[franceculture] Make thumbnail optional (closes #18807)

Co-authored-by: Sergey M. <dstftw@gmail.com>

README.md

@@ -583,7 +583,7 @@ Available for the media that is a track or a part of a music album:
  - `disc_number` (numeric): Number of the disc or other physical medium the track belongs to
  - `release_year` (numeric): Year (YYYY) when the album was released
 
-Each aforementioned sequence when referenced in an output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by a particular extractor. Such sequences will be replaced with `NA`.
+Each aforementioned sequence when referenced in an output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by a particular extractor. Such sequences will be replaced with placeholder value provided with `--output-na-placeholder` (`NA` by default).
 
 For example for `-o %(title)s-%(id)s.%(ext)s` and an mp4 video with title `youtube-dl test video` and id `BaW_jenozKcj`, this will result in a `youtube-dl test video-BaW_jenozKcj.mp4` file created in the current directory.
 

test/test_YoutubeDL.py

@@ -633,13 +633,20 @@ class TestYoutubeDL(unittest.TestCase):
             'title2': '%PATH%',
         }
 
-        def fname(templ):
-            ydl = YoutubeDL({'outtmpl': templ})
+        def fname(templ, na_placeholder='NA'):
+            params = {'outtmpl': templ}
+            if na_placeholder != 'NA':
+                params['outtmpl_na_placeholder'] = na_placeholder
+            ydl = YoutubeDL(params)
             return ydl.prepare_filename(info)
         self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
         self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
-        # Replace missing fields with 'NA'
-        self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
+        NA_TEST_OUTTMPL = '%(uploader_date)s-%(width)d-%(id)s.%(ext)s'
+        # Replace missing fields with 'NA' by default
+        self.assertEqual(fname(NA_TEST_OUTTMPL), 'NA-NA-1234.mp4')
+        # Or by provided placeholder
+        self.assertEqual(fname(NA_TEST_OUTTMPL, na_placeholder='none'), 'none-none-1234.mp4')
+        self.assertEqual(fname(NA_TEST_OUTTMPL, na_placeholder=''), '--1234.mp4')
         self.assertEqual(fname('%(height)d.%(ext)s'), '1080.mp4')
         self.assertEqual(fname('%(height)6d.%(ext)s'), '  1080.mp4')
         self.assertEqual(fname('%(height)-6d.%(ext)s'), '1080  .mp4')

youtube_dl/YoutubeDL.py

@@ -163,6 +163,7 @@ class YoutubeDL(object):
     simulate:          Do not download the video files.
     format:            Video format code. See options.py for more information.
     outtmpl:           Template for output names.
+    outtmpl_na_placeholder: Placeholder for unavailable meta fields.
     restrictfilenames: Do not allow "&" and spaces in file names
     ignoreerrors:      Do not stop on download errors.
     force_generic_extractor: Force downloader to use the generic extractor
@@ -658,7 +659,7 @@ class YoutubeDL(object):
             template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
                                  for k, v in template_dict.items()
                                  if v is not None and not isinstance(v, (list, tuple, dict)))
-            template_dict = collections.defaultdict(lambda: 'NA', template_dict)
+            template_dict = collections.defaultdict(lambda: self.params.get('outtmpl_na_placeholder', 'NA'), template_dict)
 
             outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
 
@@ -678,8 +679,8 @@ class YoutubeDL(object):
 
             # Missing numeric fields used together with integer presentation types
             # in format specification will break the argument substitution since
-            # string 'NA' is returned for missing fields. We will patch output
-            # template for missing fields to meet string presentation type.
+            # string NA placeholder is returned for missing fields. We will patch
+            # output template for missing fields to meet string presentation type.
             for numeric_field in self._NUMERIC_FIELDS:
                 if numeric_field not in template_dict:
                     # As of [1] format syntax is:

youtube_dl/__init__.py

@@ -340,6 +340,7 @@ def _real_main(argv=None):
         'format': opts.format,
         'listformats': opts.listformats,
         'outtmpl': outtmpl,
+        'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
         'autonumber_size': opts.autonumber_size,
         'autonumber_start': opts.autonumber_start,
         'restrictfilenames': opts.restrictfilenames,

youtube_dl/extractor/franceculture.py

@@ -11,7 +11,7 @@ from ..utils import (
 
 class FranceCultureIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emissions/(?:[^/]+/)*(?P<id>[^/?#&]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.franceculture.fr/emissions/carnet-nomade/rendez-vous-au-pays-des-geeks',
         'info_dict': {
             'id': 'rendez-vous-au-pays-des-geeks',
@@ -20,10 +20,14 @@ class FranceCultureIE(InfoExtractor):
             'title': 'Rendez-vous au pays des geeks',
             'thumbnail': r're:^https?://.*\.jpg$',
             'upload_date': '20140301',
-            'timestamp': 1393642916,
+            'timestamp': 1393700400,
             'vcodec': 'none',
         }
-    }
+    }, {
+        # no thumbnail
+        'url': 'https://www.franceculture.fr/emissions/la-recherche-montre-en-main/la-recherche-montre-en-main-du-mercredi-10-octobre-2018',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
@@ -36,19 +40,19 @@ class FranceCultureIE(InfoExtractor):
                     </h1>|
                     <div[^>]+class="[^"]*?(?:title-zone-diffusion|heading-zone-(?:wrapper|player-button))[^"]*?"[^>]*>
                 ).*?
-                (<button[^>]+data-asset-source="[^"]+"[^>]+>)
+                (<button[^>]+data-(?:url|asset-source)="[^"]+"[^>]+>)
             ''',
             webpage, 'video data'))
 
-        video_url = video_data['data-asset-source']
-        title = video_data.get('data-asset-title') or self._og_search_title(webpage)
+        video_url = video_data.get('data-url') or video_data['data-asset-source']
+        title = video_data.get('data-asset-title') or video_data.get('data-diffusion-title') or self._og_search_title(webpage)
 
         description = self._html_search_regex(
             r'(?s)<div[^>]+class="intro"[^>]*>.*?<h2>(.+?)</h2>',
             webpage, 'description', default=None)
         thumbnail = self._search_regex(
             r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+(?:data-dejavu-)?src="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
+            webpage, 'thumbnail', default=None)
         uploader = self._html_search_regex(
             r'(?s)<span class="author">(.*?)</span>',
             webpage, 'uploader', default=None)
@@ -64,6 +68,6 @@ class FranceCultureIE(InfoExtractor):
             'ext': ext,
             'vcodec': 'none' if ext == 'mp3' else None,
             'uploader': uploader,
-            'timestamp': int_or_none(video_data.get('data-asset-created-date')),
+            'timestamp': int_or_none(video_data.get('data-start-time')) or int_or_none(video_data.get('data-asset-created-date')),
             'duration': int_or_none(video_data.get('data-duration')),
         }

youtube_dl/options.py

@@ -689,6 +689,10 @@ def parseOpts(overrideArguments=None):
         '-o', '--output',
         dest='outtmpl', metavar='TEMPLATE',
         help=('Output filename template, see the "OUTPUT TEMPLATE" for all the info'))
+    filesystem.add_option(
+        '--output-na-placeholder',
+        dest='outtmpl_na_placeholder', metavar='PLACEHOLDER', default='NA',
+        help=('Placeholder value for unavailable meta fields in output filename template (default is "%default")'))
     filesystem.add_option(
         '--autonumber-size',
         dest='autonumber_size', metavar='NUMBER', type=int,
@@ -782,7 +786,7 @@ def parseOpts(overrideArguments=None):
     postproc.add_option(
         '-x', '--extract-audio',
         action='store_true', dest='extractaudio', default=False,
-        help='Convert video files to audio-only files (requires ffmpeg or avconv and ffprobe or avprobe)')
+        help='Convert video files to audio-only files (requires ffmpeg/avconv and ffprobe/avprobe)')
     postproc.add_option(
         '--audio-format', metavar='FORMAT', dest='audioformat', default='best',
         help='Specify audio format: "best", "aac", "flac", "mp3", "m4a", "opus", "vorbis", or "wav"; "%default" by default; No effect without -x')