1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
|
#!/usr/bin/python3
# dmedia: distributed media library
# Copyright (C) 2011 Novacut Inc
#
# This file is part of `dmedia`.
#
# `dmedia` is free software: you can redistribute it and/or modify it under
# the terms of the GNU Affero General Public License as published by the Free
# Software Foundation, either version 3 of the License, or (at your option) any
# later version.
#
# `dmedia` is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
# A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
# details.
#
# You should have received a copy of the GNU Affero General Public License along
# with `dmedia`. If not, see <http://www.gnu.org/licenses/>.
#
# Authors:
# Jason Gerard DeRose <jderose@novacut.com>
"""
A quick and dirty GStreamer extractor.
"""
import sys
from os import path
import json
import optparse
parser = optparse.OptionParser()
parser.add_option('--full',
help='do full play-through do gaurantee duration in fames and samples',
action='store_true',
default=False,
)
(options, args) = parser.parse_args()
if len(args) != 1:
print('takes exacly 1 argument FILENAME')
sys.exit(1)
filename = path.abspath(args[0])
if not path.isfile(filename):
print('not a file: {!r}'.format(filename))
sys.exit(1)
import gi
gi.require_version('Gst', '1.0')
from gi.repository import GLib, Gst
Gst.init(None)
class FakeBin(Gst.Bin):
def __init__(self, callback, doc, rate):
super().__init__()
self._callback = callback
self._doc = doc
self._q = Gst.ElementFactory.make('queue', None)
self._rate = Gst.ElementFactory.make(rate, None)
self._sink = Gst.ElementFactory.make('fakesink', None)
self.add(self._q)
self.add(self._rate)
self.add(self._sink)
self._q.link(self._rate)
self._rate.link(self._sink)
# Ghost Pads
pad = self._q.get_static_pad('sink')
self.add_pad(
Gst.GhostPad.new('sink', pad)
)
pad.connect('notify::caps', self._on_notify_caps)
def _query_duration(self, pad):
q = Gst.Query.new_duration(Gst.Format.TIME)
# FIXME: We get an intermittent failures here with this exception:
# TypeError: argument query: Expected Gst.Query, but got gi.repository.Gst.Query
# See LP:1069582
# https://bugs.launchpad.net/dmedia/+bug/1069582
try:
if pad.get_peer().query(q):
(format, duration) = q.parse_duration()
if format == Gst.Format.TIME:
return duration
except TypeError:
pass
def _on_notify_caps(self, pad, args):
caps = pad.get_current_caps()
if not caps:
return
ns = self._query_duration(pad)
self._extract(caps.get_structure(0), ns)
self._callback(self)
class VideoBin(FakeBin):
def __init__(self, callback, doc):
super().__init__(callback, doc, 'videorate')
def _extract(self, d, ns):
(success, num, denom) = d.get_fraction('framerate')
if num != 0:
self._doc['framerate'] = {
'num': num,
'denom': denom,
}
self._doc['width'] = d.get_int('width')[1]
self._doc['height'] = d.get_int('height')[1]
if ns:
self._doc['video_ns'] = ns
frames = int(
round(ns * num / denom / float(Gst.SECOND))
)
self._doc['duration']['frames'] = frames
# FIXME: Quick work-around for us not being able to thumbnail or
# playthrough the last few frames on GStreamer 1.0 on Ubuntu Trusty:
#self._doc['duration']['frames'] = max(1, frames - 15)
def _finalize(self):
d = dict(
(n, self._rate.get_property(n))
for n in ('in', 'out', 'duplicate', 'drop')
)
d['consistent'] = bool(d['in'] + d['duplicate'] - d['drop'] == d['out'])
d['matching'] = d['in'] == d['out'] == self._doc['duration']['frames']
self._doc['frames'] = d
class AudioBin(FakeBin):
def __init__(self, callback, doc):
super().__init__(callback, doc, 'audiorate')
def _extract(self, d, ns):
samplerate = d.get_int('rate')[1]
self._doc['samplerate'] = samplerate
self._doc['channels'] = d.get_int('channels')[1]
if ns:
self._doc['audio_ns'] = ns
self._doc['duration']['samples'] = samplerate * ns // Gst.SECOND
self._callback(self)
def _finalize(self):
d = dict(
(n, self._rate.get_property(n))
for n in ('in', 'out', 'add', 'drop')
)
d['consistent'] = bool(d['in'] + d['add'] - d['drop'] == d['out'])
d['matching'] = d['in'] == d['out'] == self._doc['duration']['samples']
self._doc['samples'] = d
class Extractor(object):
def __init__(self, filename, full):
self.full = full
self.doc = {'video_ns': 0, 'audio_ns': 0, 'duration': {}}
self.mainloop = GLib.MainLoop()
self.pipeline = Gst.Pipeline()
# Create bus and connect several handlers
self.bus = self.pipeline.get_bus()
self.bus.add_signal_watch()
self.bus.connect('message::eos', self.on_eos)
self.bus.connect('message::error', self.on_error)
# Create elements
self.src = Gst.ElementFactory.make('filesrc', None)
self.dec = Gst.ElementFactory.make('decodebin', None)
# Set properties
self.src.set_property('location', filename)
# Connect handler for 'new-decoded-pad' signal
self.dec.connect('pad-added', self.on_pad_added)
self.dec.connect('no-more-pads', self.on_no_more_pads)
self.typefind = self.dec.get_by_name('typefind')
self.typefind.connect('have-type', self.on_have_type)
# Add elements to pipeline
self.pipeline.add(self.src)
self.pipeline.add(self.dec)
# Link elements
self.src.link(self.dec)
self.audio = None
self.video = None
self._killed = False
def run(self):
self.pipeline.set_state(Gst.State.PLAYING)
GLib.timeout_add(2000, self.on_timeout)
self.mainloop.run()
def on_timeout(self):
self.kill()
def kill(self):
if self._killed:
return
self._killed = True
self.pipeline.set_state(Gst.State.NULL)
ns = max(self.doc.pop('video_ns'), self.doc.pop('audio_ns'))
if self.doc['duration']:
self.doc['duration']['nanoseconds'] = ns
self.doc['duration']['seconds'] = float(ns) / Gst.SECOND
if self.video is not None:
self.doc['media'] = 'video'
elif self.audio is not None:
self.doc['media'] = 'audio'
else:
del self.doc['duration']
if self.video is not None:
self.doc['media'] = 'image'
self.mainloop.quit()
def on_have_type(self, element, prop, caps):
self.doc['content_type'] = caps.to_string().split(',')[0]
def link_pad(self, pad, name):
cls = {'audio': AudioBin, 'video': VideoBin}[name]
fakebin = cls(self.on_callback, self.doc)
self.pipeline.add(fakebin)
pad.link(fakebin.get_static_pad('sink'))
fakebin.set_state(Gst.State.PLAYING)
return fakebin
def on_pad_added(self, element, pad):
string = pad.get_current_caps().to_string()
if string.startswith('audio/'):
assert self.audio is None
self.audio = self.link_pad(pad, 'audio')
elif string.startswith('video/'):
assert self.video is None
self.video = self.link_pad(pad, 'video')
def on_no_more_pads(self, element):
self.need = set(filter(None, [self.audio, self.video]))
self.have = set()
def on_callback(self, fakebin):
self.have.add(fakebin)
if self.have == self.need and not self.full:
GLib.idle_add(self.kill)
def on_eos(self, bus, msg):
for fakebin in (self.audio, self.video):
if fakebin is not None:
try:
fakebin._finalize()
except Exception:
pass
self.kill()
def on_error(self, bus, msg):
#error = msg.parse_error()[1]
self.kill()
sys.exit(2)
extractor = Extractor(filename, options.full)
extractor.run()
print(json.dumps(extractor.doc, sort_keys=True, indent=4))
|