mirror of https://github.com/FFmpeg/FFmpeg.git
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
695 lines
20 KiB
695 lines
20 KiB
/* |
|
* AviSynth/AvxSynth support |
|
* Copyright (c) 2012 AvxSynth Team. |
|
* |
|
* 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 "libavutil/internal.h" |
|
#include "libavcodec/internal.h" |
|
#include "avformat.h" |
|
#include "internal.h" |
|
|
|
/* Enable function pointer definitions for runtime loading. */ |
|
#define AVSC_NO_DECLSPEC |
|
|
|
/* Platform-specific directives for AviSynth vs AvxSynth. |
|
* |
|
* avisynth_c.h needs to be the one provided with x264, as |
|
* the one in AviSynth's CVS hasn't been updated to support |
|
* 2.6's extra colorspaces. A temporary source of that header, |
|
* installable from a GNU-style Makefile is available from |
|
* github.com/qyot27/avisynth_headers -- AvxSynth doesn't |
|
* require this kind of special treatment because like any |
|
* standard *nix application, it installs its headers |
|
* alongside its libs. */ |
|
#ifdef _WIN32 |
|
#include <windows.h> |
|
#undef EXTERN_C |
|
#include <avisynth/avisynth_c.h> |
|
#define AVISYNTH_LIB "avisynth" |
|
#define USING_AVISYNTH |
|
#else |
|
#include <dlfcn.h> |
|
#include <avxsynth/avxsynth_c.h> |
|
#if defined (__APPLE__) |
|
#define AVISYNTH_LIB "libavxsynth.dylib" |
|
#else |
|
#define AVISYNTH_LIB "libavxsynth.so" |
|
#endif |
|
|
|
#define LoadLibrary(x) dlopen(x, RTLD_NOW | RTLD_LOCAL) |
|
#define GetProcAddress dlsym |
|
#define FreeLibrary dlclose |
|
#endif |
|
|
|
typedef struct AviSynthLibrary { |
|
void *library; |
|
#define AVSC_DECLARE_FUNC(name) name ## _func name |
|
AVSC_DECLARE_FUNC(avs_bit_blt); |
|
AVSC_DECLARE_FUNC(avs_clip_get_error); |
|
AVSC_DECLARE_FUNC(avs_create_script_environment); |
|
AVSC_DECLARE_FUNC(avs_delete_script_environment); |
|
AVSC_DECLARE_FUNC(avs_get_audio); |
|
AVSC_DECLARE_FUNC(avs_get_error); |
|
AVSC_DECLARE_FUNC(avs_get_frame); |
|
AVSC_DECLARE_FUNC(avs_get_version); |
|
AVSC_DECLARE_FUNC(avs_get_video_info); |
|
AVSC_DECLARE_FUNC(avs_invoke); |
|
AVSC_DECLARE_FUNC(avs_release_clip); |
|
AVSC_DECLARE_FUNC(avs_release_value); |
|
AVSC_DECLARE_FUNC(avs_release_video_frame); |
|
AVSC_DECLARE_FUNC(avs_take_clip); |
|
#undef AVSC_DECLARE_FUNC |
|
} AviSynthLibrary; |
|
|
|
typedef struct AviSynthContext { |
|
AVS_ScriptEnvironment *env; |
|
AVS_Clip *clip; |
|
const AVS_VideoInfo *vi; |
|
|
|
/* avisynth_read_packet_video() iterates over this. */ |
|
int n_planes; |
|
const int *planes; |
|
|
|
int curr_stream; |
|
int curr_frame; |
|
int64_t curr_sample; |
|
|
|
int error; |
|
|
|
/* Linked list pointers. */ |
|
struct AviSynthContext *next; |
|
} AviSynthContext; |
|
|
|
static const int avs_planes_packed[1] = { 0 }; |
|
static const int avs_planes_grey[1] = { AVS_PLANAR_Y }; |
|
static const int avs_planes_yuv[3] = { AVS_PLANAR_Y, AVS_PLANAR_U, |
|
AVS_PLANAR_V }; |
|
|
|
/* A conflict between C++ global objects, atexit, and dynamic loading requires |
|
* us to register our own atexit handler to prevent double freeing. */ |
|
static AviSynthLibrary avs_library; |
|
static int avs_atexit_called = 0; |
|
|
|
/* Linked list of AviSynthContexts. An atexit handler destroys this list. */ |
|
static AviSynthContext *avs_ctx_list = NULL; |
|
|
|
static av_cold void avisynth_atexit_handler(void); |
|
|
|
static av_cold int avisynth_load_library(void) |
|
{ |
|
avs_library.library = LoadLibrary(AVISYNTH_LIB); |
|
if (!avs_library.library) |
|
return AVERROR_UNKNOWN; |
|
|
|
#define LOAD_AVS_FUNC(name, continue_on_fail) \ |
|
avs_library.name = \ |
|
(void *)GetProcAddress(avs_library.library, #name); \ |
|
if (!continue_on_fail && !avs_library.name) \ |
|
goto fail; |
|
|
|
LOAD_AVS_FUNC(avs_bit_blt, 0); |
|
LOAD_AVS_FUNC(avs_clip_get_error, 0); |
|
LOAD_AVS_FUNC(avs_create_script_environment, 0); |
|
LOAD_AVS_FUNC(avs_delete_script_environment, 0); |
|
LOAD_AVS_FUNC(avs_get_audio, 0); |
|
LOAD_AVS_FUNC(avs_get_error, 1); // New to AviSynth 2.6 |
|
LOAD_AVS_FUNC(avs_get_frame, 0); |
|
LOAD_AVS_FUNC(avs_get_version, 0); |
|
LOAD_AVS_FUNC(avs_get_video_info, 0); |
|
LOAD_AVS_FUNC(avs_invoke, 0); |
|
LOAD_AVS_FUNC(avs_release_clip, 0); |
|
LOAD_AVS_FUNC(avs_release_value, 0); |
|
LOAD_AVS_FUNC(avs_release_video_frame, 0); |
|
LOAD_AVS_FUNC(avs_take_clip, 0); |
|
#undef LOAD_AVS_FUNC |
|
|
|
atexit(avisynth_atexit_handler); |
|
return 0; |
|
|
|
fail: |
|
FreeLibrary(avs_library.library); |
|
return AVERROR_UNKNOWN; |
|
} |
|
|
|
/* Note that avisynth_context_create and avisynth_context_destroy |
|
* do not allocate or free the actual context! That is taken care of |
|
* by libavformat. */ |
|
static av_cold int avisynth_context_create(AVFormatContext *s) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
int ret; |
|
|
|
if (!avs_library.library) |
|
if (ret = avisynth_load_library()) |
|
return ret; |
|
|
|
avs->env = avs_library.avs_create_script_environment(3); |
|
if (avs_library.avs_get_error) { |
|
const char *error = avs_library.avs_get_error(avs->env); |
|
if (error) { |
|
av_log(s, AV_LOG_ERROR, "%s\n", error); |
|
return AVERROR_UNKNOWN; |
|
} |
|
} |
|
|
|
if (!avs_ctx_list) { |
|
avs_ctx_list = avs; |
|
} else { |
|
avs->next = avs_ctx_list; |
|
avs_ctx_list = avs; |
|
} |
|
|
|
return 0; |
|
} |
|
|
|
static av_cold void avisynth_context_destroy(AviSynthContext *avs) |
|
{ |
|
if (avs_atexit_called) |
|
return; |
|
|
|
if (avs == avs_ctx_list) { |
|
avs_ctx_list = avs->next; |
|
} else { |
|
AviSynthContext *prev = avs_ctx_list; |
|
while (prev->next != avs) |
|
prev = prev->next; |
|
prev->next = avs->next; |
|
} |
|
|
|
if (avs->clip) { |
|
avs_library.avs_release_clip(avs->clip); |
|
avs->clip = NULL; |
|
} |
|
if (avs->env) { |
|
avs_library.avs_delete_script_environment(avs->env); |
|
avs->env = NULL; |
|
} |
|
} |
|
|
|
static av_cold void avisynth_atexit_handler(void) |
|
{ |
|
AviSynthContext *avs = avs_ctx_list; |
|
|
|
while (avs) { |
|
AviSynthContext *next = avs->next; |
|
avisynth_context_destroy(avs); |
|
avs = next; |
|
} |
|
FreeLibrary(avs_library.library); |
|
|
|
avs_atexit_called = 1; |
|
} |
|
|
|
/* Create AVStream from audio and video data. */ |
|
static int avisynth_create_stream_video(AVFormatContext *s, AVStream *st) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
int planar = 0; // 0: packed, 1: YUV, 2: Y8 |
|
|
|
st->codec->codec_type = AVMEDIA_TYPE_VIDEO; |
|
st->codec->codec_id = AV_CODEC_ID_RAWVIDEO; |
|
st->codec->width = avs->vi->width; |
|
st->codec->height = avs->vi->height; |
|
|
|
st->time_base = (AVRational) { avs->vi->fps_denominator, |
|
avs->vi->fps_numerator }; |
|
st->avg_frame_rate = (AVRational) { avs->vi->fps_numerator, |
|
avs->vi->fps_denominator }; |
|
st->start_time = 0; |
|
st->duration = avs->vi->num_frames; |
|
st->nb_frames = avs->vi->num_frames; |
|
|
|
switch (avs->vi->pixel_type) { |
|
#ifdef USING_AVISYNTH |
|
case AVS_CS_YV24: |
|
st->codec->pix_fmt = AV_PIX_FMT_YUV444P; |
|
planar = 1; |
|
break; |
|
case AVS_CS_YV16: |
|
st->codec->pix_fmt = AV_PIX_FMT_YUV422P; |
|
planar = 1; |
|
break; |
|
case AVS_CS_YV411: |
|
st->codec->pix_fmt = AV_PIX_FMT_YUV411P; |
|
planar = 1; |
|
break; |
|
case AVS_CS_Y8: |
|
st->codec->pix_fmt = AV_PIX_FMT_GRAY8; |
|
planar = 2; |
|
break; |
|
#endif |
|
case AVS_CS_BGR24: |
|
st->codec->pix_fmt = AV_PIX_FMT_BGR24; |
|
break; |
|
case AVS_CS_BGR32: |
|
st->codec->pix_fmt = AV_PIX_FMT_RGB32; |
|
break; |
|
case AVS_CS_YUY2: |
|
st->codec->pix_fmt = AV_PIX_FMT_YUYV422; |
|
break; |
|
case AVS_CS_YV12: |
|
st->codec->pix_fmt = AV_PIX_FMT_YUV420P; |
|
planar = 1; |
|
break; |
|
case AVS_CS_I420: // Is this even used anywhere? |
|
st->codec->pix_fmt = AV_PIX_FMT_YUV420P; |
|
planar = 1; |
|
break; |
|
default: |
|
av_log(s, AV_LOG_ERROR, |
|
"unknown AviSynth colorspace %d\n", avs->vi->pixel_type); |
|
avs->error = 1; |
|
return AVERROR_UNKNOWN; |
|
} |
|
|
|
switch (planar) { |
|
case 2: // Y8 |
|
avs->n_planes = 1; |
|
avs->planes = avs_planes_grey; |
|
break; |
|
case 1: // YUV |
|
avs->n_planes = 3; |
|
avs->planes = avs_planes_yuv; |
|
break; |
|
default: |
|
avs->n_planes = 1; |
|
avs->planes = avs_planes_packed; |
|
} |
|
return 0; |
|
} |
|
|
|
static int avisynth_create_stream_audio(AVFormatContext *s, AVStream *st) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
|
|
st->codec->codec_type = AVMEDIA_TYPE_AUDIO; |
|
st->codec->sample_rate = avs->vi->audio_samples_per_second; |
|
st->codec->channels = avs->vi->nchannels; |
|
st->time_base = (AVRational) { 1, |
|
avs->vi->audio_samples_per_second }; |
|
st->duration = avs->vi->num_audio_samples; |
|
|
|
switch (avs->vi->sample_type) { |
|
case AVS_SAMPLE_INT8: |
|
st->codec->codec_id = AV_CODEC_ID_PCM_U8; |
|
break; |
|
case AVS_SAMPLE_INT16: |
|
st->codec->codec_id = AV_CODEC_ID_PCM_S16LE; |
|
break; |
|
case AVS_SAMPLE_INT24: |
|
st->codec->codec_id = AV_CODEC_ID_PCM_S24LE; |
|
break; |
|
case AVS_SAMPLE_INT32: |
|
st->codec->codec_id = AV_CODEC_ID_PCM_S32LE; |
|
break; |
|
case AVS_SAMPLE_FLOAT: |
|
st->codec->codec_id = AV_CODEC_ID_PCM_F32LE; |
|
break; |
|
default: |
|
av_log(s, AV_LOG_ERROR, |
|
"unknown AviSynth sample type %d\n", avs->vi->sample_type); |
|
avs->error = 1; |
|
return AVERROR_UNKNOWN; |
|
} |
|
return 0; |
|
} |
|
|
|
static int avisynth_create_stream(AVFormatContext *s) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVStream *st; |
|
int ret; |
|
int id = 0; |
|
|
|
if (avs_has_video(avs->vi)) { |
|
st = avformat_new_stream(s, NULL); |
|
if (!st) |
|
return AVERROR_UNKNOWN; |
|
st->id = id++; |
|
if (ret = avisynth_create_stream_video(s, st)) |
|
return ret; |
|
} |
|
if (avs_has_audio(avs->vi)) { |
|
st = avformat_new_stream(s, NULL); |
|
if (!st) |
|
return AVERROR_UNKNOWN; |
|
st->id = id++; |
|
if (ret = avisynth_create_stream_audio(s, st)) |
|
return ret; |
|
} |
|
return 0; |
|
} |
|
|
|
static int avisynth_open_file(AVFormatContext *s) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVS_Value arg, val; |
|
int ret; |
|
#ifdef USING_AVISYNTH |
|
char filename_ansi[MAX_PATH * 4]; |
|
wchar_t filename_wc[MAX_PATH * 4]; |
|
#endif |
|
|
|
if (ret = avisynth_context_create(s)) |
|
return ret; |
|
|
|
#ifdef USING_AVISYNTH |
|
/* Convert UTF-8 to ANSI code page */ |
|
MultiByteToWideChar(CP_UTF8, 0, s->filename, -1, filename_wc, MAX_PATH * 4); |
|
WideCharToMultiByte(CP_THREAD_ACP, 0, filename_wc, -1, filename_ansi, |
|
MAX_PATH * 4, NULL, NULL); |
|
arg = avs_new_value_string(filename_ansi); |
|
#else |
|
arg = avs_new_value_string(s->filename); |
|
#endif |
|
val = avs_library.avs_invoke(avs->env, "Import", arg, 0); |
|
if (avs_is_error(val)) { |
|
av_log(s, AV_LOG_ERROR, "%s\n", avs_as_error(val)); |
|
ret = AVERROR_UNKNOWN; |
|
goto fail; |
|
} |
|
if (!avs_is_clip(val)) { |
|
av_log(s, AV_LOG_ERROR, "AviSynth script did not return a clip\n"); |
|
ret = AVERROR_UNKNOWN; |
|
goto fail; |
|
} |
|
|
|
avs->clip = avs_library.avs_take_clip(val, avs->env); |
|
avs->vi = avs_library.avs_get_video_info(avs->clip); |
|
|
|
#ifdef USING_AVISYNTH |
|
/* libav only supports AviSynth 2.6 on Windows. Since AvxSynth |
|
* identifies itself as interface version 3 like 2.5.8, this |
|
* needs to be special-cased. */ |
|
|
|
if (avs_library.avs_get_version(avs->clip) == 3) { |
|
av_log(s, AV_LOG_ERROR, |
|
"AviSynth 2.5.8 not supported. Please upgrade to 2.6.\n"); |
|
ret = AVERROR_UNKNOWN; |
|
goto fail; |
|
} |
|
#endif |
|
|
|
/* Release the AVS_Value as it will go out of scope. */ |
|
avs_library.avs_release_value(val); |
|
|
|
if (ret = avisynth_create_stream(s)) |
|
goto fail; |
|
|
|
return 0; |
|
|
|
fail: |
|
avisynth_context_destroy(avs); |
|
return ret; |
|
} |
|
|
|
static void avisynth_next_stream(AVFormatContext *s, AVStream **st, |
|
AVPacket *pkt, int *discard) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
|
|
avs->curr_stream++; |
|
avs->curr_stream %= s->nb_streams; |
|
|
|
*st = s->streams[avs->curr_stream]; |
|
if ((*st)->discard == AVDISCARD_ALL) |
|
*discard = 1; |
|
else |
|
*discard = 0; |
|
|
|
return; |
|
} |
|
|
|
/* Copy AviSynth clip data into an AVPacket. */ |
|
static int avisynth_read_packet_video(AVFormatContext *s, AVPacket *pkt, |
|
int discard) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVS_VideoFrame *frame; |
|
unsigned char *dst_p; |
|
const unsigned char *src_p; |
|
int n, i, plane, rowsize, planeheight, pitch, bits; |
|
const char *error; |
|
|
|
if (avs->curr_frame >= avs->vi->num_frames) |
|
return AVERROR_EOF; |
|
|
|
/* This must happen even if the stream is discarded to prevent desync. */ |
|
n = avs->curr_frame++; |
|
if (discard) |
|
return 0; |
|
|
|
#ifdef USING_AVISYNTH |
|
/* Define the bpp values for the new AviSynth 2.6 colorspaces. |
|
* Since AvxSynth doesn't have these functions, special-case |
|
* it in order to avoid implicit declaration errors. */ |
|
|
|
if (avs_is_yv24(avs->vi)) |
|
bits = 24; |
|
else if (avs_is_yv16(avs->vi)) |
|
bits = 16; |
|
else if (avs_is_yv411(avs->vi)) |
|
bits = 12; |
|
else if (avs_is_y8(avs->vi)) |
|
bits = 8; |
|
else |
|
#endif |
|
bits = avs_bits_per_pixel(avs->vi); |
|
|
|
/* Without the cast to int64_t, calculation overflows at about 9k x 9k |
|
* resolution. */ |
|
pkt->size = (((int64_t)avs->vi->width * |
|
(int64_t)avs->vi->height) * bits) / 8; |
|
if (!pkt->size) |
|
return AVERROR_UNKNOWN; |
|
|
|
if (av_new_packet(pkt, pkt->size) < 0) |
|
return AVERROR(ENOMEM); |
|
|
|
pkt->pts = n; |
|
pkt->dts = n; |
|
pkt->duration = 1; |
|
pkt->stream_index = avs->curr_stream; |
|
|
|
frame = avs_library.avs_get_frame(avs->clip, n); |
|
error = avs_library.avs_clip_get_error(avs->clip); |
|
if (error) { |
|
av_log(s, AV_LOG_ERROR, "%s\n", error); |
|
avs->error = 1; |
|
av_packet_unref(pkt); |
|
return AVERROR_UNKNOWN; |
|
} |
|
|
|
dst_p = pkt->data; |
|
for (i = 0; i < avs->n_planes; i++) { |
|
plane = avs->planes[i]; |
|
src_p = avs_get_read_ptr_p(frame, plane); |
|
pitch = avs_get_pitch_p(frame, plane); |
|
|
|
rowsize = avs_get_row_size_p(frame, plane); |
|
planeheight = avs_get_height_p(frame, plane); |
|
|
|
/* Flip RGB video. */ |
|
if (avs_is_rgb24(avs->vi) || avs_is_rgb(avs->vi)) { |
|
src_p = src_p + (planeheight - 1) * pitch; |
|
pitch = -pitch; |
|
} |
|
|
|
avs_library.avs_bit_blt(avs->env, dst_p, rowsize, src_p, pitch, |
|
rowsize, planeheight); |
|
dst_p += rowsize * planeheight; |
|
} |
|
|
|
avs_library.avs_release_video_frame(frame); |
|
return 0; |
|
} |
|
|
|
static int avisynth_read_packet_audio(AVFormatContext *s, AVPacket *pkt, |
|
int discard) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVRational fps, samplerate; |
|
int samples; |
|
int64_t n; |
|
const char *error; |
|
|
|
if (avs->curr_sample >= avs->vi->num_audio_samples) |
|
return AVERROR_EOF; |
|
|
|
fps.num = avs->vi->fps_numerator; |
|
fps.den = avs->vi->fps_denominator; |
|
samplerate.num = avs->vi->audio_samples_per_second; |
|
samplerate.den = 1; |
|
|
|
if (avs_has_video(avs->vi)) { |
|
if (avs->curr_frame < avs->vi->num_frames) |
|
samples = av_rescale_q(avs->curr_frame, samplerate, fps) - |
|
avs->curr_sample; |
|
else |
|
samples = av_rescale_q(1, samplerate, fps); |
|
} else { |
|
samples = 1000; |
|
} |
|
|
|
/* After seeking, audio may catch up with video. */ |
|
if (samples <= 0) { |
|
pkt->size = 0; |
|
pkt->data = NULL; |
|
return 0; |
|
} |
|
|
|
if (avs->curr_sample + samples > avs->vi->num_audio_samples) |
|
samples = avs->vi->num_audio_samples - avs->curr_sample; |
|
|
|
/* This must happen even if the stream is discarded to prevent desync. */ |
|
n = avs->curr_sample; |
|
avs->curr_sample += samples; |
|
if (discard) |
|
return 0; |
|
|
|
pkt->size = avs_bytes_per_channel_sample(avs->vi) * |
|
samples * avs->vi->nchannels; |
|
if (!pkt->size) |
|
return AVERROR_UNKNOWN; |
|
|
|
if (av_new_packet(pkt, pkt->size) < 0) |
|
return AVERROR(ENOMEM); |
|
|
|
pkt->pts = n; |
|
pkt->dts = n; |
|
pkt->duration = samples; |
|
pkt->stream_index = avs->curr_stream; |
|
|
|
avs_library.avs_get_audio(avs->clip, pkt->data, n, samples); |
|
error = avs_library.avs_clip_get_error(avs->clip); |
|
if (error) { |
|
av_log(s, AV_LOG_ERROR, "%s\n", error); |
|
avs->error = 1; |
|
av_packet_unref(pkt); |
|
return AVERROR_UNKNOWN; |
|
} |
|
return 0; |
|
} |
|
|
|
static av_cold int avisynth_read_header(AVFormatContext *s) |
|
{ |
|
int ret; |
|
|
|
// Calling library must implement a lock for thread-safe opens. |
|
if (ret = avpriv_lock_avformat()) |
|
return ret; |
|
|
|
if (ret = avisynth_open_file(s)) { |
|
avpriv_unlock_avformat(); |
|
return ret; |
|
} |
|
|
|
avpriv_unlock_avformat(); |
|
return 0; |
|
} |
|
|
|
static int avisynth_read_packet(AVFormatContext *s, AVPacket *pkt) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVStream *st; |
|
int discard = 0; |
|
int ret; |
|
|
|
if (avs->error) |
|
return AVERROR_UNKNOWN; |
|
|
|
/* If either stream reaches EOF, try to read the other one before |
|
* giving up. */ |
|
avisynth_next_stream(s, &st, pkt, &discard); |
|
if (st->codec->codec_type == AVMEDIA_TYPE_VIDEO) { |
|
ret = avisynth_read_packet_video(s, pkt, discard); |
|
if (ret == AVERROR_EOF && avs_has_audio(avs->vi)) { |
|
avisynth_next_stream(s, &st, pkt, &discard); |
|
return avisynth_read_packet_audio(s, pkt, discard); |
|
} |
|
} else { |
|
ret = avisynth_read_packet_audio(s, pkt, discard); |
|
if (ret == AVERROR_EOF && avs_has_video(avs->vi)) { |
|
avisynth_next_stream(s, &st, pkt, &discard); |
|
return avisynth_read_packet_video(s, pkt, discard); |
|
} |
|
} |
|
|
|
return ret; |
|
} |
|
|
|
static av_cold int avisynth_read_close(AVFormatContext *s) |
|
{ |
|
if (avpriv_lock_avformat()) |
|
return AVERROR_UNKNOWN; |
|
|
|
avisynth_context_destroy(s->priv_data); |
|
avpriv_unlock_avformat(); |
|
return 0; |
|
} |
|
|
|
static int avisynth_read_seek(AVFormatContext *s, int stream_index, |
|
int64_t timestamp, int flags) |
|
{ |
|
AviSynthContext *avs = s->priv_data; |
|
AVStream *st; |
|
AVRational fps, samplerate; |
|
|
|
if (avs->error) |
|
return AVERROR_UNKNOWN; |
|
|
|
fps = (AVRational) { avs->vi->fps_numerator, |
|
avs->vi->fps_denominator }; |
|
samplerate = (AVRational) { avs->vi->audio_samples_per_second, 1 }; |
|
|
|
st = s->streams[stream_index]; |
|
if (st->codec->codec_type == AVMEDIA_TYPE_VIDEO) { |
|
/* AviSynth frame counts are signed int. */ |
|
if ((timestamp >= avs->vi->num_frames) || |
|
(timestamp > INT_MAX) || |
|
(timestamp < 0)) |
|
return AVERROR_EOF; |
|
avs->curr_frame = timestamp; |
|
if (avs_has_audio(avs->vi)) |
|
avs->curr_sample = av_rescale_q(timestamp, samplerate, fps); |
|
} else { |
|
if ((timestamp >= avs->vi->num_audio_samples) || (timestamp < 0)) |
|
return AVERROR_EOF; |
|
/* Force frame granularity for seeking. */ |
|
if (avs_has_video(avs->vi)) { |
|
avs->curr_frame = av_rescale_q(timestamp, fps, samplerate); |
|
avs->curr_sample = av_rescale_q(avs->curr_frame, samplerate, fps); |
|
} else { |
|
avs->curr_sample = timestamp; |
|
} |
|
} |
|
|
|
return 0; |
|
} |
|
|
|
AVInputFormat ff_avisynth_demuxer = { |
|
.name = "avisynth", |
|
.long_name = NULL_IF_CONFIG_SMALL("AviSynth script"), |
|
.priv_data_size = sizeof(AviSynthContext), |
|
.read_header = avisynth_read_header, |
|
.read_packet = avisynth_read_packet, |
|
.read_close = avisynth_read_close, |
|
.read_seek = avisynth_read_seek, |
|
.extensions = "avs", |
|
};
|
|
|