Create edl2scenelist.py

Wrote an example script for extracting scene lists from a video edl.

bin/edl2scenelist.py

@@ -0,0 +1,84 @@
+import pycmx
+import re
+import argparse
+import sys
+
+def all_video_edits(edl):
+    for event in edl.events:
+        for edit in event.edits:
+            if edit.channels.video:
+                yield edit
+
+
+def get_scene_name(edit, pattern):
+    scene_extractor = re.compile(pattern, re.I)
+    if edit.clip_name is None:
+        return None
+    else:
+        match_data = re.match(scene_extractor, edit.clip_name)
+        if match_data:
+            return match_data[1]
+        else:
+            return edit.clip_name
+
+def output_cmx(outfile, out_list):
+    outfile.write("TITLE:  SCENE LIST\r\n")
+    outfile.write("FCM: NON-DROP FRAME\r\n")
+
+    for o in out_list:
+        line = "%03i  AX       V     C        00:00:00:00 00:00:00:00 %s %s\r\n" % (0, o['start'],o['end'])
+        outfile.write(line)
+        outfile.write("* FROM CLIP NAME: %s\r\n" % (o['scene']) )
+
+
+
+def scene_list(infile, outfile, out_format, pattern):
+    
+    edl = pycmx.parse_cmx3600(infile)
+
+    current_scene_name = None
+    
+    grouped_edits = [ ]
+
+    for edit in all_video_edits(edl):
+        this_scene_name = get_scene_name(edit, pattern)
+        if this_scene_name is not None:
+            if current_scene_name != this_scene_name:
+                grouped_edits.append([ ])
+                current_scene_name = this_scene_name
+
+            grouped_edits[-1].append(edit)
+    
+    out_list = [ ]
+    for group in grouped_edits:
+        out_list.append({ 
+            'start': group[0].record_in, 
+            'end': group[-1].record_out,
+            'scene': get_scene_name(group[0], pattern ) }
+            )
+    
+    if out_format == 'cmx':
+        output_cmx(outfile, out_list)
+
+
+def scene_list_cli():
+    parser = argparse.ArgumentParser(description=
+            'Read video events from an input CMX EDL and output events merged into scehes.')
+    parser.add_argument('-o','--outfile', default=sys.stdout, type=argparse.FileType('w'), 
+            help='Output file. Default is stdout.')
+    parser.add_argument('-f','--format', default='cmx', type=str, 
+            help='Output format. Options are json and cmx, cmx is the default.')
+    parser.add_argument('-p','--pattern', default='V?([A-Z]*[0-9]+)',
+            help='RE pattern for extracting scene name from clip name. The default is "V?([A-Z]*[0-9]+)". ' + \
+                    'This pattern will be matched case-insensitively.')
+    parser.add_argument('input_edl', default=sys.stdin, type=argparse.FileType('r'), nargs='?', 
+            help='Input file. Default is stdin.')
+    args = parser.parse_args()
+    
+    infile = args.input_edl
+
+    scene_list(infile=infile, outfile=args.outfile , out_format=args.format, pattern=args.pattern)
+
+
+if __name__ == '__main__':
+   scene_list_cli()