vf: add vf_convert as interim replacement for vf_scale

We want to drop vf_scale, but we still need a way to auto convert
between imgfmts. In particular, vf.c will auto insert the "scale" filter
if the VO doesn't support a pixfmt.

To avoid chaos, create a new vf_convert.c filter, based on vf_scale.c,
but without the unrelicensed code parts. In particular, this filter does
not do scaling and has no options. It merely converts from one imgfmt to
another, if needed.

4 files changed, 133 insertions, 3 deletions

diff --git a/Copyright b/Copyright
index c31bf7011d..7dd4c6a27e 100644
--- a/Copyright
+++ b/Copyright
@@ -42,7 +42,6 @@ them quite central:
 - many builtin video filters (use libavfilter instead)
 - automatic rotation and stereoscopic video handling
 - Linux TV input
-- video format conversion (if not supported by the VO)
 - minor features: jack, DVD, CDDA, SMB, CACA, legacy direct3d VO
 Some of these will be fixed in the future. The intended use for LGPL mode is
 with libmpv, and currently it's not recommended to build mpv CLI in LGPL mode
diff --git a/video/filter/vf.c b/video/filter/vf.c
index c2fc4f716a..6e2cb251e0 100644
--- a/video/filter/vf.c
+++ b/video/filter/vf.c
@@ -53,6 +53,7 @@ extern const vf_info_t vf_info_pullup;
 extern const vf_info_t vf_info_sub;
 extern const vf_info_t vf_info_yadif;
 extern const vf_info_t vf_info_stereo3d;
+extern const vf_info_t vf_info_convert;
 extern const vf_info_t vf_info_lavfi;
 extern const vf_info_t vf_info_lavfi_bridge;
 extern const vf_info_t vf_info_vaapi;
@@ -81,6 +82,7 @@ static const vf_info_t *const filter_list[] = {
     &vf_info_sub,
 #endif
 
+    &vf_info_convert,
     &vf_info_lavfi,
     &vf_info_lavfi_bridge,
     &vf_info_buffer,
@@ -553,7 +555,7 @@ static void query_formats(uint8_t *fmts, struct vf_instance *vf)
 
 static bool is_conv_filter(struct vf_instance *vf)
 {
-    return vf && (strcmp(vf->info->name, "scale") == 0 || vf->autoinserted);
+    return vf && (strcmp(vf->info->name, "convert") == 0 || vf->autoinserted);
 }
 
 static const char *find_conv_filter(uint8_t *fmts_in, uint8_t *fmts_out)
@@ -569,7 +571,7 @@ static const char *find_conv_filter(uint8_t *fmts_in, uint8_t *fmts_out)
             }
         }
     }
-    return "scale";
+    return "convert";
 }
 
 static void update_formats(struct vf_chain *c, struct vf_instance *vf,
diff --git a/video/filter/vf_convert.c b/video/filter/vf_convert.c
new file mode 100644
index 0000000000..d7b4c21e56
--- /dev/null
+++ b/video/filter/vf_convert.c
@@ -0,0 +1,128 @@
+/*
+ * This file is part of mpv.
+ *
+ * mpv 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.
+ *
+ * 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 Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with mpv.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <math.h>
+#include <inttypes.h>
+#include <stdarg.h>
+#include <assert.h>
+
+#include <libswscale/swscale.h>
+
+#include "common/av_common.h"
+#include "common/msg.h"
+
+#include "options/options.h"
+
+#include "video/img_format.h"
+#include "video/mp_image.h"
+#include "video/sws_utils.h"
+#include "video/fmt-conversion.h"
+#include "vf.h"
+
+struct vf_priv_s {
+    struct mp_sws_context *sws;
+};
+
+static int find_best_out(vf_instance_t *vf, int in_format)
+{
+    int best = 0;
+    for (int out_format = IMGFMT_START; out_format < IMGFMT_END; out_format++) {
+        if (!vf_next_query_format(vf, out_format))
+            continue;
+        if (sws_isSupportedOutput(imgfmt2pixfmt(out_format)) < 1)
+            continue;
+        if (best) {
+            int candidate = mp_imgfmt_select_best(best, out_format, in_format);
+            if (candidate)
+                best = candidate;
+        } else {
+            best = out_format;
+        }
+    }
+    return best;
+}
+
+static int reconfig(struct vf_instance *vf, struct mp_image_params *in,
+                    struct mp_image_params *out)
+{
+    unsigned int best = find_best_out(vf, in->imgfmt);
+    if (!best) {
+        MP_WARN(vf, "no supported output format found\n");
+        return -1;
+    }
+
+    *out = *in;
+    out->imgfmt = best;
+
+    mp_image_params_guess_csp(out);
+
+    mp_sws_set_from_cmdline(vf->priv->sws, vf->chain->opts->vo->sws_opts);
+    vf->priv->sws->src = *in;
+    vf->priv->sws->dst = *out;
+
+    if (mp_sws_reinit(vf->priv->sws) < 0) {
+        // error...
+        MP_WARN(vf, "Couldn't init libswscale for this setup\n");
+        return -1;
+    }
+    return 0;
+}
+
+static struct mp_image *filter(struct vf_instance *vf, struct mp_image *mpi)
+{
+    struct mp_image *dmpi = vf_alloc_out_image(vf);
+    if (!dmpi)
+        return NULL;
+    mp_image_copy_attributes(dmpi, mpi);
+
+    mp_sws_scale(vf->priv->sws, dmpi, mpi);
+
+    talloc_free(mpi);
+    return dmpi;
+}
+
+static int query_format(struct vf_instance *vf, unsigned int fmt)
+{
+    if (IMGFMT_IS_HWACCEL(fmt) || sws_isSupportedInput(imgfmt2pixfmt(fmt)) < 1)
+        return 0;
+    return !!find_best_out(vf, fmt);
+}
+
+static void uninit(struct vf_instance *vf)
+{
+}
+
+static int vf_open(vf_instance_t *vf)
+{
+    vf->reconfig = reconfig;
+    vf->filter = filter;
+    vf->query_format = query_format;
+    vf->uninit = uninit;
+    vf->priv->sws = mp_sws_alloc(vf);
+    vf->priv->sws->log = vf->log;
+    return 1;
+}
+
+const vf_info_t vf_info_convert = {
+    .description = "image format conversion with libswscale",
+    .name = "convert",
+    .open = vf_open,
+    .priv_size = sizeof(struct vf_priv_s),
+};
diff --git a/wscript_build.py b/wscript_build.py
index a2971176a3..62798a1ced 100644
--- a/wscript_build.py
+++ b/wscript_build.py
@@ -374,6 +374,7 @@ def build(ctx):
         ( "video/filter/refqueue.c" ),
         ( "video/filter/vf.c" ),
         ( "video/filter/vf_buffer.c" ),
+        ( "video/filter/vf_convert.c" )
         ( "video/filter/vf_crop.c",              "gpl" ),
         ( "video/filter/vf_d3d11vpp.c",          "d3d-hwaccel" ),
         ( "video/filter/vf_dsize.c",             "gpl" ),