def swfExportNoAudio(lecture, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640*20,480*20) ; dimScale = 0.5 * (dims[0]**2 + dims[1]**2)**0.5
styles = []
obj_count = 0
depth_count = 0
swfOutput = swf.SWF(fps=250, size=dims, fname=fname)
lastMessageLength = 0
nframes = int(math.ceil(dur * SWF_FPS))
for fnum in xrange(0, nframes):
sys.stdout.write("\x08"*lastMessageLength)
message = "writing frame %d/%d" % (fnum+1, nframes)
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
shapes = []
shape_extents = []
prog = fnum / (float(SWF_FPS))
styles = styles[-1:]
for e in it.next(prog):
if type(e) == Slide:
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i+1))
depth_count = 0
elif type(e) == Stroke:
styles.append(map(lambda x: int(255*x), (e.r(), e.g(), e.b())))
last_point = None
elif type(e) == Point:
x, y, p = int(e.x() * dims[0]), int(e.y() * dims[1]), int(dimScale*e.p)
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2*last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (min(shape_extents[-1][0], last_point[0] - p), \
min(shape_extents[-1][1], last_point[1] - p), \
max(shape_extents[-1][2], last_point[0] + p), \
max(shape_extents[-1][3], last_point[1] + p))
shape_extents[-1] = (min(shape_extents[-1][0], x - p), \
min(shape_extents[-1][1], y - p), \
max(shape_extents[-1][2], x + p), \
max(shape_extents[-1][3], y + p))
else:
shapes.append([(x, y)])
shape_extents.append(2*(x, y))
last_point = (x,y)
else:
print 'WTF is a %s?' % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2)))+1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(swf.DefineShape2(obj_count, \
swf.Rectangle(*shape_extents[shapeidx]), \
swf.ShapeWithStyle( \
[], \
[swf.LineStyle(2*pos[-1], swf.RGB(*styles[shapeidx])) \
for pos in shapes[shapeidx][1:]],\
0, \
nbits, \
[swf.StyleChangeRecord( \
x = shapes[shapeidx][0][0], \
y = shapes[shapeidx][0][1])] \
+ sum([[ \
swf.StyleChangeRecord(lineStyle = swf.binary(posidx, nbits)), \
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]) \
] for posidx in xrange(1,len(shapes[shapeidx]))], []) \
+ [swf.EndShapeRecord()]
)))
depth_count += 1
swfOutput.append( \
swf.PlaceObject2( \
depth_count, \
objectID = obj_count, \
placeFlagHasCharacter = True, \
placeFlagMove = False \
))
swfOutput.newFrame()
swfOutput.close()
print
def swfExportWithRawAudio(lecture, audio_data, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640*20,480*20) ; dimScale = 0.5 * (dims[0]**2 + dims[1]**2)**0.5
styles = []
obj_count = 0
depth_count = 0
# with compression I have to compute the number of frames beforehand
# since it is part of the compressed data stream
nframes = int(sum([math.ceil((len(audio_frames)*SWF_FPS)/(44100 * 2.)) for first_ts,audio_frames in audio_data]))
swfOutput = swf.SWF(fps=SWF_FPS, size=dims, fname=fname, nframes=nframes, compression=6)
lastMessageLength = 0
fnum = 0
swfOutput.append(swf.SoundStreamHead(sscount=(44100 / SWF_FPS)))
for first_ts, audio_frames in audio_data:
it.offset = first_ts + 1
prog = 0 ; audio_sample_idx = 0
while audio_sample_idx < len(audio_frames):
sys.stdout.write("\x08"*lastMessageLength)
message = "writing frame %d / %d" % (fnum+1, nframes) ; fnum += 1
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
#
# write audio data
new_sample_idx = int((prog + 1.0/SWF_FPS) * 44100) * 2
swfOutput.append(swf.SoundStreamBlock(audio_frames[audio_sample_idx:new_sample_idx]))
audio_sample_idx = new_sample_idx
#
# write strokes
shapes = []
shape_extents = []
styles = styles[-1:]
for e in it.next(prog):
if type(e) == Slide:
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i+1))
depth_count = 0
elif type(e) == Stroke:
styles.append(map(lambda x: int(255*x), (e.r(), e.g(), e.b())))
last_point = None
elif type(e) == Point:
x, y, p = int(e.x() * dims[0]), int(e.y() * dims[1]), int(dimScale*e.p)
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2*last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (min(shape_extents[-1][0], last_point[0] - p), \
min(shape_extents[-1][1], last_point[1] - p), \
max(shape_extents[-1][2], last_point[0] + p), \
max(shape_extents[-1][3], last_point[1] + p))
shape_extents[-1] = (min(shape_extents[-1][0], x - p), \
min(shape_extents[-1][1], y - p), \
max(shape_extents[-1][2], x + p), \
max(shape_extents[-1][3], y + p))
else:
shapes.append([(x, y)])
shape_extents.append(2*(x, y))
last_point = (x,y)
else:
print 'WTF is a %s?' % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2)))+1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(swf.DefineShape2(obj_count, \
swf.Rectangle(*shape_extents[shapeidx]), \
swf.ShapeWithStyle( \
[], \
[swf.LineStyle(2*pos[-1], swf.RGB(*styles[shapeidx])) \
for pos in shapes[shapeidx][1:]],\
0, \
nbits, \
[swf.StyleChangeRecord( \
x = shapes[shapeidx][0][0], \
y = shapes[shapeidx][0][1])] \
+ sum([[ \
swf.StyleChangeRecord(lineStyle = swf.binary(posidx, nbits)), \
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]) \
] for posidx in xrange(1,len(shapes[shapeidx]))], []) \
+ [swf.EndShapeRecord()]
)))
depth_count += 1
swfOutput.append( \
swf.PlaceObject2( \
depth_count, \
objectID = obj_count, \
placeFlagHasCharacter = True, \
placeFlagMove = False \
))
swfOutput.newFrame()
prog += 1./SWF_FPS
swfOutput.close()
print
def swfExportWithMP3(lecture, audio_data, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640*20,480*20) ; dimScale = 0.5 * (dims[0]**2 + dims[1]**2)**0.5
styles = []
obj_count = 0
depth_count = 0
# with compression I have to compute the number of frames beforehand
# since it is part of the compressed data stream
nframes = int(sum([math.ceil((len(audio_frames)*SWF_FPS)/(44100 * 2.)) for first_ts,audio_frames in audio_data]))
swfOutput = swf.SWF(fps=SWF_FPS, size=dims, fname=fname, nframes=nframes, compression=6)
lastMessageLength = 0
fnum = 0
swfOutput.append(swf.SoundStreamHead(sscount=(44100 / SWF_FPS)))
for first_ts, audio_frames in audio_data:
it.offset = first_ts + 1
prog = 0 ; audio_sample_idx = 0
import os, fcntl
lameInput = os.mkfifo("lamein.fifo")
lameOR, lameOW = os.pipe()
lameProcess = subprocess.Popen( \
"unbuffer /usr/bin/lame -r -s 44.1 --bitwidth 16 --signed --little-endian -m m --cbr -b 128 -t lamein.fifo -", stdout = lameOW, bufsize=20, shell=True)
lameInputFile = open("lamein.fifo", "w")
fcntl.fcntl(lameOR, fcntl.F_SETFL, fcntl.fcntl(lameOR, fcntl.F_GETFL) | os.O_NONBLOCK)
while audio_sample_idx < len(audio_frames):
sys.stdout.write("\x08"*lastMessageLength)
message = "writing frame %d / %d" % (fnum+1, nframes) ; fnum += 1
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
#
# write audio data
new_sample_idx = int((prog + 1.0/SWF_FPS) * 44100) * 2
# os.write(lameIW, audio_frames[audio_sample_idx:new_sample_idx])
currentBlock = audio_frames[audio_sample_idx:new_sample_idx]
lameInputFile.write(currentBlock + "\x00"*(new_sample_idx - len(currentBlock)))
lameInputFile.flush()
try:
count = 0
xx = "0"
while xx != "":
xx = os.read(lameOR, 1000)
count += len(xx)
# print "read %d (total %d bytes)" % (len(xx), count)
except OSError:
pass
finally:
print "read a total of %d bytes" % (count)
audio_sample_idx = new_sample_idx
#
# write strokes
shapes = []
shape_extents = []
styles = styles[-1:]
for e in it.next(prog):
if type(e) == Slide:
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i+1))
depth_count = 0
elif type(e) == Stroke:
styles.append(map(lambda x: int(255*x), (e.r(), e.g(), e.b())))
last_point = None
elif type(e) == Point:
x, y, p = int(e.x() * dims[0]), int(e.y() * dims[1]), int(dimScale*e.p)
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2*last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (min(shape_extents[-1][0], last_point[0] - p), \
min(shape_extents[-1][1], last_point[1] - p), \
max(shape_extents[-1][2], last_point[0] + p), \
max(shape_extents[-1][3], last_point[1] + p))
shape_extents[-1] = (min(shape_extents[-1][0], x - p), \
min(shape_extents[-1][1], y - p), \
max(shape_extents[-1][2], x + p), \
max(shape_extents[-1][3], y + p))
else:
shapes.append([(x, y)])
shape_extents.append(2*(x, y))
last_point = (x,y)
else:
print 'WTF is a %s?' % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2)))+1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(swf.DefineShape2(obj_count, \
swf.Rectangle(*shape_extents[shapeidx]), \
swf.ShapeWithStyle( \
[], \
[swf.LineStyle(2*pos[-1], swf.RGB(*styles[shapeidx])) \
for pos in shapes[shapeidx][1:]],\
0, \
nbits, \
[swf.StyleChangeRecord( \
x = shapes[shapeidx][0][0], \
y = shapes[shapeidx][0][1])] \
+ sum([[ \
swf.StyleChangeRecord(lineStyle = swf.binary(posidx, nbits)), \
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]) \
] for posidx in xrange(1,len(shapes[shapeidx]))], []) \
+ [swf.EndShapeRecord()]
)))
depth_count += 1
swfOutput.append( \
swf.PlaceObject2( \
depth_count, \
objectID = obj_count, \
placeFlagHasCharacter = True, \
placeFlagMove = False \
))
swfOutput.newFrame()
prog += 1./SWF_FPS
lameProcess.terminate()
os.unlink("lamein.fifo")
swfOutput.close()
print
def _to_swf_no_audio(lecture, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640 * 20, 480 * 20)
dimScale = 0.25 * (dims[0] ** 2 + dims[1] ** 2) ** 0.5
styles = []
obj_count = 0
depth_count = 0
swfOutput = swf.SWF(fps=250, size=dims, fname=fname)
lastMessageLength = 0
nframes = int(math.ceil(dur * SWF_FPS))
thickness = 0.01
for fnum in xrange(0, nframes):
sys.stdout.write("\x08" * lastMessageLength)
message = "writing frame %d/%d" % (fnum + 1, nframes)
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
shapes = []
shape_extents = []
prog = fnum / (float(SWF_FPS))
oldStyles = styles[-1:]
styles = []
for e in it.next(prog):
if isinstance(e, Slide):
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i + 1))
depth_count = 0
elif isinstance(e, Stroke):
styles.append(map(lambda x: int(255 * x), (e.r(), e.g(), e.b())))
thickness = e.thickness
last_point = None
elif isinstance(e, Point):
x, y = int(e.x() * dims[0]), int(e.y() * dims[1])
p = int(dimScale * thickness * e.p)
if len(styles) == 0:
styles = oldStyles
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2 * last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (
min(shape_extents[-1][0], last_point[0] - p),
min(shape_extents[-1][1], last_point[1] - p),
max(shape_extents[-1][2], last_point[0] + p),
max(shape_extents[-1][3], last_point[1] + p),
)
shape_extents[-1] = (
min(shape_extents[-1][0], x - p),
min(shape_extents[-1][1], y - p),
max(shape_extents[-1][2], x + p),
max(shape_extents[-1][3], y + p),
)
else:
shapes.append([(x, y)])
shape_extents.append(2 * (x, y))
last_point = (x, y)
else:
print "WTF is a %s?" % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2))) + 1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(
swf.DefineShape2(
obj_count,
swf.Rectangle(*shape_extents[shapeidx]),
swf.ShapeWithStyle(
[],
[swf.LineStyle(2 * pos[-1], swf.RGB(*styles[shapeidx])) for pos in shapes[shapeidx][1:]],
0,
nbits,
[swf.StyleChangeRecord(x=shapes[shapeidx][0][0], y=shapes[shapeidx][0][1])]
+ sum(
[
[
swf.StyleChangeRecord(lineStyle=swf.binary(posidx, nbits)),
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]),
]
for posidx in xrange(1, len(shapes[shapeidx]))
],
[],
)
+ [swf.EndShapeRecord()],
),
)
)
depth_count += 1
swfOutput.append(
swf.PlaceObject2(depth_count, objectID=obj_count, placeFlagHasCharacter=True, placeFlagMove=False)
)
swfOutput.newFrame()
swfOutput.close()
def _to_swf_raw_audio(lecture, audio_data, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640 * 20, 480 * 20)
dimScale = 0.25 * (dims[0] ** 2 + dims[1] ** 2) ** 0.5
styles = []
obj_count = 0
depth_count = 0
# with compression I have to compute the number of frames beforehand
# since it is part of the compressed data stream
nframes = int(
sum([math.ceil((len(audio_frames) * SWF_FPS) / (44100 * 2.0)) for first_ts, audio_frames in audio_data])
)
swfOutput = swf.SWF(fps=SWF_FPS, size=dims, fname=fname, nframes=nframes, compression=6)
lastMessageLength = 0
fnum = 0
swfOutput.append(swf.SoundStreamHead(sscount=(44100 / SWF_FPS)))
for first_ts, audio_frames in audio_data:
it.seek(first_ts + 1)
prog = 0
audio_sample_idx = 0
while audio_sample_idx < len(audio_frames):
sys.stdout.write("\x08" * lastMessageLength)
message = "writing frame %d / %d" % (fnum + 1, nframes)
fnum += 1
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
#
# write audio data
new_sample_idx = int((prog + 1.0 / SWF_FPS) * 44100) * 2
swfOutput.append(swf.SoundStreamBlock(audio_frames[audio_sample_idx:new_sample_idx]))
audio_sample_idx = new_sample_idx
#
# write strokes
thickness = 0.01
shapes = []
shape_extents = []
oldStyles = styles[-1:]
styles = []
for e in it.next(prog):
if isinstance(e, Slide):
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i + 1))
depth_count = 0
elif isinstance(e, Stroke):
styles.append(map(lambda x: int(255 * x), (e.r(), e.g(), e.b())))
thickness = e.thickness
last_point = None
elif isinstance(e, Point):
x, y = int(e.x() * dims[0]), int(e.y() * dims[1])
p = int(dimScale * thickness * e.p)
if len(styles) == 0:
styles = oldStyles
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2 * last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (
min(shape_extents[-1][0], last_point[0] - p),
min(shape_extents[-1][1], last_point[1] - p),
max(shape_extents[-1][2], last_point[0] + p),
max(shape_extents[-1][3], last_point[1] + p),
)
shape_extents[-1] = (
min(shape_extents[-1][0], x - p),
min(shape_extents[-1][1], y - p),
max(shape_extents[-1][2], x + p),
max(shape_extents[-1][3], y + p),
)
else:
shapes.append([(x, y)])
shape_extents.append(2 * (x, y))
last_point = (x, y)
else:
print "WTF is a %s?" % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2))) + 1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(
swf.DefineShape2(
obj_count,
swf.Rectangle(*shape_extents[shapeidx]),
swf.ShapeWithStyle(
[],
[swf.LineStyle(2 * pos[-1], swf.RGB(*styles[shapeidx])) for pos in shapes[shapeidx][1:]],
0,
nbits,
[swf.StyleChangeRecord(x=shapes[shapeidx][0][0], y=shapes[shapeidx][0][1])]
+ sum(
[
[
swf.StyleChangeRecord(lineStyle=swf.binary(posidx, nbits)),
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]),
]
for posidx in xrange(1, len(shapes[shapeidx]))
],
[],
)
+ [swf.EndShapeRecord()],
),
)
)
depth_count += 1
swfOutput.append(
swf.PlaceObject2(depth_count, objectID=obj_count, placeFlagHasCharacter=True, placeFlagMove=False)
)
swfOutput.newFrame()
prog += 1.0 / SWF_FPS
swfOutput.close()
def swfExportWithMP3(lecture, audio_data, fname):
it = iter(lecture)
first_ts = lecture.get_time_of_first_event()
last_ts = lecture.get_time_of_last_event()
dur = last_ts - first_ts
dims = (640 * 20, 480 * 20)
dimScale = 0.5 * (dims[0] ** 2 + dims[1] ** 2) ** 0.5
styles = []
obj_count = 0
depth_count = 0
# with compression I have to compute the number of frames beforehand
# since it is part of the compressed data stream
nframes = int(
sum([math.ceil((len(audio_frames) * SWF_FPS) / (44100 * 2.0)) for first_ts, audio_frames in audio_data])
)
swfOutput = swf.SWF(fps=SWF_FPS, size=dims, fname=fname, nframes=nframes, compression=6)
lastMessageLength = 0
fnum = 0
swfOutput.append(swf.SoundStreamHead(sscount=(44100 / SWF_FPS)))
for first_ts, audio_frames in audio_data:
it.offset = first_ts + 1
prog = 0
audio_sample_idx = 0
import os, fcntl
lameInput = os.mkfifo("lamein.fifo")
lameOR, lameOW = os.pipe()
lameProcess = subprocess.Popen(
"unbuffer /usr/bin/lame -r -s 44.1 --bitwidth 16 --signed --little-endian -m m --cbr -b 128 -t lamein.fifo -",
stdout=lameOW,
bufsize=20,
shell=True,
)
lameInputFile = open("lamein.fifo", "w")
fcntl.fcntl(lameOR, fcntl.F_SETFL, fcntl.fcntl(lameOR, fcntl.F_GETFL) | os.O_NONBLOCK)
while audio_sample_idx < len(audio_frames):
sys.stdout.write("\x08" * lastMessageLength)
message = "writing frame %d / %d" % (fnum + 1, nframes)
fnum += 1
sys.stdout.write(message)
lastMessageLength = len(message)
sys.stdout.flush()
#
# write audio data
new_sample_idx = int((prog + 1.0 / SWF_FPS) * 44100) * 2
# os.write(lameIW, audio_frames[audio_sample_idx:new_sample_idx])
currentBlock = audio_frames[audio_sample_idx:new_sample_idx]
lameInputFile.write(currentBlock + "\x00" * (new_sample_idx - len(currentBlock)))
lameInputFile.flush()
try:
count = 0
xx = "0"
while xx != "":
xx = os.read(lameOR, 1000)
count += len(xx)
# print "read %d (total %d bytes)" % (len(xx), count)
except OSError:
pass
finally:
print "read a total of %d bytes" % (count)
audio_sample_idx = new_sample_idx
#
# write strokes
shapes = []
shape_extents = []
styles = styles[-1:]
for e in it.next(prog):
if type(e) == Slide:
styles = []
shapes = []
for i in xrange(depth_count):
swfOutput.append(swf.RemoveObject2(i + 1))
depth_count = 0
elif type(e) == Stroke:
styles.append(map(lambda x: int(255 * x), (e.r(), e.g(), e.b())))
last_point = None
elif type(e) == Point:
x, y, p = int(e.x() * dims[0]), int(e.y() * dims[1]), int(dimScale * e.p)
if last_point:
if len(shapes) == 0:
shapes.append([last_point])
shape_extents.append(2 * last_point)
shapes[-1].append((x - last_point[0], y - last_point[1], p))
shape_extents[-1] = (
min(shape_extents[-1][0], last_point[0] - p),
min(shape_extents[-1][1], last_point[1] - p),
max(shape_extents[-1][2], last_point[0] + p),
max(shape_extents[-1][3], last_point[1] + p),
)
shape_extents[-1] = (
min(shape_extents[-1][0], x - p),
min(shape_extents[-1][1], y - p),
max(shape_extents[-1][2], x + p),
max(shape_extents[-1][3], y + p),
)
else:
shapes.append([(x, y)])
shape_extents.append(2 * (x, y))
last_point = (x, y)
else:
print "WTF is a %s?" % str(type(e)) # Unhandled.
for shapeidx in xrange(len(shapes)):
nbits = int(math.ceil(math.log(max(len(shapes[shapeidx]) - 1, 0.5), 2))) + 1
if len(shapes[shapeidx]) <= 1:
continue
obj_count += 1
swfOutput.append(
swf.DefineShape2(
obj_count,
swf.Rectangle(*shape_extents[shapeidx]),
swf.ShapeWithStyle(
[],
[swf.LineStyle(2 * pos[-1], swf.RGB(*styles[shapeidx])) for pos in shapes[shapeidx][1:]],
0,
nbits,
[swf.StyleChangeRecord(x=shapes[shapeidx][0][0], y=shapes[shapeidx][0][1])]
+ sum(
[
[
swf.StyleChangeRecord(lineStyle=swf.binary(posidx, nbits)),
swf.StraightEdgeRecord(*shapes[shapeidx][posidx][:-1]),
]
for posidx in xrange(1, len(shapes[shapeidx]))
],
[],
)
+ [swf.EndShapeRecord()],
),
)
)
depth_count += 1
swfOutput.append(
swf.PlaceObject2(depth_count, objectID=obj_count, placeFlagHasCharacter=True, placeFlagMove=False)
)
swfOutput.newFrame()
prog += 1.0 / SWF_FPS
lameProcess.terminate()
os.unlink("lamein.fifo")
swfOutput.close()
print
