typefind: Add typefinder for AV1 obu/annexb raw streams

AV1 specification [1] define 2 AV1 raw bitstreams storage
formats without containers:
- OBU in chapiter 5.2
- Annexb in chapiter 11.

Implement a detection function for the both cases mostly
by testing OBU forbidden, type, has_size_field and reserved bits.
For annexb case testing if temporal unit size, frame unit size and
obu length are valid. If they are check that the first OBU is
a temporal delimiter.

[1] https://aomediacodec.github.io/av1-spec/av1-spec.pdf

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8502>

subprojects/gst-plugins-base/gst/typefind/gsttypefindfunctions.c

@@ -6777,6 +6777,154 @@ aa_type_find (GstTypeFind * tf, gpointer private)
   }
 }
 
+/*** video/x-av1 ***/
+
+static GstStaticCaps av1_caps =
+    GST_STATIC_CAPS
+    ("video/x-av1,stream-format={obu-stream, annexb},alignment=none");
+
+#define AV1_CAPS gst_static_caps_get(&av1_caps)
+
+static gboolean
+av1_is_valid_obu_type (guint obu_type)
+{
+  if ((obu_type > 0 && obu_type < 9) || obu_type == 15)
+    return TRUE;
+
+  return FALSE;
+}
+
+static gboolean
+av1_leb128 (const guint8 * data, guint32 * retval, gint * read_bytes)
+{
+  guint8 leb128_byte = 0;
+  guint64 value = 0;
+  gint i;
+
+  *retval = 0;
+  *read_bytes = 0;
+
+  for (i = 0; i < 8; i++) {
+    leb128_byte = data[i];
+    value |= (((gint) leb128_byte & 0x7f) << (i * 7));
+    if (!(leb128_byte & 0x80))
+      break;
+
+    if (i == 7 && leb128_byte & 0x80)
+      return FALSE;
+  }
+
+  if (i == 8)
+    return FALSE;
+
+  /* check for bitstream conformance see chapter 4.10.5 */
+  if (value < G_MAXUINT32) {
+    *retval = (guint32) value;
+    *read_bytes = i + 1;
+    return TRUE;
+  }
+
+  return FALSE;
+}
+
+static gboolean
+av1_is_valid_obu (const guint8 * data, guint * obu_type, gint * read_bytes)
+{
+  gboolean obu_forbidden_bit;
+  gboolean obu_extension_flag;
+  gboolean obu_has_size_field;
+  gboolean obu_reserved_1bit;
+  int offset = 1;
+
+  *obu_type = 0;
+  *read_bytes = 0;
+
+  /* Detect OBU header */
+  obu_forbidden_bit = !!(data[0] & 0x80);
+  if (obu_forbidden_bit)
+    return FALSE;
+
+  *obu_type = (data[0] & 0x78) >> 3;
+  obu_extension_flag = !!(data[0] & 0x4);
+  obu_has_size_field = !!(data[0] & 0x2);
+  obu_reserved_1bit = !!(data[0] & 0x1);
+
+  /* if obu_extension_flag is set temporal_id (3 bits)
+   * spatial_id (2 bits) and extension_header_reserved_3bits (3 bits)
+   * field are coded in the header so OBU size field is
+   * 1 byte after */
+  if (obu_extension_flag)
+    offset++;
+
+  *read_bytes += offset;
+
+  if (av1_is_valid_obu_type (*obu_type) && !obu_reserved_1bit) {
+    if (obu_has_size_field) {
+      guint32 obu_size;
+      gint bytes;
+
+      if (!av1_leb128 (data + offset, &obu_size, &bytes))
+        return FALSE;
+      *read_bytes += bytes;
+    }
+
+    return TRUE;
+  }
+
+  return FALSE;
+}
+
+static void
+av1_type_find (GstTypeFind * tf, gpointer unused)
+{
+  guint32 temporal_unit_size;
+  guint32 frame_unit_size;
+  guint32 obu_length;
+  const guint8 *data;
+  guint obu_type;
+  gint read_bytes;
+  gint offset = 0;
+
+  data = gst_type_find_peek (tf, 0, 25);
+  if (!data)
+    return;
+
+  if (av1_is_valid_obu (data, &obu_type, &read_bytes)) {
+    gst_type_find_suggest_simple (tf, GST_TYPE_FIND_MINIMUM, "video/x-av1",
+        "stream-format", G_TYPE_STRING, "obu-stream",
+        "alignment", G_TYPE_STRING, "none", NULL);
+    return;
+  }
+
+  if (!av1_leb128 (data, &temporal_unit_size, &read_bytes))
+    return;
+  offset += read_bytes;
+
+  if (!av1_leb128 (data + offset, &frame_unit_size, &read_bytes))
+    return;
+  offset += read_bytes;
+
+  if (frame_unit_size > temporal_unit_size)
+    return;
+
+  if (!av1_leb128 (data + offset, &obu_length, &read_bytes))
+    return;
+  offset += read_bytes;
+
+  if (obu_length > frame_unit_size)
+    return;
+
+  if (!av1_is_valid_obu (data + offset, &obu_type, &read_bytes))
+    return;
+  offset += read_bytes;
+
+  /* The first OBU must be a temporal delimiter */
+  if (obu_type == 2)
+    gst_type_find_suggest_simple (tf, GST_TYPE_FIND_MINIMUM, "video/x-av1",
+        "stream-format", G_TYPE_STRING, "annexb",
+        "alignment", G_TYPE_STRING, "none", NULL);
+}
+
 /*Type find definition by functions */
 GST_TYPE_FIND_REGISTER_DEFINE (musepack, "audio/x-musepack", GST_RANK_PRIMARY,
     musepack_type_find, "mpc,mpp,mp+", MUSEPACK_CAPS, NULL, NULL);
@@ -7042,3 +7190,5 @@ GST_TYPE_FIND_REGISTER_DEFINE (wsaud, "application/x-wsaud", GST_RANK_MARGINAL,
     wsaud_type_find, "wsaud", WSAUD_CAPS, NULL, NULL);
 GST_TYPE_FIND_REGISTER_DEFINE (wsvqa, "application/x-wsvqa", GST_RANK_MARGINAL,
     wsvqa_type_find, "wsvqa", WSVQA_CAPS, NULL, NULL);
+GST_TYPE_FIND_REGISTER_DEFINE (av1, "video/x-av1", GST_RANK_MARGINAL,
+    av1_type_find, "av1", AV1_CAPS, NULL, NULL);

subprojects/gst-plugins-base/gst/typefind/gsttypefindfunctionsplugin.c

@@ -214,6 +214,7 @@ plugin_init (GstPlugin * plugin)
   GST_TYPE_FIND_REGISTER (film_cpk, plugin);
   GST_TYPE_FIND_REGISTER (gxf, plugin);
   GST_TYPE_FIND_REGISTER (iff, plugin);
+  GST_TYPE_FIND_REGISTER (av1, plugin);
 
   return TRUE;
 }

subprojects/gst-plugins-base/gst/typefind/gsttypefindfunctionsplugin.h

@@ -220,5 +220,6 @@ GST_TYPE_FIND_REGISTER_DECLARE (vmd);
 GST_TYPE_FIND_REGISTER_DECLARE (wc3movie);
 GST_TYPE_FIND_REGISTER_DECLARE (wsaud);
 GST_TYPE_FIND_REGISTER_DECLARE (wsvqa);
+GST_TYPE_FIND_REGISTER_DECLARE (av1);
 
 #endif //__GST_TYPE_FIND_FUNCTIONS_PLUGIN_H__