From ba17598d061bf316c449052fcabbc55e6010c83f Mon Sep 17 00:00:00 2001
From: luxcem <a@luxcem.fr>
Date: Mon, 15 Jul 2019 20:15:25 +0200
Subject: [PATCH] [arteradio] Add new extractor

---
 youtube_dl/extractor/arteradio.py  | 135 +++++++++++++++++++++++++++++
 youtube_dl/extractor/extractors.py |   4 +
 2 files changed, 139 insertions(+)
 create mode 100644 youtube_dl/extractor/arteradio.py

diff --git a/youtube_dl/extractor/arteradio.py b/youtube_dl/extractor/arteradio.py
new file mode 100644
index 000000000..76fde70bc
--- /dev/null
+++ b/youtube_dl/extractor/arteradio.py
@@ -0,0 +1,135 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    month_by_name,
+    int_or_none,
+    str_or_none,
+)
+
+
+class ArteRadioIE(InfoExtractor):
+    """ArteRadio sound extractor."""
+    IE_NAME = 'arteradio'
+    _VALID_URL = r'https?://(?:www\.)?arteradio\.com/son/(?P<id>\d+)/(.*)'
+    _CDN_URL = 'https://download.www.arte.tv/permanent/arteradio/sites/default/files/sons/'
+
+    _TESTS = [{
+        'url': 'https://www.arteradio.com/son/616458/la_bas_si_j_y_suis_plus',
+        'md5': 'ae9219bbcfbb258ab5d5ba877708e2a9',
+        'info_dict': {
+            'id': '616458',
+            'ext': 'mp3',
+            'title': 'Là-bas si j\'y suis plus | ARTE Radio',
+            'upload_date': '20140911',
+            'description': 'md5:863c01af898a02681a2f543d32031566',
+            'vcodec': 'none',
+            'duration': 917,
+        },
+    }, {
+        'url': 'https://www.arteradio.com/son/61661758/bande_annonce_beatmakers_saison_2',
+        'md5': 'a4678484b374e35faf47a555860a0a4f',
+        'info_dict': {
+            'id': '61661758',
+            'ext': 'mp3',
+            'title': 'Bande-annonce Beatmakers, saison 2 | ARTE Radio',
+            'upload_date': '20190627',
+            'description': 'md5:d52a0fd2fcc88e2d4b1cd0f2a5092fd1',
+            'vcodec': 'none',
+            'duration': 56,
+            'thumbnail': 'https://www.arteradio.com/sites/default/files/beatmakers_s2_1.jpg'
+        },
+    }]
+
+    def _extract_date(self, webpage):
+        # Fetching date
+        upload_date_str = self._html_search_regex(
+            r'<h5[^>]*>Mise en ligne.+<\/h5>\s+<p>(.+)<\/p>',
+            webpage, 'upload_date_str', fatal=False, default=None)
+        if not upload_date_str:
+            return None
+        try:
+            day, month, year = upload_date_str.split(' ')
+            day = '{:02d}'.format(int_or_none(day))
+            month = '{:02d}'.format(month_by_name(month, lang='fr'))
+        except (ValueError, TypeError):
+            return None
+        return ''.join((year, month, day))
+
+    def _extract_data_from_button(self, button):
+        meta_data = dict(re.findall(r'data-([-\w]+?)=\"(.+?)\"', button))
+        # If no sound-href is found we cannot extract the link
+        try:
+            url = self._CDN_URL + meta_data['sound-href']
+        except KeyError:
+            raise ExtractorError('No audio found')
+        return {
+            'id': meta_data.get('sound-id'),
+            'duration': int_or_none(meta_data.get('duration-seconds')),
+            'url': url,
+            'href': meta_data.get('href'),
+            'position': meta_data.get('position-serie'),
+            'thumbnail': meta_data.get('image-url'),
+            'vcodec': 'none',
+        }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        button = self._html_search_regex(
+            r'<button([^>]+data-sound-id=\"{}\"[^>]*)>'.format(video_id),
+            webpage, 'button')
+        audio_data = self._extract_data_from_button(button)
+        result = {
+            'id': audio_data['id'] or video_id,
+            'duration': audio_data['duration'],
+            'url': audio_data['url'],
+            'thumbnail': audio_data['thumbnail'],
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'upload_date': str_or_none(self._extract_date(webpage)),
+        }
+        result.update(audio_data)
+        return result
+
+
+class ArteRadioSerieIE(ArteRadioIE):
+    """ArteRadio serie extractor."""
+    IE_NAME = 'arteradio.com:playlist'
+    _VALID_URL = r'https?://(?:www\.)?arteradio\.com/serie/(?P<id>.+)'
+
+    _TESTS = [{
+        'url': 'https://www.arteradio.com/serie/crackopolis',
+        'md5': '',
+        'info_dict': {
+            'id': 'crackopolis',
+            'title': 'CRACKOPOLIS | ARTE Radio',
+            'description': 'md5:1b4665891ef07bef17c98d692435d177',
+        },
+        'playlist_count': 16
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        webpage = self._download_webpage(url, playlist_id)
+        button_entries = re.findall(
+            r'<button([^>]+data-serie-url=\"/serie/{}\"[^>]*)>'.format(playlist_id),
+            webpage)
+        entries = [self._extract_data_from_button(button) for button in button_entries]
+        # The title for an item is not in the item meta data
+        # We generate the title from the item url
+        for index, entry in enumerate(entries):
+            position = entry.get('position') or index
+            try:
+                title = entry.get('href', '').split('/')[3]
+            except IndexError:
+                title = playlist_id
+            entry['title'] = position + '_' + title
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        return self.playlist_result(entries, playlist_id, title, description)
diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
index a202a0449..8fdef738a 100644
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -62,6 +62,10 @@ from .arte import (
     ArteTVEmbedIE,
     ArteTVPlaylistIE,
 )
+from .arteradio import (
+    ArteRadioIE,
+    ArteRadioSerieIE,
+)
 from .asiancrush import (
     AsianCrushIE,
     AsianCrushPlaylistIE,