Merge dc179db221331a6e63c7e1e7bd569fc07b0841ec into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

youtube_dl/YoutubeDL.py

@@ -122,6 +122,7 @@ from .postprocessor import (
     FFmpegFixupM4aPP,
     FFmpegFixupStretchedPP,
     FFmpegMergerPP,
+    FFmpegConcatPP,
     FFmpegPostProcessor,
     get_postprocessor,
 )
@@ -1206,6 +1207,20 @@ class YoutubeDL(object):
             entry_result = self.__process_iterable_entry(entry, download, extra)
             # TODO: skip failed (empty) entries?
             playlist_results.append(entry_result)
+
+        if self.params.get('concat_playlist', False):
+            concat_pp = FFmpegConcatPP(self)
+            ie_result['__postprocessors'] = [concat_pp]
+            ie_result['__files_to_concat'] = list(map(lambda e: e['_filename'], entries))
+
+            filename_wo_ext = self.prepare_filename(ie_result)
+
+            # Ensure filename always has a correct extension for successful merge
+            ie_result['ext'] = self.params.get('merge_output_format') or entries[0]['ext']
+            ie_result['_filename'] = filename = '%s.%s' % (filename_wo_ext, ie_result['ext'])
+
+            self.post_process(filename, ie_result)
+
         ie_result['entries'] = playlist_results
         self.to_screen('[download] Finished downloading playlist: %s' % playlist)
         return ie_result
@@ -1858,6 +1873,7 @@ class YoutubeDL(object):
                 new_info = dict(info_dict)
                 new_info.update(format)
                 self.process_info(new_info)
+                format.update(new_info)
         # We update the info dict with the best quality format (backwards compatibility)
         info_dict.update(formats_to_download[-1])
         return info_dict

youtube_dl/__init__.py

@@ -411,6 +411,7 @@ def _real_main(argv=None):
         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
         'encoding': opts.encoding,
         'extract_flat': opts.extract_flat,
+        'concat_playlist': opts.concat_playlist,
         'mark_watched': opts.mark_watched,
         'merge_output_format': opts.merge_output_format,
         'postprocessors': postprocessors,

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError:

youtube_dl/options.py

@@ -183,6 +183,11 @@ def parseOpts(overrideArguments=None):
         action='store_const', dest='extract_flat', const='in_playlist',
         default=False,
         help='Do not extract the videos of a playlist, only list them.')
+    general.add_option(
+        '--concat-playlist',
+        action='store_true', dest='concat_playlist', default=False,
+        help='Concatenate all videos in a playlist into a single video file. Useful for services which split up full '
+             'episodes into multiple segments.')
     general.add_option(
         '--mark-watched',
         action='store_true', dest='mark_watched', default=False,

youtube_dl/postprocessor/__init__.py

@@ -9,6 +9,7 @@ from .ffmpeg import (
     FFmpegFixupM3u8PP,
     FFmpegFixupM4aPP,
     FFmpegMergerPP,
+    FFmpegConcatPP,
     FFmpegMetadataPP,
     FFmpegVideoConvertorPP,
     FFmpegSubtitlesConvertorPP,
@@ -31,6 +32,7 @@ __all__ = [
     'FFmpegFixupM4aPP',
     'FFmpegFixupStretchedPP',
     'FFmpegMergerPP',
+    'FFmpegConcatPP',
     'FFmpegMetadataPP',
     'FFmpegPostProcessor',
     'FFmpegSubtitlesConvertorPP',

youtube_dl/postprocessor/embedthumbnail.py

@@ -84,7 +84,7 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
 
             self._downloader.to_screen('[ffmpeg] Adding thumbnail to "%s"' % filename)
 
-            self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options)
+            self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options, [])
 
             if not self._already_have_thumbnail:
                 os.remove(encodeFilename(thumbnail_filename))

youtube_dl/postprocessor/ffmpeg.py

@@ -5,7 +5,6 @@ import subprocess
 import time
 import re
 
-
 from .common import AudioConversionError, PostProcessor
 
 from ..compat import compat_open as open
@@ -24,7 +23,6 @@ from ..utils import (
     replace_extension,
 )
 
-
 EXT_TO_OUT_FORMATS = {
     'aac': 'adts',
     'flac': 'flac',
@@ -196,7 +194,7 @@ class FFmpegPostProcessor(PostProcessor):
                 return mobj.group(1)
         return None
 
-    def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
+    def run_ffmpeg_multiple_files(self, input_paths, out_path, opts, file_opts):
         self.check_version()
 
         oldest_mtime = min(
@@ -207,6 +205,7 @@ class FFmpegPostProcessor(PostProcessor):
         files_cmd = []
         for path in input_paths:
             files_cmd.extend([
+                *file_opts,
                 encodeArgument('-i'),
                 encodeFilename(self._ffmpeg_filename_argument(path), True)
             ])
@@ -231,8 +230,8 @@ class FFmpegPostProcessor(PostProcessor):
             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 run_ffmpeg(self, path, out_path, opts, file_opts):
+        self.run_ffmpeg_multiple_files([path], out_path, opts, file_opts)
 
     def _ffmpeg_filename_argument(self, fn):
         # Always use 'file:' because the filename may contain ':' (ffmpeg
@@ -270,7 +269,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
             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 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'
@@ -315,7 +315,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
                 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
+        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
@@ -353,14 +354,16 @@ class FFmpegVideoConvertorPP(FFmpegPostProcessor):
     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))
+            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._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
@@ -419,7 +422,7 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
 
         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)
+        self.run_ffmpeg_multiple_files(input_files, temp_filename, opts, [])
         os.remove(encodeFilename(filename))
         os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
@@ -502,7 +505,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
                 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)
+        self.run_ffmpeg_multiple_files(in_filenames, temp_filename, options, [])
         if chapters:
             os.remove(metadata_filename)
         os.remove(encodeFilename(filename))
@@ -516,7 +519,7 @@ class FFmpegMergerPP(FFmpegPostProcessor):
         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)
+        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
 
@@ -538,6 +541,29 @@ class FFmpegMergerPP(FFmpegPostProcessor):
         return True
 
 
+class FFmpegConcatPP(FFmpegPostProcessor):
+    def run(self, info):
+        filename = info['filepath']
+
+        list_filename = prepend_extension(filename, 'list')
+        temp_filename = prepend_extension(filename, 'temp')
+
+        with open(list_filename, 'wt') as f:
+            f.write('ffconcat version 1.0\n')
+            for file in info['__files_to_concat']:
+                f.write("file '%s'\n" % self._ffmpeg_filename_argument(file))
+
+        file_opts = ['-f', 'concat', '-safe', '0']
+        opts = ['-c', 'copy']
+        self._downloader.to_screen('[ffmpeg] Concatenating files into "%s"' % filename)
+        self.run_ffmpeg(list_filename, temp_filename, opts, file_opts)
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+        return info['__files_to_concat'], info
+
+    def can_merge(self):
+        return True
+
+
 class FFmpegFixupStretchedPP(FFmpegPostProcessor):
     def run(self, info):
         stretched_ratio = info.get('stretched_ratio')
@@ -583,7 +609,7 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
 
             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)
+            self.run_ffmpeg(filename, temp_filename, options, [])
 
             os.remove(encodeFilename(filename))
             os.rename(encodeFilename(temp_filename), encodeFilename(filename))