sub: make use of libavcodec subtitle converters

This allows using some formats that were not supported until now, like
WebVTT.

We still prefer the internal subtitle reader (subreader.c), because
1. Libav, and 2. random things which we probably want to keep, such as
control over formatting, codepage stuff, or various mysterious
postprecessing done in that code.

5 files changed, 152 insertions, 0 deletions

diff --git a/Makefile b/Makefile
index d3e203a852..389a519055 100644
--- a/Makefile
+++ b/Makefile
@@ -231,6 +231,7 @@ SOURCES = talloc.c \
           sub/find_subfiles.c \
           sub/img_convert.c \
           sub/sd_lavc.c \
+          sub/sd_lavc_conv.c \
           sub/sd_microdvd.c \
           sub/sd_movtext.c \
           sub/sd_spu.c \
diff --git a/sub/dec_sub.c b/sub/dec_sub.c
index 49ecd5c009..ec9af94abb 100644
--- a/sub/dec_sub.c
+++ b/sub/dec_sub.c
@@ -35,6 +35,7 @@ extern const struct sd_functions sd_spu;
 extern const struct sd_functions sd_movtext;
 extern const struct sd_functions sd_srt;
 extern const struct sd_functions sd_microdvd;
+extern const struct sd_functions sd_lavc_conv;
 
 static const struct sd_functions *sd_list[] = {
 #ifdef CONFIG_ASS
@@ -45,6 +46,7 @@ static const struct sd_functions *sd_list[] = {
     &sd_movtext,
     &sd_srt,
     &sd_microdvd,
+    &sd_lavc_conv,
     NULL
 };
 
@@ -159,6 +161,14 @@ static void read_sub_data(struct dec_sub *sub, struct sub_data *subdata)
         sub_decode(sub, &pkt);
     }
 
+    struct sd *sd = sub_get_last_sd(sub);
+    // Hack for broken FFmpeg packet format: make sd_ass keep the subtitle
+    // events on reset(), even though broken FFmpeg ASS packets were received
+    // (from sd_lavc_conv.c). Normally, these events are removed on seek/reset,
+    // but this is obviously unwanted in this case.
+    if (sd && sd->driver->fix_events)
+        sd->driver->fix_events(sd);
+
     talloc_free(temp);
 }
 
diff --git a/sub/sd.h b/sub/sd.h
index dbb6af835f..20dce8003d 100644
--- a/sub/sd.h
+++ b/sub/sd.h
@@ -44,6 +44,8 @@ struct sd_functions {
     void (*reset)(struct sd *sd);
     void (*uninit)(struct sd *sd);
 
+    void (*fix_events)(struct sd *sd);
+
     // decoder
     void (*get_bitmaps)(struct sd *sd, struct mp_osd_res dim, double pts,
                         struct sub_bitmaps *res);
diff --git a/sub/sd_ass.c b/sub/sd_ass.c
index 405cef323a..d8951df96f 100644
--- a/sub/sd_ass.c
+++ b/sub/sd_ass.c
@@ -254,6 +254,12 @@ static char *get_text(struct sd *sd, double pts)
     return ctx->last_text;
 }
 
+static void fix_events(struct sd *sd)
+{
+    struct sd_ass_priv *ctx = sd->priv;
+    ctx->flush_on_seek = false;
+}
+
 static void reset(struct sd *sd)
 {
     struct sd_ass_priv *ctx = sd->priv;
@@ -281,6 +287,7 @@ const struct sd_functions sd_ass = {
     .decode = decode,
     .get_bitmaps = get_bitmaps,
     .get_text = get_text,
+    .fix_events = fix_events,
     .reset = reset,
     .uninit = uninit,
 };
diff --git a/sub/sd_lavc_conv.c b/sub/sd_lavc_conv.c
new file mode 100644
index 0000000000..e45e743499
--- /dev/null
+++ b/sub/sd_lavc_conv.c
@@ -0,0 +1,132 @@
+/*
+ * This file is part of mpv.
+ *
+ * mpv is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * mpv 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with mpv.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdlib.h>
+#include <assert.h>
+
+#include <libavcodec/avcodec.h>
+#include <libavutil/intreadwrite.h>
+#include <libavutil/common.h>
+
+#include "talloc.h"
+#include "core/mp_msg.h"
+#include "core/av_common.h"
+#include "sd.h"
+
+struct sd_lavc_priv {
+    AVCodecContext *avctx;
+};
+
+static bool supports_format(const char *format)
+{
+    enum AVCodecID cid = mp_codec_to_av_codec_id(format);
+    const AVCodecDescriptor *desc = avcodec_descriptor_get(cid);
+    // These are documented to support AVSubtitleRect->ass.
+    return desc && (desc->props & AV_CODEC_PROP_TEXT_SUB);
+}
+
+static int init(struct sd *sd)
+{
+    struct sd_lavc_priv *priv = talloc_zero(NULL, struct sd_lavc_priv);
+    AVCodecContext *avctx = NULL;
+    AVCodec *codec = avcodec_find_decoder(mp_codec_to_av_codec_id(sd->codec));
+    if (!codec)
+        goto error;
+    avctx = avcodec_alloc_context3(codec);
+    if (!avctx)
+        goto error;
+    avctx->extradata_size = sd->extradata_len;
+    avctx->extradata = sd->extradata;
+    if (avcodec_open2(avctx, codec, NULL) < 0)
+        goto error;
+    // Documented as "set by libavcodec", but there is no other way
+    avctx->time_base = (AVRational) {1, 1000};
+    priv->avctx = avctx;
+    sd->priv = priv;
+    sd->output_codec = "ass";
+    sd->output_extradata = avctx->subtitle_header;
+    sd->output_extradata_len = avctx->subtitle_header_size;
+    return 0;
+
+ error:
+    mp_msg(MSGT_SUBREADER, MSGL_ERR,
+           "Could not open libavcodec subtitle converter\n");
+    av_free(avctx);
+    talloc_free(priv);
+    return -1;
+}
+
+static void decode(struct sd *sd, struct demux_packet *packet)
+{
+    struct sd_lavc_priv *priv = sd->priv;
+    AVCodecContext *avctx = priv->avctx;
+    double ts = av_q2d(av_inv_q(avctx->time_base));
+    AVSubtitle sub = {0};
+    AVPacket pkt;
+    int ret, got_sub;
+
+    av_init_packet(&pkt);
+    pkt.data = packet->buffer;
+    pkt.size = packet->len;
+    pkt.pts = packet->pts == MP_NOPTS_VALUE ? AV_NOPTS_VALUE : packet->pts * ts;
+    pkt.duration = packet->duration * ts;
+
+    ret = avcodec_decode_subtitle2(avctx, &sub, &got_sub, &pkt);
+    if (ret < 0) {
+        mp_msg(MSGT_OSD, MSGL_ERR, "Error decoding subtitle\n");
+    } else if (got_sub) {
+        for (int i = 0; i < sub.num_rects; i++) {
+            char *ass_line = sub.rects[i]->ass;
+            if (!ass_line)
+                break;
+            // This might contain embedded timestamps, using the "old" ffmpeg
+            // ASS packet format, in which case pts/duration might be ignored
+            // at a later point.
+            sd_conv_add_packet(sd, ass_line, strlen(ass_line),
+                               packet->pts, packet->duration);
+        }
+    }
+
+    av_free_packet(&pkt);
+    avsubtitle_free(&sub);
+}
+
+static void reset(struct sd *sd)
+{
+    struct sd_lavc_priv *priv = sd->priv;
+
+    avcodec_flush_buffers(priv->avctx);
+    sd_conv_def_reset(sd);
+}
+
+static void uninit(struct sd *sd)
+{
+    struct sd_lavc_priv *priv = sd->priv;
+
+    avcodec_close(priv->avctx);
+    av_free(priv->avctx);
+    talloc_free(priv);
+}
+
+const struct sd_functions sd_lavc_conv = {
+    .supports_format = supports_format,
+    .init = init,
+    .decode = decode,
+    .get_converted = sd_conv_def_get_converted,
+    .reset = reset,
+    .uninit = uninit,
+};