = p->data[i] + (linesize * (v * mb_y + y)) + (h * mb_x + x); //FIXME optimize this crap
+//printf("%d %d %d %d %8X\n", mb_x, mb_y, x, y, ptr);
+                                PREDICT(pred, ptr[-linesize-1], ptr[-linesize], ptr[-1], predictor);
+
+                                if(i==0)
+                                    mjpeg_encode_dc(s, (int8_t)(*ptr - pred), m->huff_size_dc_luminance, m->huff_code_dc_luminance); //FIXME ugly
+                                else
+                                    mjpeg_encode_dc(s, (int8_t)(*ptr - pred), m->huff_size_dc_chrominance, m->huff_code_dc_chrominance);
+                            }
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+    emms_c();
+
+    mjpeg_picture_trailer(s);
+    s->picture_number++;
+
+    flush_put_bits(&s->pb);
+    return pbBufPtr(&s->pb) - s->pb.buf;
+//    return (put_bits_count(&f->pb)+7)/8;
+}
+
+
+AVCodec ljpeg_encoder = { //FIXME avoid MPV_* lossless jpeg shouldnt need them
+    "ljpeg",
+    CODEC_TYPE_VIDEO,
+    CODEC_ID_LJPEG,
+    sizeof(MpegEncContext),
+    MPV_encode_init,
+    encode_picture_lossless,
+    MPV_encode_end,
+};