yt-dlp/youtube_dl/PostProcessor.py

import os
import subprocess
import sys
import time


from .utils import (
    compat_subprocess_get_DEVNULL,
    encodeFilename,
    PostProcessingError,
    shell_quote,
    subtitles_filename,
)


class PostProcessor(object):
    """Post Processor class.

    PostProcessor objects can be added to downloaders with their
    add_post_processor() method. When the downloader has finished a
    successful download, it will take its internal chain of PostProcessors
    and start calling the run() method on each one of them, first with
    an initial argument and then with the returned value of the previous
    PostProcessor.

    The chain will be stopped if one of them ever returns None or the end
    of the chain is reached.

    PostProcessor objects follow a "mutual registration" process similar
    to InfoExtractor objects.
    """

    _downloader = None

    def __init__(self, downloader=None):
        self._downloader = downloader

    def set_downloader(self, downloader):
        """Sets the downloader for this PP."""
        self._downloader = downloader

    def run(self, information):
        """Run the PostProcessor.

        The "information" argument is a dictionary like the ones
        composed by InfoExtractors. The only difference is that this
        one has an extra field called "filepath" that points to the
        downloaded file.

        This method returns a tuple, the first element of which describes
        whether the original file should be kept (i.e. not deleted - None for
        no preference), and the second of which is the updated information.

        In addition, this method may raise a PostProcessingError
        exception if post processing fails.
        """
        return None, information # by default, keep file and do nothing

class FFmpegPostProcessorError(PostProcessingError):
    pass

class AudioConversionError(PostProcessingError):
    pass


class FFmpegPostProcessor(PostProcessor):
    def __init__(self,downloader=None):
        PostProcessor.__init__(self, downloader)
        self._exes = self.detect_executables()

    @staticmethod
    def detect_executables():
        def executable(exe):
            try:
                subprocess.Popen([exe, '-version'], stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate()
            except OSError:
                return False
            return exe
        programs = ['avprobe', 'avconv', 'ffmpeg', 'ffprobe']
        return dict((program, executable(program)) for program in programs)

    def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
        if not self._exes['ffmpeg'] and not self._exes['avconv']:
            raise FFmpegPostProcessorError(u'ffmpeg or avconv not found. Please install one.')

        files_cmd = []
        for path in input_paths:
            files_cmd.extend(['-i', encodeFilename(path)])
        cmd = ([self._exes['avconv'] or self._exes['ffmpeg'], '-y'] + files_cmd
               + opts +
               [encodeFilename(self._ffmpeg_filename_argument(out_path))])

        if self._downloader.params.get('verbose', False):
            self._downloader.to_screen(u'[debug] ffmpeg command line: %s' % shell_quote(cmd))
        p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
        stdout,stderr = p.communicate()
        if p.returncode != 0:
            stderr = stderr.decode('utf-8', 'replace')
            msg = stderr.strip().split('\n')[-1]
            raise FFmpegPostProcessorError(msg)

    def run_ffmpeg(self, path, out_path, opts):
        self.run_ffmpeg_multiple_files([path], out_path, opts)

    def _ffmpeg_filename_argument(self, fn):
        # ffmpeg broke --, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details
        if fn.startswith(u'-'):
            return u'./' + fn
        return fn


class FFmpegExtractAudioPP(FFmpegPostProcessor):
    def __init__(self, downloader=None, preferredcodec=None, preferredquality=None, nopostoverwrites=False):
        FFmpegPostProcessor.__init__(self, downloader)
        if preferredcodec is None:
            preferredcodec = 'best'
        self._preferredcodec = preferredcodec
        self._preferredquality = preferredquality
        self._nopostoverwrites = nopostoverwrites

    def get_audio_codec(self, path):
        if not self._exes['ffprobe'] and not self._exes['avprobe']:
            raise PostProcessingError(u'ffprobe or avprobe not found. Please install one.')
        try:
            cmd = [self._exes['avprobe'] or self._exes['ffprobe'], '-show_streams', encodeFilename(self._ffmpeg_filename_argument(path))]
            handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE)
            output = handle.communicate()[0]
            if handle.wait() != 0:
                return None
        except (IOError, OSError):
            return None
        audio_codec = None
        for line in output.decode('ascii', 'ignore').split('\n'):
            if line.startswith('codec_name='):
                audio_codec = line.split('=')[1].strip()
            elif line.strip() == 'codec_type=audio' and audio_codec is not None:
                return audio_codec
        return None

    def run_ffmpeg(self, path, out_path, codec, more_opts):
        if not self._exes['ffmpeg'] and not self._exes['avconv']:
            raise AudioConversionError('ffmpeg or avconv not found. Please install one.')
        if codec is None:
            acodec_opts = []
        else:
            acodec_opts = ['-acodec', codec]
        opts = ['-vn'] + acodec_opts + more_opts
        try:
            FFmpegPostProcessor.run_ffmpeg(self, path, out_path, opts)
        except FFmpegPostProcessorError as err:
            raise AudioConversionError(err.msg)

    def run(self, information):
        path = information['filepath']

        filecodec = self.get_audio_codec(path)
        if filecodec is None:
            raise PostProcessingError(u'WARNING: unable to obtain file audio codec with ffprobe')

        more_opts = []
        if self._preferredcodec == 'best' or self._preferredcodec == filecodec or (self._preferredcodec == 'm4a' and filecodec == 'aac'):
            if filecodec == 'aac' and self._preferredcodec in ['m4a', 'best']:
                # Lossless, but in another container
                acodec = 'copy'
                extension = 'm4a'
                more_opts = [self._exes['avconv'] and '-bsf:a' or '-absf', 'aac_adtstoasc']
            elif filecodec in ['aac', 'mp3', 'vorbis', 'opus']:
                # Lossless if possible
                acodec = 'copy'
                extension = filecodec
                if filecodec == 'aac':
                    more_opts = ['-f', 'adts']
                if filecodec == 'vorbis':
                    extension = 'ogg'
            else:
                # MP3 otherwise.
                acodec = 'libmp3lame'
                extension = 'mp3'
                more_opts = []
                if self._preferredquality is not None:
                    if int(self._preferredquality) < 10:
                        more_opts += [self._exes['avconv'] and '-q:a' or '-aq', self._preferredquality]
                    else:
                        more_opts += [self._exes['avconv'] and '-b:a' or '-ab', self._preferredquality + 'k']
        else:
            # We convert the audio (lossy)
            acodec = {'mp3': 'libmp3lame', 'aac': 'aac', 'm4a': 'aac', 'opus': 'opus', 'vorbis': 'libvorbis', 'wav': None}[self._preferredcodec]
            extension = self._preferredcodec
            more_opts = []
            if self._preferredquality is not None:
                # The opus codec doesn't support the -aq option
                if int(self._preferredquality) < 10 and extension != 'opus':
                    more_opts += [self._exes['avconv'] and '-q:a' or '-aq', self._preferredquality]
                else:
                    more_opts += [self._exes['avconv'] and '-b:a' or '-ab', self._preferredquality + 'k']
            if self._preferredcodec == 'aac':
                more_opts += ['-f', 'adts']
            if self._preferredcodec == 'm4a':
                more_opts += [self._exes['avconv'] and '-bsf:a' or '-absf', 'aac_adtstoasc']
            if self._preferredcodec == 'vorbis':
                extension = 'ogg'
            if self._preferredcodec == 'wav':
                extension = 'wav'
                more_opts += ['-f', 'wav']

        prefix, sep, ext = path.rpartition(u'.') # not os.path.splitext, since the latter does not work on unicode in all setups
        new_path = prefix + sep + extension

        # If we download foo.mp3 and convert it to... foo.mp3, then don't delete foo.mp3, silly.
        if new_path == path:
            self._nopostoverwrites = True

        try:
            if self._nopostoverwrites and os.path.exists(encodeFilename(new_path)):
                self._downloader.to_screen(u'[youtube] Post-process file %s exists, skipping' % new_path)
            else:
                self._downloader.to_screen(u'[' + (self._exes['avconv'] and 'avconv' or 'ffmpeg') + '] Destination: ' + new_path)
                self.run_ffmpeg(path, new_path, acodec, more_opts)
        except:
            etype,e,tb = sys.exc_info()
            if isinstance(e, AudioConversionError):
                msg = u'audio conversion failed: ' + e.msg
            else:
                msg = u'error running ' + (self._exes['avconv'] and 'avconv' or 'ffmpeg')
            raise PostProcessingError(msg)

        # Try to update the date time for extracted audio file.
        if information.get('filetime') is not None:
            try:
                os.utime(encodeFilename(new_path), (time.time(), information['filetime']))
            except:
                self._downloader.report_warning(u'Cannot update utime of audio file')

        information['filepath'] = new_path
        return self._nopostoverwrites,information


class FFmpegVideoConvertor(FFmpegPostProcessor):
    def __init__(self, downloader=None,preferedformat=None):
        super(FFmpegVideoConvertor, self).__init__(downloader)
        self._preferedformat=preferedformat

    def run(self, information):
        path = information['filepath']
        prefix, sep, ext = path.rpartition(u'.')
        outpath = prefix + sep + self._preferedformat
        if information['ext'] == self._preferedformat:
            self._downloader.to_screen(u'[ffmpeg] Not converting video file %s - already is in target format %s' % (path, self._preferedformat))
            return True,information
        self._downloader.to_screen(u'['+'ffmpeg'+'] Converting video from %s to %s, Destination: ' % (information['ext'], self._preferedformat) +outpath)
        self.run_ffmpeg(path, outpath, [])
        information['filepath'] = outpath
        information['format'] = self._preferedformat
        information['ext'] = self._preferedformat
        return False,information


class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
    # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
    _lang_map = {
        'aa': 'aar',
        'ab': 'abk',
        'ae': 'ave',
        'af': 'afr',
        'ak': 'aka',
        'am': 'amh',
        'an': 'arg',
        'ar': 'ara',
        'as': 'asm',
        'av': 'ava',
        'ay': 'aym',
        'az': 'aze',
        'ba': 'bak',
        'be': 'bel',
        'bg': 'bul',
        'bh': 'bih',
        'bi': 'bis',
        'bm': 'bam',
        'bn': 'ben',
        'bo': 'bod',
        'br': 'bre',
        'bs': 'bos',
        'ca': 'cat',
        'ce': 'che',
        'ch': 'cha',
        'co': 'cos',
        'cr': 'cre',
        'cs': 'ces',
        'cu': 'chu',
        'cv': 'chv',
        'cy': 'cym',
        'da': 'dan',
        'de': 'deu',
        'dv': 'div',
        'dz': 'dzo',
        'ee': 'ewe',
        'el': 'ell',
        'en': 'eng',
        'eo': 'epo',
        'es': 'spa',
        'et': 'est',
        'eu': 'eus',
        'fa': 'fas',
        'ff': 'ful',
        'fi': 'fin',
        'fj': 'fij',
        'fo': 'fao',
        'fr': 'fra',
        'fy': 'fry',
        'ga': 'gle',
        'gd': 'gla',
        'gl': 'glg',
        'gn': 'grn',
        'gu': 'guj',
        'gv': 'glv',
        'ha': 'hau',
        'he': 'heb',
        'hi': 'hin',
        'ho': 'hmo',
        'hr': 'hrv',
        'ht': 'hat',
        'hu': 'hun',
        'hy': 'hye',
        'hz': 'her',
        'ia': 'ina',
        'id': 'ind',
        'ie': 'ile',
        'ig': 'ibo',
        'ii': 'iii',
        'ik': 'ipk',
        'io': 'ido',
        'is': 'isl',
        'it': 'ita',
        'iu': 'iku',
        'ja': 'jpn',
        'jv': 'jav',
        'ka': 'kat',
        'kg': 'kon',
        'ki': 'kik',
        'kj': 'kua',
        'kk': 'kaz',
        'kl': 'kal',
        'km': 'khm',
        'kn': 'kan',
        'ko': 'kor',
        'kr': 'kau',
        'ks': 'kas',
        'ku': 'kur',
        'kv': 'kom',
        'kw': 'cor',
        'ky': 'kir',
        'la': 'lat',
        'lb': 'ltz',
        'lg': 'lug',
        'li': 'lim',
        'ln': 'lin',
        'lo': 'lao',
        'lt': 'lit',
        'lu': 'lub',
        'lv': 'lav',
        'mg': 'mlg',
        'mh': 'mah',
        'mi': 'mri',
        'mk': 'mkd',
        'ml': 'mal',
        'mn': 'mon',
        'mr': 'mar',
        'ms': 'msa',
        'mt': 'mlt',
        'my': 'mya',
        'na': 'nau',
        'nb': 'nob',
        'nd': 'nde',
        'ne': 'nep',
        'ng': 'ndo',
        'nl': 'nld',
        'nn': 'nno',
        'no': 'nor',
        'nr': 'nbl',
        'nv': 'nav',
        'ny': 'nya',
        'oc': 'oci',
        'oj': 'oji',
        'om': 'orm',
        'or': 'ori',
        'os': 'oss',
        'pa': 'pan',
        'pi': 'pli',
        'pl': 'pol',
        'ps': 'pus',
        'pt': 'por',
        'qu': 'que',
        'rm': 'roh',
        'rn': 'run',
        'ro': 'ron',
        'ru': 'rus',
        'rw': 'kin',
        'sa': 'san',
        'sc': 'srd',
        'sd': 'snd',
        'se': 'sme',
        'sg': 'sag',
        'si': 'sin',
        'sk': 'slk',
        'sl': 'slv',
        'sm': 'smo',
        'sn': 'sna',
        'so': 'som',
        'sq': 'sqi',
        'sr': 'srp',
        'ss': 'ssw',
        'st': 'sot',
        'su': 'sun',
        'sv': 'swe',
        'sw': 'swa',
        'ta': 'tam',
        'te': 'tel',
        'tg': 'tgk',
        'th': 'tha',
        'ti': 'tir',
        'tk': 'tuk',
        'tl': 'tgl',
        'tn': 'tsn',
        'to': 'ton',
        'tr': 'tur',
        'ts': 'tso',
        'tt': 'tat',
        'tw': 'twi',
        'ty': 'tah',
        'ug': 'uig',
        'uk': 'ukr',
        'ur': 'urd',
        'uz': 'uzb',
        've': 'ven',
        'vi': 'vie',
        'vo': 'vol',
        'wa': 'wln',
        'wo': 'wol',
        'xh': 'xho',
        'yi': 'yid',
        'yo': 'yor',
        'za': 'zha',
        'zh': 'zho',
        'zu': 'zul',
    }

    def __init__(self, downloader=None, subtitlesformat='srt'):
        super(FFmpegEmbedSubtitlePP, self).__init__(downloader)
        self._subformat = subtitlesformat

    @classmethod
    def _conver_lang_code(cls, code):
        """Convert language code from ISO 639-1 to ISO 639-2/T"""
        return cls._lang_map.get(code[:2])

    def run(self, information):
        if information['ext'] != u'mp4':
            self._downloader.to_screen(u'[ffmpeg] Subtitles can only be embedded in mp4 files')
            return True, information
        if not information.get('subtitles'):
            self._downloader.to_screen(u'[ffmpeg] There aren\'t any subtitles to embed') 
            return True, information

        sub_langs = [key for key in information['subtitles']]
        filename = information['filepath']
        input_files = [filename] + [subtitles_filename(filename, lang, self._subformat) for lang in sub_langs]

        opts = ['-map', '0:0', '-map', '0:1', '-c:v', 'copy', '-c:a', 'copy']
        for (i, lang) in enumerate(sub_langs):
            opts.extend(['-map', '%d:0' % (i+1), '-c:s:%d' % i, 'mov_text'])
            lang_code = self._conver_lang_code(lang)
            if lang_code is not None:
                opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code])
        opts.extend(['-f', 'mp4'])

        temp_filename = filename + u'.temp'
        self._downloader.to_screen(u'[ffmpeg] Embedding subtitles in \'%s\'' % filename)
        self.run_ffmpeg_multiple_files(input_files, temp_filename, opts)
        os.remove(encodeFilename(filename))
        os.rename(encodeFilename(temp_filename), encodeFilename(filename))

        return True, information


class FFmpegMetadataPP(FFmpegPostProcessor):
    def run(self, info):
        metadata = {}
        if info.get('title') is not None:
            metadata['title'] = info['title']
        if info.get('upload_date') is not None:
            metadata['date'] = info['upload_date']
        if info.get('uploader') is not None:
            metadata['artist'] = info['uploader']
        elif info.get('uploader_id') is not None:
            metadata['artist'] = info['uploader_id']

        if not metadata:
            self._downloader.to_screen(u'[ffmpeg] There isn\'t any metadata to add')
            return True, info

        filename = info['filepath']
        ext = os.path.splitext(filename)[1][1:]
        temp_filename = filename + u'.temp'

        options = ['-c', 'copy']
        for (name, value) in metadata.items():
            options.extend(['-metadata', '%s=%s' % (name, value)])
        options.extend(['-f', ext])

        self._downloader.to_screen(u'[ffmpeg] Adding metadata to \'%s\'' % filename)
        self.run_ffmpeg(filename, temp_filename, options)
        os.remove(encodeFilename(filename))
        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
        return True, info


class XAttrMetadataPP(PostProcessor):

    #
    # More info about extended attributes for media:
    #   http://freedesktop.org/wiki/CommonExtendedAttributes/
    #   http://www.freedesktop.org/wiki/PhreedomDraft/
    #   http://dublincore.org/documents/usageguide/elements.shtml
    #
    # TODO:
    #  * capture youtube keywords and put them in 'user.dublincore.subject' (comma-separated)
    #  * figure out which xattrs can be used for 'duration', 'thumbnail', 'resolution'
    #

    def run(self, info):
        """ Set extended attributes on downloaded file (if xattr support is found). """

        from .utils import hyphenate_date

        # This mess below finds the best xattr tool for the job and creates a
        # "write_xattr" function.
        try:
            # try the pyxattr module...
            import xattr
            def write_xattr(path, key, value):
                return xattr.setxattr(path, key, value)

        except ImportError:

            if os.name == 'posix':
                def which(bin):
                    for dir in os.environ["PATH"].split(":"):
                        path = os.path.join(dir, bin)
                        if os.path.exists(path):
                            return path

                user_has_setfattr = which("setfattr")
                user_has_xattr    = which("xattr")

                if user_has_setfattr or user_has_xattr:

                    def write_xattr(path, key, value):
                        import errno
                        potential_errors = {
                            # setfattr: /tmp/blah: Operation not supported
                            "Operation not supported": errno.EOPNOTSUPP,
                            # setfattr: ~/blah: No such file or directory
                            # xattr: No such file: ~/blah
                            "No such file": errno.ENOENT,
                        }

                        if user_has_setfattr:
                            cmd = ['setfattr', '-n', key, '-v', value, path]
                        elif user_has_xattr:
                            cmd = ['xattr', '-w', key, value, path]

                        try:
                            output = subprocess.check_output(cmd, stderr=subprocess.STDOUT)
                        except subprocess.CalledProcessError as e:
                            errorstr = e.output.strip().decode()
                            for potential_errorstr, potential_errno in potential_errors.items():
                                if errorstr.find(potential_errorstr) > -1:
                                    e = OSError(potential_errno, potential_errorstr)
                                    e.__cause__ = None
                                    raise e
                            raise # Reraise unhandled error

                else:
                    # On Unix, and can't find pyxattr, setfattr, or xattr.
                    if sys.platform.startswith('linux'):
                        self._downloader.report_error("Couldn't find a tool to set the xattrs. Install either the python 'pyxattr' or 'xattr' modules, or the GNU 'attr' package (which contains the 'setfattr' tool).")
                    elif sys.platform == 'darwin':
                        self._downloader.report_error("Couldn't find a tool to set the xattrs. Install either the python 'xattr' module, or the 'xattr' binary.")
            else:
                # Write xattrs to NTFS Alternate Data Streams: http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
                def write_xattr(path, key, value):
                    assert(key.find(":") < 0)
                    assert(path.find(":") < 0)
                    assert(os.path.exists(path))

                    f = open(path+":"+key, "w")
                    f.write(value)
                    f.close()

        # Write the metadata to the file's xattrs
        self._downloader.to_screen('[metadata] Writing metadata to file\'s xattrs...')

        filename = info['filepath']

        try:
            xattr_mapping = {
                'user.xdg.referrer.url':       'webpage_url',
                # 'user.xdg.comment':            'description',
                'user.dublincore.title':       'title',
                'user.dublincore.date':        'upload_date',
                'user.dublincore.description': 'description',
                'user.dublincore.contributor': 'uploader',
                'user.dublincore.format':      'format',
            }

            for xattrname, infoname in xattr_mapping.items():

                value = info.get(infoname)

                if value:
                    if infoname == "upload_date":
                        value = hyphenate_date(value)

                    write_xattr(filename, xattrname, value)

            return True, info

        except OSError:
            self._downloader.report_error("This filesystem doesn't support extended attributes. (You may have to enable them in your /etc/fstab)")
            return False, info