On Thu, May 16, 2013 at 11:04:08AM +0300, Martin Storsjö wrote:
> On Thu, 16 May 2013, Evert Taube wrote:
>
> >Vad ska ni säga om det?
> >
>
> >From b16ca98863708da7efa83dcceba9faaa49a4255b Mon Sep 17 00:00:00 2001
> >From: Evert Taube <[email protected]>
> >Date: Thu, 16 May 2013 07:21:08 +0200
> >Subject: [PATCH] Apple Intermediate Codec decoder
>
> Is the pseudonym really necessary here? Yes, it's a good joke though. :-)
It's a tradition ;)
> >
> >---
> > Changelog | 1 +
> > configure | 1 +
> > doc/general.texi | 1 +
> > libavcodec/Makefile | 1 +
> > libavcodec/aic.c | 477
> > +++++++++++++++++++++++++++++++++++++++++++++++
> > libavcodec/allcodecs.c | 1 +
> > libavcodec/avcodec.h | 1 +
> > libavcodec/codec_desc.c | 7 +
> > libavcodec/version.h | 2 +-
> > libavformat/isom.c | 2 +
> > 10 files changed, 493 insertions(+), 1 deletion(-)
> > create mode 100644 libavcodec/aic.c
> >
> >diff --git a/Changelog b/Changelog
> >index 97e7a32..64d8a00 100644
> >--- a/Changelog
> >+++ b/Changelog
> >@@ -18,6 +18,7 @@ version 10:
> > transcoding audio
> > - Matroska muxer can now put the index at the beginning of the file.
> > - avconv -deinterlace option removed, the yadif filter should be used
> > instead
> >+- Apple Intermediate Codec decoder
> >
> >
> > version 9:
> >diff --git a/configure b/configure
> >index c84050c..1668fac 100755
> >--- a/configure
> >+++ b/configure
> >@@ -1526,6 +1526,7 @@ aac_latm_decoder_select="aac_decoder aac_latm_parser"
> > ac3_decoder_select="mdct ac3dsp ac3_parser dsputil"
> > ac3_encoder_select="mdct ac3dsp dsputil"
> > ac3_fixed_encoder_select="mdct ac3dsp dsputil"
> >+aic_decoder_select="dsputil"
> > alac_encoder_select="lpc"
> > als_decoder_select="dsputil"
> > amrnb_decoder_select="lsp"
>
> I hope you did test standalone compilation to check that this really
> is enough
>
> >diff --git a/doc/general.texi b/doc/general.texi
> >index ee87e9f..6a049f4 100644
> >--- a/doc/general.texi
> >+++ b/doc/general.texi
> >@@ -430,6 +430,7 @@ following image formats are supported:
> > @item AMV Video @tab @tab X
> > @tab Used in Chinese MP3 players.
> > @item ANSI/ASCII art @tab @tab X
> >+@item Apple Intermediate Codec @tab @tab X
> > @item Apple MJPEG-B @tab @tab X
> > @item Apple ProRes @tab X @tab X
> > @item Apple QuickDraw @tab @tab X
> >diff --git a/libavcodec/Makefile b/libavcodec/Makefile
> >index 2c3522e..85e430f 100644
> >--- a/libavcodec/Makefile
> >+++ b/libavcodec/Makefile
> >@@ -83,6 +83,7 @@ OBJS-$(CONFIG_AC3_DECODER) += ac3dec.o
> >ac3dec_data.o ac3.o kbdwin.o
> > OBJS-$(CONFIG_AC3_ENCODER) += ac3enc_float.o ac3enc.o ac3tab.o \
> > ac3.o kbdwin.o
> > OBJS-$(CONFIG_AC3_FIXED_ENCODER) += ac3enc_fixed.o ac3enc.o ac3tab.o
> > ac3.o
> >+OBJS-$(CONFIG_AIC_DECODER) += aic.o
> > OBJS-$(CONFIG_ALAC_DECODER) += alac.o alac_data.o
> > OBJS-$(CONFIG_ALAC_ENCODER) += alacenc.o alac_data.o
> > OBJS-$(CONFIG_ALS_DECODER) += alsdec.o bgmc.o mpeg4audio.o
> >diff --git a/libavcodec/aic.c b/libavcodec/aic.c
> >new file mode 100644
> >index 0000000..5cd5e4e
> >--- /dev/null
> >+++ b/libavcodec/aic.c
> >@@ -0,0 +1,477 @@
> >+/*
> >+ * Apple Intermediate Codec decoder
> >+ *
> >+ * Copyright (c) 2013 Karl-Alfred, Fritiof Andersson och jag
> >+ *
> >+ * This file is part of Libav.
> >+ *
> >+ * Libav 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.
> >+ *
> >+ * Libav 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 Libav; if not, write to the Free Software
> >+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301
> >USA
> >+ */
> >+
> >+#include "avcodec.h"
> >+#include "bytestream.h"
> >+#include "dsputil.h"
> >+#include "internal.h"
> >+#include "get_bits.h"
> >+#include "unary.h"
> >+#include "golomb.h"
> >+
> >+#define AIC_HDR_SIZE 24
> >+
> >+enum AICBands {
> >+ COEFF_LUMA = 0,
> >+ COEFF_CHROMA,
> >+ COEFF_LUMA_EXT,
> >+ COEFF_CHROMA_EXT,
> >+ NUM_BANDS
> >+};
> >+
> >+static const int aic_num_band_coeffs[NUM_BANDS] = { 64, 32, 192, 96 };
> >+
> >+static const int aic_band_off[NUM_BANDS] = { 0, 64, 96, 288 };
> >+
> >
> >[ tables skipped for brevity]
> >
> >+
> >+static const uint8_t *aic_scan[NUM_BANDS] = {
> >+ aic_y_scan, aic_c_scan, aic_y_ext_scan,aic_c_ext_scan
>
> nit: Add a space after the last comma as well
yes, a leftover
> >+};
> >+
> >+typedef struct AICContext {
> >+ AVCodecContext *avctx;
> >+ AVFrame *frame;
> >+ DSPContext dsp;
> >+ ScanTable scantable;
> >+
> >+ int num_x_slices;
> >+ int slice_width;
> >+ int mb_width, mb_height;
> >+ int quant;
> >+ int interlaced;
> >+
> >+ int16_t *slice_data;
> >+ int16_t *data_ptr[NUM_BANDS];
> >+
> >+ DECLARE_ALIGNED(16, int16_t, block)[64];
> >+} AICContext;
> >+
> >+static int aic_decode_header(AICContext *ctx, const uint8_t *src, int size)
> >+{
> >+ uint32_t frame_size;
> >+ int width, height;
> >+
> >+ if (src[0] != 1) {
> >+ av_log(ctx->avctx, AV_LOG_ERROR, "Invalid version %d\n", src[0]);
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+ if (src[1] != AIC_HDR_SIZE - 2) {
> >+ av_log(ctx->avctx, AV_LOG_ERROR, "Invalid header size %d\n",
> >src[1]);
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+ frame_size = AV_RB32(src + 2);
> >+ width = AV_RB16(src + 6);
> >+ height = AV_RB16(src + 8);
> >+ if (frame_size > size) {
> >+ av_log(ctx->avctx, AV_LOG_ERROR, "Frame size should be %d got %d\n",
> >+ frame_size, size);
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+ if (width != ctx->avctx->width || height != ctx->avctx->height) {
> >+ av_log(ctx->avctx, AV_LOG_ERROR,
> >+ "Picture dimension changed: old: %d x %d, new: %d x %d\n",
> >+ ctx->avctx->width, ctx->avctx->height, width, height);
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+ ctx->quant = src[15];
> >+ ctx->interlaced = ((src[16] >> 4) == 3);
> >+
> >+ return 0;
> >+}
> >+
> >+#define GET_CODE(val, type, add_bits) \
> >+ do { \
> >+ if (type) \
> >+ val = get_ue_golomb(gb); \
> >+ else \
> >+ val = get_unary(gb, 1, 31); \
> >+ if (add_bits) \
> >+ val = (val << add_bits) + get_bits(gb, add_bits); \
> >+ } while (0)
>
> nit: Perhaps you could align the backslashes here?
sure
> >+
> >+static int aic_decode_coeffs(GetBitContext *gb, int16_t *dst,
> >+ int band, int slice_width)
> >+{
> >+ int has_skips, coeff_type, coeff_bits, skip_type, skip_bits;
> >+ const int num_coeffs = aic_num_band_coeffs[band];
> >+ const uint8_t *scan = aic_scan[band];
> >+ int mb, idx, val;
> >+
> >+ has_skips = get_bits1(gb);
> >+ coeff_type = get_bits1(gb);
> >+ coeff_bits = get_bits(gb, 3);
> >+ // XXX: det är inte så snabb
>
> No swedish comments please :-)
Varf^H^H^H^HOK
> >+ if (has_skips) {
> >+ skip_type = get_bits1(gb);
> >+ skip_bits = get_bits(gb, 3);
> >+
> >+ for (mb = 0; mb < slice_width; mb++) {
> >+ idx = -1;
> >+ do {
> >+ GET_CODE(val, skip_type, skip_bits);
> >+ idx += val + 1;
> >+ if (idx >= num_coeffs)
> >+ break;
> >+ GET_CODE(val, coeff_type, coeff_bits);
> >+ val++;
> >+ if (val >= 0x10000)
> >+ return AVERROR_INVALIDDATA;
> >+ dst[scan[idx]] = val;
> >+ } while (idx < num_coeffs - 1);
> >+ dst += num_coeffs;
> >+ }
> >+ } else {
> >+ for (mb = 0; mb < slice_width; mb++) {
> >+ for (idx = 0; idx < num_coeffs; idx++) {
> >+ GET_CODE(val, coeff_type, coeff_bits);
> >+ if (val >= 0x10000)
> >+ return AVERROR_INVALIDDATA;
> >+ dst[scan[idx]] = val;
> >+ }
> >+ dst += num_coeffs;
> >+ }
> >+ }
> >+ return 0;
> >+}
> >+
> >+static void recombine_block(int16_t *dst, const uint8_t *scan,
> >+ int16_t **base, int16_t **ext)
> >+{
> >+ int i, j;
> >+
> >+ for (i = 0; i < 4; i++) {
> >+ for (j = 0; j < 4; j++)
> >+ dst[scan[i * 8 + j]] = (*base)[j];
> >+ for (j = 0; j < 4; j++)
> >+ dst[scan[i * 8 + j + 4]] = (*ext)[j];
> >+ *base += 4;
> >+ *ext += 4;
> >+ }
> >+ for (; i < 8; i++) {
> >+ for (j = 0; j < 8; j++)
> >+ dst[scan[i * 8 + j]] = (*ext)[j];
> >+ *ext += 8;
> >+ }
> >+}
> >+
> >+static void recombine_block_il(int16_t *dst, const uint8_t *scan,
> >+ int16_t **base, int16_t **ext,
> >+ int block_no)
> >+{
> >+ int i, j;
> >+
> >+ if (block_no < 2) {
> >+ for (i = 0; i < 8; i++) {
> >+ for (j = 0; j < 4; j++)
> >+ dst[scan[i * 8 + j]] = (*base)[j];
> >+ for (j = 0; j < 4; j++)
> >+ dst[scan[i * 8 + j + 4]] = (*ext)[j];
> >+ *base += 4;
> >+ *ext += 4;
> >+ }
> >+ } else {
> >+ for (i = 0; i < 64; i++)
> >+ dst[scan[i]] = (*ext)[i];
> >+ *ext += 64;
> >+ }
> >+}
> >+
> >+static void unquant_block(int16_t *block, int q)
> >+{
> >+ int i;
> >+
> >+ for (i = 0; i < 64; i++) {
> >+ int val = (uint16_t)block[i];
> >+ int sign = val & 1;
> >+
> >+ block[i] = (((val >> 1) ^ -sign) * q * aic_quant_matrix[i] >> 4)
> >+ + sign;
> >+ }
> >+}
> >+
> >+static int aic_decode_slice(AICContext *ctx, int mb_x, int mb_y,
> >+ const uint8_t *src, int src_size)
> >+{
> >+ GetBitContext gb;
> >+ int ret, i, mb, blk;
> >+ int slice_width = FFMIN(ctx->slice_width, ctx->mb_width - mb_x);
> >+ uint8_t *Y, *C[2];
> >+ uint8_t *dst;
> >+ int16_t *base_y = ctx->data_ptr[COEFF_LUMA];
> >+ int16_t *base_c = ctx->data_ptr[COEFF_CHROMA];
> >+ int16_t *ext_y = ctx->data_ptr[COEFF_LUMA_EXT];
> >+ int16_t *ext_c = ctx->data_ptr[COEFF_CHROMA_EXT];
> >+ const int ystride = ctx->frame->linesize[0];
> >+
> >+ Y = ctx->frame->data[0] + mb_x * 16 + mb_y * 16 * ystride;
> >+ for (i = 0; i < 2; i++)
> >+ C[i] = ctx->frame->data[i + 1] + mb_x * 8
> >+ + mb_y * 8 * ctx->frame->linesize[i + 1];
>
> nit: Don't we usually leave the operator, the + in this case, on the
> previous line? (The same thing seems to be in a few other places as well.)
Both variants seems to be common.
> >+ init_get_bits(&gb, src, src_size * 8);
> >+
> >+ memset(ctx->slice_data, 0, sizeof(*ctx->slice_data) * slice_width *
> >384);
> >+ for (i = 0; i < NUM_BANDS; i++)
> >+ if ((ret = aic_decode_coeffs(&gb, ctx->data_ptr[i],
> >+ i, slice_width)) < 0)
> >+ return ret;
> >+
> >+ for (mb = 0; mb < slice_width; mb++) {
> >+ for (blk = 0; blk < 4; blk++) {
> >+ if (!ctx->interlaced)
> >+ recombine_block(ctx->block, ctx->scantable.permutated,
> >+ &base_y, &ext_y);
> >+ else
> >+ recombine_block_il(ctx->block, ctx->scantable.permutated,
> >+ &base_y, &ext_y, blk);
> >+ unquant_block(ctx->block, ctx->quant);
> >+ ctx->dsp.idct(ctx->block);
> >+
> >+ if (!ctx->interlaced) {
> >+ dst = Y + (blk & 1) * 8 * ystride + (blk >> 1) * 8;
> >+ ctx->dsp.put_signed_pixels_clamped(ctx->block, dst,
> >+ ystride);
> >+ } else {
> >+ dst = Y + (blk & 1) * 8 + (blk >> 1) * ystride;
> >+ ctx->dsp.put_signed_pixels_clamped(ctx->block, dst,
> >+ ystride * 2);
> >+ }
> >+ }
> >+ Y += 16;
> >+
> >+ for (blk = 0; blk < 2; blk++) {
> >+ recombine_block(ctx->block, ctx->scantable.permutated,
> >+ &base_c, &ext_c);
> >+ unquant_block(ctx->block, ctx->quant);
> >+ ctx->dsp.idct(ctx->block);
> >+ ctx->dsp.put_signed_pixels_clamped(ctx->block, C[blk],
> >+ ctx->frame->linesize[blk +
> >1]);
> >+ C[blk] += 8;
> >+ }
> >+ }
> >+
> >+ return 0;
> >+}
> >+
> >+static int aic_decode_frame(AVCodecContext *avctx, void *data, int
> >*got_frame,
> >+ AVPacket *avpkt)
> >+{
> >+ AICContext *ctx = avctx->priv_data;
> >+ const uint8_t *buf = avpkt->data;
> >+ int buf_size = avpkt->size;
> >+ GetByteContext gb;
> >+ uint32_t off;
> >+ int x, y, ret;
> >+ int slice_size;
> >+
> >+ ctx->frame = data;
> >+ ctx->frame->pict_type = AV_PICTURE_TYPE_I;
> >+ ctx->frame->key_frame = 1;
> >+
> >+ off = FFALIGN(AIC_HDR_SIZE + ctx->num_x_slices * ctx->mb_height * 2, 4);
> >+
> >+ if (buf_size < off) {
> >+ av_log(avctx, AV_LOG_ERROR, "Too small frame\n");
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+
> >+ if ((ret = aic_decode_header(ctx, buf, buf_size)) < 0)
> >+ return ret;
> >+
> >+ if ((ret = ff_get_buffer(avctx, ctx->frame, 0)) < 0)
> >+ return ret;
> >+
> >+ bytestream2_init(&gb, buf + AIC_HDR_SIZE,
> >+ ctx->num_x_slices * ctx->mb_height * 2);
> >+
> >+ for (y = 0; y < ctx->mb_height; y++) {
> >+ for (x = 0; x < ctx->mb_width; x += ctx->slice_width) {
> >+ slice_size = bytestream2_get_le16(&gb) * 4;
> >+ if (slice_size + off > buf_size || !slice_size) {
> >+ av_log(avctx, AV_LOG_ERROR, "Incorrect slice size\n");
> >+ return AVERROR_INVALIDDATA;
> >+ }
> >+
> >+ if ((ret = aic_decode_slice(ctx, x, y,
> >+ buf + off, slice_size)) < 0)
> >+ return ret;
> >+
> >+ off += slice_size;
> >+ }
> >+ }
> >+
> >+ *got_frame = 1;
> >+
> >+ return avpkt->size;
> >+}
> >+
> >+static av_cold int aic_decode_init(AVCodecContext *avctx)
> >+{
> >+ AICContext *ctx = avctx->priv_data;
> >+ int i;
> >+ uint8_t scan[64];
> >+
> >+ ctx->avctx = avctx;
> >+
> >+ avctx->pix_fmt = AV_PIX_FMT_YUV420P;
> >+
> >+ ff_dsputil_init(&ctx->dsp, avctx);
> >+
> >+ for (i = 0; i < 64; i++)
> >+ scan[i] = i;
> >+ ff_init_scantable(ctx->dsp.idct_permutation, &ctx->scantable, scan);
> >+
> >+ ctx->mb_width = FFALIGN(avctx->width, 16) >> 4;
> >+ ctx->mb_height = FFALIGN(avctx->height, 16) >> 4;
> >+
> >+ ctx->num_x_slices = 16;
> >+ ctx->slice_width = ctx->mb_width / 16;
> >+ for (i = 1; i < 32; i++) {
> >+ if (!(ctx->mb_width % i) && (ctx->mb_width / i < 32)) {
> >+ ctx->slice_width = ctx->mb_width / i;
> >+ ctx->num_x_slices = i;
> >+ break;
> >+ }
> >+ }
> >+
> >+ ctx->slice_data = av_malloc(ctx->slice_width * (256 + 128)
> >+ * sizeof(*ctx->slice_data));
>
> In aic_decode_slice, you wrote a literal 384 instead of 256 + 128.
> Please keep it consistent between the two... Or perhaps a TOT_NUM_BAND_COEFS
> define?
I'll add a define.
> >+ if (!ctx->slice_data) {
> >+ av_log(avctx, AV_LOG_ERROR, "Error allocating slice buffer\n");
> >+
> >+ av_freep(&ctx->slice_data);
>
> This feels a bit redundant given that you've just checked that it's null
Indeed, a leftover from the time there were two buffers.
> >+
> >+ return AVERROR(ENOMEM);
> >+ }
> >+
> >+ for (i = 0; i < NUM_BANDS; i++)
> >+ ctx->data_ptr[i] = ctx->slice_data + ctx->slice_width
> >+ * aic_band_off[i];
> >+
> >+ return 0;
> >+}
> >+
> >+static av_cold int aic_decode_close(AVCodecContext *avctx)
> >+{
> >+ AICContext *ctx = avctx->priv_data;
> >+
> >+ av_freep(&ctx->slice_data);
> >+
> >+ return 0;
> >+}
> >+
> >+AVCodec ff_aic_decoder = {
> >+ .name = "aic",
> >+ .type = AVMEDIA_TYPE_VIDEO,
> >+ .id = AV_CODEC_ID_AIC,
> >+ .priv_data_size = sizeof(AICContext),
> >+ .init = aic_decode_init,
> >+ .close = aic_decode_close,
> >+ .decode = aic_decode_frame,
> >+ .capabilities = CODEC_CAP_DR1,
> >+ .long_name = NULL_IF_CONFIG_SMALL("Apple Intermediate Codec")
> >+};
> >diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
> >index 45fea3a..3706c91 100644
> >--- a/libavcodec/allcodecs.c
> >+++ b/libavcodec/allcodecs.c
> >@@ -96,6 +96,7 @@ void avcodec_register_all(void)
> > REGISTER_ENCODER(A64MULTI, a64multi);
> > REGISTER_ENCODER(A64MULTI5, a64multi5);
> > REGISTER_DECODER(AASC, aasc);
> >+ REGISTER_DECODER(AIC, aic);
> > REGISTER_DECODER(AMV, amv);
> > REGISTER_DECODER(ANM, anm);
> > REGISTER_DECODER(ANSI, ansi);
> >diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
> >index 968f9e5..0f2c733 100644
> >--- a/libavcodec/avcodec.h
> >+++ b/libavcodec/avcodec.h
> >@@ -267,6 +267,7 @@ enum AVCodecID {
> > AV_CODEC_ID_CLLC,
> > AV_CODEC_ID_MSS2,
> > AV_CODEC_ID_VP9,
> >+ AV_CODEC_ID_AIC,
> >
> > /* various PCM "codecs" */
> > AV_CODEC_ID_FIRST_AUDIO = 0x10000, ///< A dummy id pointing at the
> > start of audio codecs
> >diff --git a/libavcodec/codec_desc.c b/libavcodec/codec_desc.c
> >index fe2cbab..6dda00f 100644
> >--- a/libavcodec/codec_desc.c
> >+++ b/libavcodec/codec_desc.c
> >@@ -1207,6 +1207,13 @@ static const AVCodecDescriptor codec_descriptors[] = {
> > .long_name = NULL_IF_CONFIG_SMALL("MS Windows Media Video V9
> > Screen"),
> > .props = AV_CODEC_PROP_INTRA_ONLY | AV_CODEC_PROP_LOSSY,
> > },
> >+ {
> >+ .id = AV_CODEC_ID_AIC,
> >+ .type = AVMEDIA_TYPE_VIDEO,
> >+ .name = "aic",
> >+ .long_name = NULL_IF_CONFIG_SMALL("Apple Intermediate Codec"),
> >+ .props = AV_CODEC_PROP_INTRA_ONLY | AV_CODEC_PROP_LOSSY,
> >+ },
> >
> > /* various PCM "codecs" */
> > {
> >diff --git a/libavcodec/version.h b/libavcodec/version.h
> >index 2dc0f4d..afc2d35 100644
> >--- a/libavcodec/version.h
> >+++ b/libavcodec/version.h
> >@@ -27,7 +27,7 @@
> > */
> >
> > #define LIBAVCODEC_VERSION_MAJOR 55
> >-#define LIBAVCODEC_VERSION_MINOR 4
> >+#define LIBAVCODEC_VERSION_MINOR 5
> > #define LIBAVCODEC_VERSION_MICRO 0
> >
> > #define LIBAVCODEC_VERSION_INT AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
> >diff --git a/libavformat/isom.c b/libavformat/isom.c
> >index ccf4ca1..2756544 100644
> >--- a/libavformat/isom.c
> >+++ b/libavformat/isom.c
> >@@ -224,6 +224,8 @@ const AVCodecTag ff_codec_movvideo_tags[] = {
> > { AV_CODEC_ID_PRORES, MKTAG('a', 'p', 'c', 'o') }, /* Apple ProRes 422
> > Proxy */
> > { AV_CODEC_ID_PRORES, MKTAG('a', 'p', '4', 'h') }, /* Apple ProRes 4444
> > */
> >
> >+ { AV_CODEC_ID_AIC, MKTAG('i', 'c', 'o', 'd') },
> >+
> > { AV_CODEC_ID_NONE, 0 },
> > };
>
> Would this perhaps warrant a micro bump in lavf? I don't know...
It never did and it shouldn't - it does not change API or add some new
(de)muxer.
> All in all, looks very good to me, I don't want to re-review it just because
> of the minor stylistic nits I pointed out. Bonuses such as fate tests
> can come later...
I'll send an updated version just in case.
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
