diff options
author | James Taylor <user234683@users.noreply.github.com> | 2019-12-19 21:33:54 -0800 |
---|---|---|
committer | James Taylor <user234683@users.noreply.github.com> | 2019-12-19 21:33:54 -0800 |
commit | b4406df9cf33c53b6e942e6a5c72d955f57c4b5f (patch) | |
tree | 4de0082ac9eb26a05188dd424835ea50b1483113 /youtube_dl/postprocessor | |
parent | b614fcdb8579ba29fccfa47eab1e2965cfb0beaa (diff) | |
parent | 6b7a1212e30b713453aa7d2b3a7122e97689dad0 (diff) | |
download | yt-local-b4406df9cf33c53b6e942e6a5c72d955f57c4b5f.tar.lz yt-local-b4406df9cf33c53b6e942e6a5c72d955f57c4b5f.tar.xz yt-local-b4406df9cf33c53b6e942e6a5c72d955f57c4b5f.zip |
Merge branch 'modular-data-extract'
Commits in this branch are prefixed with "Extraction:"
This branch refactors data extraction. All such functionality has been moved to the yt_data_extract module.
Responses from requests are given to the module and it parses them into a consistent, more useful format.
The dependency on youtube-dl has also been dropped and this functionality has been built from scratch for these reasons:
(1) I've noticed youtube-dl breaks more often than invidious (which uses watch page extraction built from scratch) in response to changes from Youtube, so I'm hoping what I wrote will also be less brittle.
(2) Such breakage is inconvenient because I have to manually merge the fixes since I had to make changes to youtube-dl to make it do things such as extracting related videos.
(3) I have no control over error handling and request pooling with youtube-dl, since it does all the requests (these would require intrusive changes I don't want to maintain).
(4) I will now be able to finally display the number of comments and whether comments are disabled without making additional requests.
Diffstat (limited to 'youtube_dl/postprocessor')
-rw-r--r-- | youtube_dl/postprocessor/__init__.py | 40 | ||||
-rw-r--r-- | youtube_dl/postprocessor/common.py | 69 | ||||
-rw-r--r-- | youtube_dl/postprocessor/embedthumbnail.py | 93 | ||||
-rw-r--r-- | youtube_dl/postprocessor/execafterdownload.py | 31 | ||||
-rw-r--r-- | youtube_dl/postprocessor/ffmpeg.py | 613 | ||||
-rw-r--r-- | youtube_dl/postprocessor/metadatafromtitle.py | 48 | ||||
-rw-r--r-- | youtube_dl/postprocessor/xattrpp.py | 79 |
7 files changed, 0 insertions, 973 deletions
diff --git a/youtube_dl/postprocessor/__init__.py b/youtube_dl/postprocessor/__init__.py deleted file mode 100644 index 3ea5183..0000000 --- a/youtube_dl/postprocessor/__init__.py +++ /dev/null @@ -1,40 +0,0 @@ -from __future__ import unicode_literals - -from .embedthumbnail import EmbedThumbnailPP -from .ffmpeg import ( - FFmpegPostProcessor, - FFmpegEmbedSubtitlePP, - FFmpegExtractAudioPP, - FFmpegFixupStretchedPP, - FFmpegFixupM3u8PP, - FFmpegFixupM4aPP, - FFmpegMergerPP, - FFmpegMetadataPP, - FFmpegVideoConvertorPP, - FFmpegSubtitlesConvertorPP, -) -from .xattrpp import XAttrMetadataPP -from .execafterdownload import ExecAfterDownloadPP -from .metadatafromtitle import MetadataFromTitlePP - - -def get_postprocessor(key): - return globals()[key + 'PP'] - - -__all__ = [ - 'EmbedThumbnailPP', - 'ExecAfterDownloadPP', - 'FFmpegEmbedSubtitlePP', - 'FFmpegExtractAudioPP', - 'FFmpegFixupM3u8PP', - 'FFmpegFixupM4aPP', - 'FFmpegFixupStretchedPP', - 'FFmpegMergerPP', - 'FFmpegMetadataPP', - 'FFmpegPostProcessor', - 'FFmpegSubtitlesConvertorPP', - 'FFmpegVideoConvertorPP', - 'MetadataFromTitlePP', - 'XAttrMetadataPP', -] diff --git a/youtube_dl/postprocessor/common.py b/youtube_dl/postprocessor/common.py deleted file mode 100644 index 599dd1d..0000000 --- a/youtube_dl/postprocessor/common.py +++ /dev/null @@ -1,69 +0,0 @@ -from __future__ import unicode_literals - -import os - -from ..utils import ( - PostProcessingError, - cli_configuration_args, - encodeFilename, -) - - -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. - - Optionally PostProcessor can use a list of additional command-line arguments - with self._configuration_args. - """ - - _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 is a list of the files - that can be deleted, and the second of which is the updated - information. - - In addition, this method may raise a PostProcessingError - exception if post processing fails. - """ - return [], information # by default, keep file and do nothing - - def try_utime(self, path, atime, mtime, errnote='Cannot update utime of file'): - try: - os.utime(encodeFilename(path), (atime, mtime)) - except Exception: - self._downloader.report_warning(errnote) - - def _configuration_args(self, default=[]): - return cli_configuration_args(self._downloader.params, 'postprocessor_args', default) - - -class AudioConversionError(PostProcessingError): - pass diff --git a/youtube_dl/postprocessor/embedthumbnail.py b/youtube_dl/postprocessor/embedthumbnail.py deleted file mode 100644 index 56be914..0000000 --- a/youtube_dl/postprocessor/embedthumbnail.py +++ /dev/null @@ -1,93 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - - -import os -import subprocess - -from .ffmpeg import FFmpegPostProcessor - -from ..utils import ( - check_executable, - encodeArgument, - encodeFilename, - PostProcessingError, - prepend_extension, - shell_quote -) - - -class EmbedThumbnailPPError(PostProcessingError): - pass - - -class EmbedThumbnailPP(FFmpegPostProcessor): - def __init__(self, downloader=None, already_have_thumbnail=False): - super(EmbedThumbnailPP, self).__init__(downloader) - self._already_have_thumbnail = already_have_thumbnail - - def run(self, info): - filename = info['filepath'] - temp_filename = prepend_extension(filename, 'temp') - - if not info.get('thumbnails'): - self._downloader.to_screen('[embedthumbnail] There aren\'t any thumbnails to embed') - return [], info - - thumbnail_filename = info['thumbnails'][-1]['filename'] - - if not os.path.exists(encodeFilename(thumbnail_filename)): - self._downloader.report_warning( - 'Skipping embedding the thumbnail because the file is missing.') - return [], info - - if info['ext'] == 'mp3': - options = [ - '-c', 'copy', '-map', '0', '-map', '1', - '-metadata:s:v', 'title="Album cover"', '-metadata:s:v', 'comment="Cover (Front)"'] - - self._downloader.to_screen('[ffmpeg] Adding thumbnail to "%s"' % filename) - - self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options) - - if not self._already_have_thumbnail: - os.remove(encodeFilename(thumbnail_filename)) - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - - elif info['ext'] in ['m4a', 'mp4']: - if not check_executable('AtomicParsley', ['-v']): - raise EmbedThumbnailPPError('AtomicParsley was not found. Please install.') - - cmd = [encodeFilename('AtomicParsley', True), - encodeFilename(filename, True), - encodeArgument('--artwork'), - encodeFilename(thumbnail_filename, True), - encodeArgument('-o'), - encodeFilename(temp_filename, True)] - - self._downloader.to_screen('[atomicparsley] Adding thumbnail to "%s"' % filename) - - if self._downloader.params.get('verbose', False): - self._downloader.to_screen('[debug] AtomicParsley command line: %s' % shell_quote(cmd)) - - p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE) - stdout, stderr = p.communicate() - - if p.returncode != 0: - msg = stderr.decode('utf-8', 'replace').strip() - raise EmbedThumbnailPPError(msg) - - if not self._already_have_thumbnail: - os.remove(encodeFilename(thumbnail_filename)) - # for formats that don't support thumbnails (like 3gp) AtomicParsley - # won't create to the temporary file - if b'No changes' in stdout: - self._downloader.report_warning('The file format doesn\'t support embedding a thumbnail') - else: - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - else: - raise EmbedThumbnailPPError('Only mp3 and m4a/mp4 are supported for thumbnail embedding for now.') - - return [], info diff --git a/youtube_dl/postprocessor/execafterdownload.py b/youtube_dl/postprocessor/execafterdownload.py deleted file mode 100644 index 64dabe7..0000000 --- a/youtube_dl/postprocessor/execafterdownload.py +++ /dev/null @@ -1,31 +0,0 @@ -from __future__ import unicode_literals - -import subprocess - -from .common import PostProcessor -from ..compat import compat_shlex_quote -from ..utils import ( - encodeArgument, - PostProcessingError, -) - - -class ExecAfterDownloadPP(PostProcessor): - def __init__(self, downloader, exec_cmd): - super(ExecAfterDownloadPP, self).__init__(downloader) - self.exec_cmd = exec_cmd - - def run(self, information): - cmd = self.exec_cmd - if '{}' not in cmd: - cmd += ' {}' - - cmd = cmd.replace('{}', compat_shlex_quote(information['filepath'])) - - self._downloader.to_screen('[exec] Executing command: %s' % cmd) - retCode = subprocess.call(encodeArgument(cmd), shell=True) - if retCode != 0: - raise PostProcessingError( - 'Command returned error code %d' % retCode) - - return [], information diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py deleted file mode 100644 index 757b496..0000000 --- a/youtube_dl/postprocessor/ffmpeg.py +++ /dev/null @@ -1,613 +0,0 @@ -from __future__ import unicode_literals - -import io -import os -import subprocess -import time -import re - - -from .common import AudioConversionError, PostProcessor - -from ..compat import ( - compat_subprocess_get_DEVNULL, -) -from ..utils import ( - encodeArgument, - encodeFilename, - get_exe_version, - is_outdated_version, - PostProcessingError, - prepend_extension, - shell_quote, - subtitles_filename, - dfxp2srt, - ISO639Utils, - replace_extension, -) - - -EXT_TO_OUT_FORMATS = { - 'aac': 'adts', - 'flac': 'flac', - 'm4a': 'ipod', - 'mka': 'matroska', - 'mkv': 'matroska', - 'mpg': 'mpeg', - 'ogv': 'ogg', - 'ts': 'mpegts', - 'wma': 'asf', - 'wmv': 'asf', -} -ACODECS = { - 'mp3': 'libmp3lame', - 'aac': 'aac', - 'flac': 'flac', - 'm4a': 'aac', - 'opus': 'libopus', - 'vorbis': 'libvorbis', - 'wav': None, -} - - -class FFmpegPostProcessorError(PostProcessingError): - pass - - -class FFmpegPostProcessor(PostProcessor): - def __init__(self, downloader=None): - PostProcessor.__init__(self, downloader) - self._determine_executables() - - def check_version(self): - if not self.available: - raise FFmpegPostProcessorError('ffmpeg or avconv not found. Please install one.') - - required_version = '10-0' if self.basename == 'avconv' else '1.0' - if is_outdated_version( - self._versions[self.basename], required_version): - warning = 'Your copy of %s is outdated, update %s to version %s or newer if you encounter any errors.' % ( - self.basename, self.basename, required_version) - if self._downloader: - self._downloader.report_warning(warning) - - @staticmethod - def get_versions(downloader=None): - return FFmpegPostProcessor(downloader)._versions - - def _determine_executables(self): - programs = ['avprobe', 'avconv', 'ffmpeg', 'ffprobe'] - prefer_ffmpeg = True - - self.basename = None - self.probe_basename = None - - self._paths = None - self._versions = None - if self._downloader: - prefer_ffmpeg = self._downloader.params.get('prefer_ffmpeg', True) - location = self._downloader.params.get('ffmpeg_location') - if location is not None: - if not os.path.exists(location): - self._downloader.report_warning( - 'ffmpeg-location %s does not exist! ' - 'Continuing without avconv/ffmpeg.' % (location)) - self._versions = {} - return - elif not os.path.isdir(location): - basename = os.path.splitext(os.path.basename(location))[0] - if basename not in programs: - self._downloader.report_warning( - 'Cannot identify executable %s, its basename should be one of %s. ' - 'Continuing without avconv/ffmpeg.' % - (location, ', '.join(programs))) - self._versions = {} - return None - location = os.path.dirname(os.path.abspath(location)) - if basename in ('ffmpeg', 'ffprobe'): - prefer_ffmpeg = True - - self._paths = dict( - (p, os.path.join(location, p)) for p in programs) - self._versions = dict( - (p, get_exe_version(self._paths[p], args=['-version'])) - for p in programs) - if self._versions is None: - self._versions = dict( - (p, get_exe_version(p, args=['-version'])) for p in programs) - self._paths = dict((p, p) for p in programs) - - if prefer_ffmpeg is False: - prefs = ('avconv', 'ffmpeg') - else: - prefs = ('ffmpeg', 'avconv') - for p in prefs: - if self._versions[p]: - self.basename = p - break - - if prefer_ffmpeg is False: - prefs = ('avprobe', 'ffprobe') - else: - prefs = ('ffprobe', 'avprobe') - for p in prefs: - if self._versions[p]: - self.probe_basename = p - break - - @property - def available(self): - return self.basename is not None - - @property - def executable(self): - return self._paths[self.basename] - - @property - def probe_available(self): - return self.probe_basename is not None - - @property - def probe_executable(self): - return self._paths[self.probe_basename] - - def get_audio_codec(self, path): - if not self.probe_available: - raise PostProcessingError('ffprobe or avprobe not found. Please install one.') - try: - cmd = [ - encodeFilename(self.probe_executable, True), - encodeArgument('-show_streams'), - encodeFilename(self._ffmpeg_filename_argument(path), True)] - if self._downloader.params.get('verbose', False): - self._downloader.to_screen('[debug] %s command line: %s' % (self.basename, shell_quote(cmd))) - handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE, stdin=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_multiple_files(self, input_paths, out_path, opts): - self.check_version() - - oldest_mtime = min( - os.stat(encodeFilename(path)).st_mtime for path in input_paths) - - opts += self._configuration_args() - - files_cmd = [] - for path in input_paths: - files_cmd.extend([ - encodeArgument('-i'), - encodeFilename(self._ffmpeg_filename_argument(path), True) - ]) - cmd = ([encodeFilename(self.executable, True), encodeArgument('-y')] + - files_cmd + - [encodeArgument(o) for o in opts] + - [encodeFilename(self._ffmpeg_filename_argument(out_path), True)]) - - if self._downloader.params.get('verbose', False): - self._downloader.to_screen('[debug] ffmpeg command line: %s' % shell_quote(cmd)) - p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=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) - self.try_utime(out_path, oldest_mtime, oldest_mtime) - - def run_ffmpeg(self, path, out_path, opts): - self.run_ffmpeg_multiple_files([path], out_path, opts) - - def _ffmpeg_filename_argument(self, fn): - # Always use 'file:' because the filename may contain ':' (ffmpeg - # interprets that as a protocol) or can start with '-' (-- is broken in - # ffmpeg, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details) - # Also leave '-' intact in order not to break streaming to stdout. - return 'file:' + fn if fn != '-' else 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 run_ffmpeg(self, path, out_path, codec, more_opts): - 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('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 = ['-bsf:a', 'aac_adtstoasc'] - elif filecodec in ['aac', 'flac', '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 += ['-q:a', self._preferredquality] - else: - more_opts += ['-b:a', self._preferredquality + 'k'] - else: - # We convert the audio (lossy if codec is lossy) - acodec = ACODECS[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 += ['-q:a', self._preferredquality] - else: - more_opts += ['-b:a', self._preferredquality + 'k'] - if self._preferredcodec == 'aac': - more_opts += ['-f', 'adts'] - if self._preferredcodec == 'm4a': - more_opts += ['-bsf:a', 'aac_adtstoasc'] - if self._preferredcodec == 'vorbis': - extension = 'ogg' - if self._preferredcodec == 'wav': - extension = 'wav' - more_opts += ['-f', 'wav'] - - prefix, sep, ext = path.rpartition('.') # not os.path.splitext, since the latter does not work on unicode in all setups - new_path = prefix + sep + extension - - information['filepath'] = new_path - information['ext'] = extension - - # If we download foo.mp3 and convert it to... foo.mp3, then don't delete foo.mp3, silly. - if (new_path == path or - (self._nopostoverwrites and os.path.exists(encodeFilename(new_path)))): - self._downloader.to_screen('[ffmpeg] Post-process file %s exists, skipping' % new_path) - return [], information - - try: - self._downloader.to_screen('[ffmpeg] Destination: ' + new_path) - self.run_ffmpeg(path, new_path, acodec, more_opts) - except AudioConversionError as e: - raise PostProcessingError( - 'audio conversion failed: ' + e.msg) - except Exception: - raise PostProcessingError('error running ' + self.basename) - - # Try to update the date time for extracted audio file. - if information.get('filetime') is not None: - self.try_utime( - new_path, time.time(), information['filetime'], - errnote='Cannot update utime of audio file') - - return [path], information - - -class FFmpegVideoConvertorPP(FFmpegPostProcessor): - def __init__(self, downloader=None, preferedformat=None): - super(FFmpegVideoConvertorPP, self).__init__(downloader) - self._preferedformat = preferedformat - - def run(self, information): - path = information['filepath'] - if information['ext'] == self._preferedformat: - self._downloader.to_screen('[ffmpeg] Not converting video file %s - already is in target format %s' % (path, self._preferedformat)) - return [], information - options = [] - if self._preferedformat == 'avi': - options.extend(['-c:v', 'libxvid', '-vtag', 'XVID']) - prefix, sep, ext = path.rpartition('.') - outpath = prefix + sep + self._preferedformat - self._downloader.to_screen('[' + 'ffmpeg' + '] Converting video from %s to %s, Destination: ' % (information['ext'], self._preferedformat) + outpath) - self.run_ffmpeg(path, outpath, options) - information['filepath'] = outpath - information['format'] = self._preferedformat - information['ext'] = self._preferedformat - return [path], information - - -class FFmpegEmbedSubtitlePP(FFmpegPostProcessor): - def run(self, information): - if information['ext'] not in ('mp4', 'webm', 'mkv'): - self._downloader.to_screen('[ffmpeg] Subtitles can only be embedded in mp4, webm or mkv files') - return [], information - subtitles = information.get('requested_subtitles') - if not subtitles: - self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to embed') - return [], information - - filename = information['filepath'] - - ext = information['ext'] - sub_langs = [] - sub_filenames = [] - webm_vtt_warn = False - - for lang, sub_info in subtitles.items(): - sub_ext = sub_info['ext'] - if ext != 'webm' or ext == 'webm' and sub_ext == 'vtt': - sub_langs.append(lang) - sub_filenames.append(subtitles_filename(filename, lang, sub_ext)) - else: - if not webm_vtt_warn and ext == 'webm' and sub_ext != 'vtt': - webm_vtt_warn = True - self._downloader.to_screen('[ffmpeg] Only WebVTT subtitles can be embedded in webm files') - - if not sub_langs: - return [], information - - input_files = [filename] + sub_filenames - - opts = [ - '-map', '0', - '-c', 'copy', - # Don't copy the existing subtitles, we may be running the - # postprocessor a second time - '-map', '-0:s', - ] - if information['ext'] == 'mp4': - opts += ['-c:s', 'mov_text'] - for (i, lang) in enumerate(sub_langs): - opts.extend(['-map', '%d:0' % (i + 1)]) - lang_code = ISO639Utils.short2long(lang) - if lang_code is not None: - opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code]) - - temp_filename = prepend_extension(filename, 'temp') - self._downloader.to_screen('[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 sub_filenames, information - - -class FFmpegMetadataPP(FFmpegPostProcessor): - def run(self, info): - metadata = {} - - def add(meta_list, info_list=None): - if not info_list: - info_list = meta_list - if not isinstance(meta_list, (list, tuple)): - meta_list = (meta_list,) - if not isinstance(info_list, (list, tuple)): - info_list = (info_list,) - for info_f in info_list: - if info.get(info_f) is not None: - for meta_f in meta_list: - metadata[meta_f] = info[info_f] - break - - add('title', ('track', 'title')) - add('date', 'upload_date') - add(('description', 'comment'), 'description') - add('purl', 'webpage_url') - add('track', 'track_number') - add('artist', ('artist', 'creator', 'uploader', 'uploader_id')) - add('genre') - add('album') - add('album_artist') - add('disc', 'disc_number') - - if not metadata: - self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add') - return [], info - - filename = info['filepath'] - temp_filename = prepend_extension(filename, 'temp') - in_filenames = [filename] - options = [] - - if info['ext'] == 'm4a': - options.extend(['-vn', '-acodec', 'copy']) - else: - options.extend(['-c', 'copy']) - - for (name, value) in metadata.items(): - options.extend(['-metadata', '%s=%s' % (name, value)]) - - chapters = info.get('chapters', []) - if chapters: - metadata_filename = replace_extension(filename, 'meta') - with io.open(metadata_filename, 'wt', encoding='utf-8') as f: - def ffmpeg_escape(text): - return re.sub(r'(=|;|#|\\|\n)', r'\\\1', text) - - metadata_file_content = ';FFMETADATA1\n' - for chapter in chapters: - metadata_file_content += '[CHAPTER]\nTIMEBASE=1/1000\n' - metadata_file_content += 'START=%d\n' % (chapter['start_time'] * 1000) - metadata_file_content += 'END=%d\n' % (chapter['end_time'] * 1000) - chapter_title = chapter.get('title') - if chapter_title: - metadata_file_content += 'title=%s\n' % ffmpeg_escape(chapter_title) - f.write(metadata_file_content) - in_filenames.append(metadata_filename) - options.extend(['-map_metadata', '1']) - - self._downloader.to_screen('[ffmpeg] Adding metadata to \'%s\'' % filename) - self.run_ffmpeg_multiple_files(in_filenames, temp_filename, options) - if chapters: - os.remove(metadata_filename) - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - return [], info - - -class FFmpegMergerPP(FFmpegPostProcessor): - def run(self, info): - filename = info['filepath'] - temp_filename = prepend_extension(filename, 'temp') - args = ['-c', 'copy', '-map', '0:v:0', '-map', '1:a:0'] - self._downloader.to_screen('[ffmpeg] Merging formats into "%s"' % filename) - self.run_ffmpeg_multiple_files(info['__files_to_merge'], temp_filename, args) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - return info['__files_to_merge'], info - - def can_merge(self): - # TODO: figure out merge-capable ffmpeg version - if self.basename != 'avconv': - return True - - required_version = '10-0' - if is_outdated_version( - self._versions[self.basename], required_version): - warning = ('Your copy of %s is outdated and unable to properly mux separate video and audio files, ' - 'youtube-dl will download single file media. ' - 'Update %s to version %s or newer to fix this.') % ( - self.basename, self.basename, required_version) - if self._downloader: - self._downloader.report_warning(warning) - return False - return True - - -class FFmpegFixupStretchedPP(FFmpegPostProcessor): - def run(self, info): - stretched_ratio = info.get('stretched_ratio') - if stretched_ratio is None or stretched_ratio == 1: - return [], info - - filename = info['filepath'] - temp_filename = prepend_extension(filename, 'temp') - - options = ['-c', 'copy', '-aspect', '%f' % stretched_ratio] - self._downloader.to_screen('[ffmpeg] Fixing aspect ratio in "%s"' % filename) - self.run_ffmpeg(filename, temp_filename, options) - - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - - return [], info - - -class FFmpegFixupM4aPP(FFmpegPostProcessor): - def run(self, info): - if info.get('container') != 'm4a_dash': - return [], info - - filename = info['filepath'] - temp_filename = prepend_extension(filename, 'temp') - - options = ['-c', 'copy', '-f', 'mp4'] - self._downloader.to_screen('[ffmpeg] Correcting container in "%s"' % filename) - self.run_ffmpeg(filename, temp_filename, options) - - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - - return [], info - - -class FFmpegFixupM3u8PP(FFmpegPostProcessor): - def run(self, info): - filename = info['filepath'] - if self.get_audio_codec(filename) == 'aac': - temp_filename = prepend_extension(filename, 'temp') - - options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc'] - self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename) - self.run_ffmpeg(filename, temp_filename, options) - - os.remove(encodeFilename(filename)) - os.rename(encodeFilename(temp_filename), encodeFilename(filename)) - return [], info - - -class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): - def __init__(self, downloader=None, format=None): - super(FFmpegSubtitlesConvertorPP, self).__init__(downloader) - self.format = format - - def run(self, info): - subs = info.get('requested_subtitles') - filename = info['filepath'] - new_ext = self.format - new_format = new_ext - if new_format == 'vtt': - new_format = 'webvtt' - if subs is None: - self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to convert') - return [], info - self._downloader.to_screen('[ffmpeg] Converting subtitles') - sub_filenames = [] - for lang, sub in subs.items(): - ext = sub['ext'] - if ext == new_ext: - self._downloader.to_screen( - '[ffmpeg] Subtitle file for %s is already in the requested format' % new_ext) - continue - old_file = subtitles_filename(filename, lang, ext) - sub_filenames.append(old_file) - new_file = subtitles_filename(filename, lang, new_ext) - - if ext in ('dfxp', 'ttml', 'tt'): - self._downloader.report_warning( - 'You have requested to convert dfxp (TTML) subtitles into another format, ' - 'which results in style information loss') - - dfxp_file = old_file - srt_file = subtitles_filename(filename, lang, 'srt') - - with open(dfxp_file, 'rb') as f: - srt_data = dfxp2srt(f.read()) - - with io.open(srt_file, 'wt', encoding='utf-8') as f: - f.write(srt_data) - old_file = srt_file - - subs[lang] = { - 'ext': 'srt', - 'data': srt_data - } - - if new_ext == 'srt': - continue - else: - sub_filenames.append(srt_file) - - self.run_ffmpeg(old_file, new_file, ['-f', new_format]) - - with io.open(new_file, 'rt', encoding='utf-8') as f: - subs[lang] = { - 'ext': new_ext, - 'data': f.read(), - } - - return sub_filenames, info diff --git a/youtube_dl/postprocessor/metadatafromtitle.py b/youtube_dl/postprocessor/metadatafromtitle.py deleted file mode 100644 index f5c14d9..0000000 --- a/youtube_dl/postprocessor/metadatafromtitle.py +++ /dev/null @@ -1,48 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import PostProcessor - - -class MetadataFromTitlePP(PostProcessor): - def __init__(self, downloader, titleformat): - super(MetadataFromTitlePP, self).__init__(downloader) - self._titleformat = titleformat - self._titleregex = (self.format_to_regex(titleformat) - if re.search(r'%\(\w+\)s', titleformat) - else titleformat) - - def format_to_regex(self, fmt): - r""" - Converts a string like - '%(title)s - %(artist)s' - to a regex like - '(?P<title>.+)\ \-\ (?P<artist>.+)' - """ - lastpos = 0 - regex = '' - # replace %(..)s with regex group and escape other string parts - for match in re.finditer(r'%\((\w+)\)s', fmt): - regex += re.escape(fmt[lastpos:match.start()]) - regex += r'(?P<' + match.group(1) + '>.+)' - lastpos = match.end() - if lastpos < len(fmt): - regex += re.escape(fmt[lastpos:]) - return regex - - def run(self, info): - title = info['title'] - match = re.match(self._titleregex, title) - if match is None: - self._downloader.to_screen( - '[fromtitle] Could not interpret title of video as "%s"' - % self._titleformat) - return [], info - for attribute, value in match.groupdict().items(): - info[attribute] = value - self._downloader.to_screen( - '[fromtitle] parsed %s: %s' - % (attribute, value if value is not None else 'NA')) - - return [], info diff --git a/youtube_dl/postprocessor/xattrpp.py b/youtube_dl/postprocessor/xattrpp.py deleted file mode 100644 index b0aed9c..0000000 --- a/youtube_dl/postprocessor/xattrpp.py +++ /dev/null @@ -1,79 +0,0 @@ -from __future__ import unicode_literals - -from .common import PostProcessor -from ..compat import compat_os_name -from ..utils import ( - hyphenate_date, - write_xattr, - XAttrMetadataError, - XAttrUnavailableError, -) - - -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). """ - - # 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', - } - - num_written = 0 - for xattrname, infoname in xattr_mapping.items(): - - value = info.get(infoname) - - if value: - if infoname == 'upload_date': - value = hyphenate_date(value) - - byte_value = value.encode('utf-8') - write_xattr(filename, xattrname, byte_value) - num_written += 1 - - return [], info - - except XAttrUnavailableError as e: - self._downloader.report_error(str(e)) - return [], info - - except XAttrMetadataError as e: - if e.reason == 'NO_SPACE': - self._downloader.report_warning( - 'There\'s no disk space left, disk quota exceeded or filesystem xattr limit exceeded. ' + - (('Some ' if num_written else '') + 'extended attributes are not written.').capitalize()) - elif e.reason == 'VALUE_TOO_LONG': - self._downloader.report_warning( - 'Unable to write extended attributes due to too long values.') - else: - msg = 'This filesystem doesn\'t support extended attributes. ' - if compat_os_name == 'nt': - msg += 'You need to use NTFS.' - else: - msg += '(You may have to enable them in your /etc/fstab)' - self._downloader.report_error(msg) - return [], info |