timeside-launch

#! /usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright (c) 2007-2014 Parisson SARL

# This file is part of TimeSide.

# TimeSide is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 2 of the License, or
# (at your option) any later version.

# TimeSide 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 General Public License for more details.

# You should have received a copy of the GNU General Public License
# along with TimeSide.  If not, see <http://www.gnu.org/licenses/>.

# Authors:
#   Paul Brossier <piem@piem.org>


"""This script runs a timeside pipeline on a collection of media files. The
pipeline can be configured using command line options or a configuration file.
"""

import sys, os.path

usage = "usage: %s [options] -C file.conf file1.wav [file2.wav ...]" % sys.argv[0]
usage += "\n help: %s -h" % sys.argv[0]

def parse_config(path):
    import yaml
    return yaml.load(open(path))

def parse_args():
    from optparse import OptionParser
    parser = OptionParser(usage = usage)
    parser.add_option("-v","--verbose",
            action="store_true", dest="verbose", default=False,
            help="be verbose")
    parser.add_option("-q","--quiet",
            action="store_true", dest="quiet", default=False,
            help="be quiet")
    parser.add_option("-C", "--conf", action = "store",
            dest = "config_file",
            help="configuration file",
            metavar = "<config_file>")

    parser.add_option("-s", "--samplerate", action = "store",
            dest = "samplerate", type = int,
            help="samplerate at which to run the pipeline",
            default = None,
            metavar = "<samplerate>")
    parser.add_option("-c", "--channels", action = "store",
            dest = "channels", type = int,
            help="number of channels to run the pipeline with",
            default = None,
            metavar = "<channels>")
    parser.add_option("-b", "--blocksize", action = "store",
            dest = "blocksize", type = int,
            help="blocksize at which to run the pipeline",
            default = None,
            metavar = "<blocksize>")

    parser.add_option("-a", "--analyzers", action = "store",
            dest = "analyzers", type = str,
            help="analyzers in the pipeline",
            default = 'Level',
            metavar = "<analyzers>")
    parser.add_option("-g", "--graphers", action = "store",
            dest = "graphers", type = str,
            help="graphers in the pipeline",
            default = [],
            metavar = "<graphers>")
    parser.add_option("-e", "--encoders", action = "store",
            dest = "encoders", type = str,
            help="encoders in the pipeline",
            default = [],
            metavar = "<encoders>")

    parser.add_option("-R", "--results-formats", action = "store",
            dest = "r_formats", type = str,
            help = "list of results output formats for the analyzers results",
            default = 'yaml',
            metavar = "<formats>")
    parser.add_option("-I", "--images-formats", action = "store",
            dest = "i_formats", type = str,
            help = "list of graph output formats for the analyzers results",
            default = 'png',
            metavar = "<formats>")

    parser.add_option("-o", "--ouput-directory", action = "store",
            dest = "outputdir", type = str,
            help="output directory",
            default = None,
            metavar = "<outputdir>")

    (options, args) = parser.parse_args()

    if options.analyzers:
        options.analyzers = options.analyzers.split(',')
    if options.graphers:
        options.graphers = options.graphers.split(',')
    if options.encoders:
        options.encoders = options.encoders.split(',')
    if options.r_formats:
        options.r_formats = options.r_formats.split(',')
        known_r_formats = ['json', 'yaml', 'xml', 'hdf5']
        for f in options.r_formats:
            if f not in known_r_formats:
                raise ValueError("unknown result format %s, possible values %s" % (f, known_r_formats))
    if options.i_formats:
        options.i_formats = options.i_formats.split(',')
        known_i_formats = ['png', 'jpg', 'tiff']
        for f in options.i_formats:
            if f not in known_i_formats:
                raise ValueError("unknown graph output format %s, possible values %s" % (f, known_i_formats))

    if options.config_file:
        if not os.path.isfile(options.config_file):
            print("ERROR: configuration file not found:", options.config_file)
            sys.exit(1)
        config = parse_config(options.config_file)
        for key in config:
            if not hasattr(options, key) or not getattr(options,key):
                setattr(options, key, config[key])

    if options.outputdir == None:
        import tempfile
        options.outputdir = tempfile.mkdtemp('-timeside')
    if not os.path.isdir(options.outputdir):
        os.makedirs(options.outputdir)

    return options, args

if __name__ == '__main__':
    options, args = parse_args()
    # load timeside after parse_args, to avoid gstreamer hijacking
    import timeside.core

    if len(args) == 0:
        print(usage)
        sys.exit(1)

    if not options.quiet:
        for a in dir(options):
            if not callable(getattr(options,a)) and not a.startswith('_'):
                print(a + ":", getattr(options,a))

    verbose = options.verbose and not options.quiet
    channels = options.channels
    samplerate = options.samplerate
    blocksize = options.blocksize
    outputdir = options.outputdir
    r_formats = options.r_formats
    i_formats = options.i_formats
    analyzers = options.analyzers
    graphers = options.graphers
    encoders = options.encoders

    all_decoders = timeside.core.processor.processors(timeside.core.api.IDecoder)
    all_analyzers = timeside.core.processor.processors(timeside.core.api.IAnalyzer)
    all_graphers = timeside.core.processor.processors(timeside.core.api.IGrapher)
    all_encoders = timeside.core.processor.processors(timeside.core.api.IEncoder)

    def match_id_or_class(id_or_class, processors):
        matches = list(filter(lambda x: x.__name__ == id_or_class, processors))
        matches += list(filter(lambda x: x.id() == id_or_class , processors))
        matches += list(filter(lambda x: hasattr(x,'file_extension') \
                and x.file_extension() == id_or_class , processors))
        if not len(matches):
            msg  = 'ERROR: could not find \'%s\'.' % id_or_class
            msg += ' possible values:' + repr(processors)
            raise ValueError(msg)
        elif len(matches) > 1:
            msg  = 'ERROR: too many matches for \'%s\'.' % id_or_class
            msg += ' matched values:' + repr(matches)
            raise ValueError(msg)
        else:
            return matches[0]

    def match_analyzer(id_or_class):
        return match_id_or_class(id_or_class, all_analyzers)
    def match_grapher(id_or_class):
        return match_id_or_class(id_or_class, all_graphers)
    def match_encoder(id_or_class):
        return match_id_or_class(id_or_class, all_encoders)

    # create instances of analyzers and graphers
    analyzers = map(match_analyzer, analyzers)
    graphers = map(match_grapher, graphers)
    encoders = map(match_encoder, encoders)

    def process_file(path):
        #import uuid
        #from timeside.plugins.decoder.utils import get_uri
        from timeside.core import get_processor

        decoder = get_processor('file_decoder')(path)

        #file_uuid = str(uuid.uuid5(uuid.NAMESPACE_URL, get_uri(path) ))
        file_uuid = decoder.sha1

        result_dir = os.path.join(outputdir, file_uuid)
        if not os.path.isdir(result_dir):
            os.makedirs(result_dir)


        #pipe.setup(channels = channels, samplerate = samplerate, blocksize = blocksize)
        pipe = decoder
        _analyzers = [a() for a in analyzers]
        _graphers = [g() for g in graphers]
        _encoders = [e(os.path.join(result_dir, file_uuid + '.' + e.file_extension())) for e in encoders]

        for a in _analyzers:
            pipe = pipe | a
        for g in _graphers:
            pipe = pipe | g
        for e in _encoders:
            pipe = pipe | e
        pipe.run(channels = channels, samplerate = samplerate, blocksize = blocksize)

        if len(_analyzers):
            for res_uuid, result in pipe.results.items():

                for f in r_formats:
                    result_path = os.path.join(result_dir, res_uuid + '.' + f)
                    getattr(result,'to_'+f)(result_path)
                    if verbose : print('saved', result_path)
        if len(_graphers):
            for g in _graphers:
                for f in i_formats:
                    graph_path = os.path.join(result_dir, g.uuid() + '.' + f)
                    g.render(graph_path)
                    if verbose : print('saved', graph_path)
        if len(_encoders):
            for e in _encoders:
                if verbose : print('saved', e.filename)

    for path in args:
        process_file (path)