player/play-back.desktop

@@ -1,8 +0,0 @@
-[Desktop Entry]
-Type=Application
-Exec=/srv/pandora/t_for_time/player/player.py --mode main --playlist /srv/t_for_time/render/back.m3u
-Hidden=false
-NoDisplay=false
-X-GNOME-Autostart-enabled=true
-Name=t-for-time
-Comment=

player/play-front.desktop

@@ -1,8 +0,0 @@
-[Desktop Entry]
-Type=Application
-Exec=/srv/pandora/t_for_time/player/player.py --mode peer --playlist /srv/t_for_time/render/front.m3u
-Hidden=false
-NoDisplay=false
-X-GNOME-Autostart-enabled=true
-Name=t-for-time
-Comment=

player/player.py

@@ -1,197 +0,0 @@
-#!/usr/bin/python3
-import argparse
-import os
-import socket
-import collections
-import time
-from threading import Thread
-
-import mpv
-
-
-import logging
-logger = logging.getLogger('t_for_time')
-
-SYNC_TOLERANCE = 0.05
-SYNC_GRACE_TIME = 5
-SYNC_JUMP_AHEAD = 1
-PORT = 9067
-DEBUG = False
-
-
-def mpv_log(loglevel, component, message):
-    logger.info('[{}] {}: {}'.format(loglevel, component, message))
-
-
-class Main:
-    playlist_current_pos = -1
-    time_pos = -1
-
-class Sync(Thread):
-    active = True
-    is_main = True
-    ready = False
-    destination = "255.255.255.255"
-
-    def __init__(self, *args, **kwargs):
-        self.is_main = kwargs.get('mode', 'main') == 'main'
-        self.sock = self.init_socket()
-        self.main = Main()
-        if self.is_main:
-            self.socket_enable_broadcast()
-        font_size = 28
-        font = 'Menlo'
-
-        self.mpv = mpv.MPV(
-            log_handler=mpv_log, input_default_bindings=True,
-            input_vo_keyboard=True, sub_text_font_size=font_size, sub_text_font=font,
-        )
-        self.mpv.observe_property('time-pos', self.time_pos_cb)
-        self.mpv.fullscreen = kwargs.get('fullscreen', False)
-        self.mpv.loop = 'inf'
-        self.mpv.loop_file = 'no'
-        self.mpv.register_key_binding('q', self.q_binding)
-        self.playlist = kwargs['playlist']
-        print(self.playlist)
-        self.mpv.loadlist(self.playlist)
-        self.deviations = collections.deque(maxlen=10)
-        if not self.is_main:
-            self.mpv.pause = False
-            time.sleep(0.1)
-            self.mpv.pause = True
-            self.sync_to_main()
-        self.ready = True
-        Thread.__init__(self)
-        self.daemon = True
-        self.start()
-
-    def run(self):
-        while self.active:
-            if self.is_main:
-                pass
-            else:
-                self.read_position_main()
-                #self.adjust_position()
-
-    def q_binding(self, *args):
-        self.stop()
-        self.mpv.stop()
-
-
-    def stop(self, *args):
-        self.active = False
-        if self.sock:
-            self.sock.close()
-            self.sock = None
-
-    def time_pos_cb(self, pos, *args, **kwargs):
-        if self.is_main:
-            self.send_position_local()
-        elif self.ready:
-            self.adjust_position()
-
-
-    def init_socket(self):
-        sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM, 0)
-        sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
-        sock.bind(("0.0.0.0", PORT))
-        return sock
-
-    #
-    # main specific
-    #
-    def socket_enable_broadcast(self):
-        self.sock.setsockopt(socket.SOL_SOCKET, socket.SO_BROADCAST, 1)
-        self.sock.connect((self.destination, PORT))
-
-    def send_position_local(self):
-        if not self.active:
-            return
-        try:
-            msg = (
-                "%0.4f %s"
-                % (self.mpv.time_pos, self.mpv.playlist_current_pos)
-            ).encode()
-        except:
-            return
-        try:
-            self.sock.send(msg)
-        except socket.error as e:
-            logger.error("send failed: %s", e)
-
-    #
-    # follower specific
-    #
-
-    def read_position_main(self):
-        data = self.sock.recvfrom(1024)[0].decode().split(" ", 1)
-        self.main.time_pos = float(data[0])
-        self.main.playlist_current_pos = int(data[1])
-
-    def adjust_position(self):
-        if self.mpv.time_pos is not None:
-            deviation = self.main.time_pos - self.mpv.time_pos
-            self.deviations.append(deviation)
-            median_deviation = self.median(list(self.deviations))
-            frames = deviation / 0.04
-            median_frames = median_deviation / 0.04
-            if time.time() - self.last_sync > SYNC_GRACE_TIME and abs(median_deviation) > SYNC_TOLERANCE:
-                print('need to sync %0.05f (%d)  median %0.05f (%d)' % (deviation, frames, median_deviation, median_frames))
-                self.sync_to_main()
-
-    def median(self, lst):
-        quotient, remainder = divmod(len(lst), 2)
-        if remainder:
-            return sorted(lst)[quotient]
-        return float(sum(sorted(lst)[quotient - 1:quotient + 1]) / 2.0)
-
-    def sync_to_main(self):
-        self.read_position_main()
-        #print(self.main.playlist_current_pos)
-        if self.main.playlist_current_pos != self.mpv.playlist_current_pos:
-            self.mpv.playlist_play_index(self.main.playlist_current_pos)
-            self.mpv.pause = False
-            time.sleep(0.1)
-        self.mpv.pause = True
-        pos = self.main.time_pos + SYNC_JUMP_AHEAD
-        #print(pos, self.mpv.playlist_current_pos, self.mpv.time_pos)
-        self.mpv.seek(pos, 'absolute', 'exact')
-        time.sleep(0.1)
-        self.read_position_main()
-        sync_timer = time.time() # - 10 * 0.04
-        deviation = self.main.time_pos - self.mpv.time_pos
-        while self.active:
-            #print(deviation, abs(deviation) - (time.time() - sync_timer))
-            if abs(deviation) - (time.time() - sync_timer) < 0:
-                self.mpv.pause = False
-                break
-        self.last_sync = time.time()
-
-
-def main():
-    prefix = os.path.expanduser('~/Videos/t_for_time')
-
-    parser = argparse.ArgumentParser(description='t_for_time sync player')
-    parser.add_argument('--mode', help='ip of peer', default="peer")
-    parser.add_argument('--playlist', default='/srv/t_for_time/render/128/front.m3u', help="m3u")
-    parser.add_argument('--prefix', help='video location', default=prefix)
-    parser.add_argument('--window', action='store_true', help='run in window', default=False)
-    parser.add_argument('--debug', action='store_true', help='debug', default=False)
-    args = parser.parse_args()
-
-    DEBUG = args.debug
-    if DEBUG:
-        log_format = '%(asctime)s:%(levelname)s:%(name)s:%(message)s'
-        logging.basicConfig(level=logging.DEBUG, format=log_format)
-    base = os.path.dirname(os.path.abspath(__file__))
-    #os.chdir(base)
-
-    player = Sync(mode=args.mode, playlist=args.playlist, fullscreen=not args.window)
-    while player.active:
-        player.mpv.wait_for_playback()
-    player.stop()
-    del player.mpv
-
-
-if __name__ == "__main__":
-    main()

player/saxophone-loop.desktop

@@ -1,8 +0,0 @@
-[Desktop Entry]
-Type=Application
-Exec=/usr/bin/mpv --quiet --loop /srv/t_for_time/render/Saxophone-5.1.mp4
-Hidden=false
-NoDisplay=false
-X-GNOME-Autostart-enabled=true
-Name=loop
-Comment=

render.py

@@ -77,7 +77,6 @@ def compose(clips, target=150, base=1024, voice_over=None):
             vo2 = vo_keys.pop(chance(seq, len(vo_keys)))
             voice_overs.append(voice_over[vo1])
             if voice_over[vo1]["duration"] + voice_over[vo2]["duration"] < target:
-                print("adding second vo")
                 voice_overs.append(voice_over[vo2])
         vo_min = sum([vo['duration'] for vo in voice_overs])
         if vo_min > target:
@@ -155,8 +154,7 @@ def compose(clips, target=150, base=1024, voice_over=None):
         else:
             transparency_back = transparency
             transparency_front = 0
-        #transparency_original = seq() / 10
+        transparency_original = seq() / 10
-        transparency_original = 1
         if "background" in clip:
             scene['front']['V1'].append({
                 'duration': clip['duration'],
@@ -209,7 +207,7 @@ def compose(clips, target=150, base=1024, voice_over=None):
             'duration': clip['duration'],
             'src': foley,
         })
-    print("scene duration %0.3f (target: %0.3f, vo_min: %0.3f)" % (length, target, vo_min))
+    print("scene duration %0.3f (target: %0.3f)" % (length, target))
     return scene
 
 def get_scene_duration(scene):