getduration.py

from __future__ import print_function

import sys
import os

import gi
gi.require_version('Gst', '1.0')
from gi.repository import GObject, Gst


def on_message(bus, message, udata):
    pipeline, loop = udata

    if message.type == Gst.MessageType.EOS:
        pipeline.set_state(Gst.State.NULL)
        loop.quit()
    elif message.type == Gst.MessageType.ERROR:
        print(message.parse_error())
        pipeline.set_state(Gst.State.NULL)
        loop.quit()

    return True


def on_pad_added(element, pad, udata):
    pipeline, sink = udata
    pad.link(sink.get_static_pad('sink'))

    # Get the duration of the source file.
    duration, success = pad.query_duration(Gst.Format.TIME)
    print(duration, success)


def decode(input_file):
    loop = GObject.MainLoop()

    pipeline = Gst.Pipeline()
    uridecodebin = Gst.ElementFactory.make('uridecodebin')
    sink = Gst.ElementFactory.make('fakesink')
    pipeline.add(uridecodebin)
    pipeline.add(sink)

    uridecodebin.link(sink)

    uridecodebin.set_property('uri', input_file)
    uridecodebin.connect('pad-added', on_pad_added, (pipeline, sink))

    bus = pipeline.get_bus()
    bus.add_watch(0, on_message, (pipeline, loop))

    pipeline.set_state(Gst.State.PLAYING)
    loop.run()


if __name__ == '__main__':
    Gst.init()
    input_file = 'file://' + os.path.abspath(sys.argv[1])
    decode(input_file)

This short program just opens a media file and queries its duration. It works for some files, but not for others.

Specifically, here's a file for which it works, image.mp3, and a file for which it fails, named full.mp3. Here's what happens when the script is run on the two files:

$ python getduration.py ../beets/test/rsrc/image.mp3 
True 1044897959
$ python getduration.py ../beets/test/rsrc/full.mp3
False 0

... indicating that the query_duration() call fails for the second file.

The two files are very similar: short MP3s containing one second of silence. The questions are: Why does the query fail on one file but not the other? Is there anything we can do to make it work for all files?

I got these results using GStreamer 1.4.5 on Arch Linux. The result is the same on both Python 2 and Python 3.