Now store metadata info from processing into the media type.

This comes in several parts:
 - Store the metadata from gstreamer during processing
 - Add a new JSONEncoded field to the VideoData table
 - And, of course, add a migration for that field!

This commit sponsored by Julius Tuomisto.  Thank you, Julius!

mediagoblin/media_types/video/migrations.py

@@ -14,4 +14,19 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
+from mediagoblin.db.migration_tools import RegisterMigration, inspect_table
+
+from sqlalchemy import MetaData, Column, Unicode
+
 MIGRATIONS = {}
+
+@RegisterMigration(1, MIGRATIONS)
+def add_orig_metadata_column(db_conn):
+    metadata = MetaData(bind=db_conn.bind)
+
+    vid_data = inspect_table(metadata, "video__mediadata")
+
+    col = Column('orig_metadata', Unicode,
+                 default=None, nullable=True)
+    col.create(vid_data)
+    db_conn.commit()

mediagoblin/media_types/video/models.py

@@ -20,12 +20,29 @@ from mediagoblin.db.base import Base
 from sqlalchemy import (
     Column, Integer, SmallInteger, ForeignKey)
 from sqlalchemy.orm import relationship, backref
+from mediagoblin.db.extratypes import JSONEncoded
 
 
 BACKREF_NAME = "video__media_data"
 
 
 class VideoData(Base):
+    """
+    Attributes:
+     - media_data: the originating media entry (of course)
+     - width: width of the transcoded video
+     - height: height of the transcoded video
+     - orig_metadata: A loose json structure containing metadata gstreamer
+         pulled from the original video.
+         This field is NOT GUARANTEED to exist!
+    
+         Likely metadata extracted:
+           "videoheight", "videolength", "videowidth",
+           "audiorate", "audiolength", "audiochannels", "audiowidth",
+           "mimetype", "tags"
+     
+         TODO: document the above better.
+    """
     __tablename__ = "video__mediadata"
 
     # The primary key *and* reference to the main media_entry
@@ -38,6 +55,8 @@ class VideoData(Base):
     width = Column(SmallInteger)
     height = Column(SmallInteger)
 
+    orig_metadata = Column(JSONEncoded)
+
 
 DATA_MODEL = VideoData
 MODELS = [VideoData]

mediagoblin/media_types/video/processing.py

@@ -86,8 +86,12 @@ def process_video(proc_state):
             mgg.global_config['media:medium']['max_width'],
             mgg.global_config['media:medium']['max_height'])
 
+        # Extract metadata and keep a record of it
         metadata = transcoders.VideoTranscoder().discover(queued_filename)
+        store_metadata(entry, metadata)
 
+        # Figure out whether or not we need to transcode this video or
+        # if we can skip it
         if skip_transcode(metadata):
             _log.debug('Skipping transcoding')
             # Just push the submitted file to the tmp_dst
@@ -152,3 +156,27 @@ def process_video(proc_state):
 
     # Remove queued media file from storage and database
     proc_state.delete_queue_file()
+
+
+def store_metadata(media_entry, metadata):
+    """
+    Store metadata from this video for this media entry.
+    """
+    # Let's pull out the easy, not having to be converted ones first
+    stored_metadata = dict(
+        [(key, metadata[key])
+         for key in [
+                 "videoheight", "videolength", "videowidth",
+                 "audiorate", "audiolength", "audiochannels", "audiowidth",
+                 "mimetype", "tags"]
+         if key in metadata])
+
+    # We have to convert videorate into a sequence because it's a
+    # special type normally..
+
+    if "videorate" in metadata:
+        videorate = metadata["videorate"]
+        stored_metadata["videorate"] = [videorate.num, videorate.denom]
+
+    media_entry.media_data_init(
+        orig_metadata=stored_metadata)