]>
Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/npr.py
1 from __future__
import unicode_literals
3 from .common
import InfoExtractor
10 class NprIE(InfoExtractor
):
11 _VALID_URL
= r
'https?://(?:www\.)?npr\.org/(?:sections/[^/]+/)?\d{4}/\d{2}/\d{2}/(?P<id>\d+)'
13 'url': 'https://www.npr.org/sections/allsongs/2015/10/21/449974205/new-music-from-beach-house-chairlift-cmj-discoveries-and-more',
16 'title': 'New Music From Beach House, Chairlift, CMJ Discoveries And More'
20 'url': 'https://www.npr.org/sections/deceptivecadence/2015/10/09/446928052/music-from-the-shadows-ancient-armenian-hymns-and-piano-jazz',
23 'title': "Songs We Love: Tigran Hamasyan, 'Your Mercy is Boundless'"
26 'md5': '12fa60cb2d3ed932f53609d4aeceabf1',
30 'title': 'Your Mercy is Boundless (Bazum en Qo gtutyunqd)',
35 # mutlimedia, not media title
36 'url': 'https://www.npr.org/2017/06/19/533198237/tigers-jaw-tiny-desk-concert',
39 'title': 'Tigers Jaw: Tiny Desk Concert',
42 'md5': '12fa60cb2d3ed932f53609d4aeceabf1',
46 'title': 'Tigers Jaw: Tiny Desk Concert',
50 'expected_warnings': ['Failed to download m3u8 information'],
53 def _real_extract(self
, url
):
54 playlist_id
= self
._match
_id
(url
)
56 story
= self
._download
_json
(
57 'http://api.npr.org/query', playlist_id
, query
={
59 'fields': 'audio,multimedia,title',
61 'apiKey': 'MDAzMzQ2MjAyMDEyMzk4MTU1MDg3ZmM3MQ010',
62 })['list']['story'][0]
63 playlist_title
= story
.get('title', {}).get('$text')
65 KNOWN_FORMATS
= ('threegp', 'm3u8', 'smil', 'mp4', 'mp3')
66 quality
= qualities(KNOWN_FORMATS
)
69 for media
in story
.get('audio', []) + story
.get('multimedia', []):
70 media_id
= media
['id']
73 for format_id
, formats_entry
in media
.get('format', {}).items():
76 if isinstance(formats_entry
, list):
77 formats_entry
= formats_entry
[0]
78 format_url
= formats_entry
.get('$text')
81 if format_id
in KNOWN_FORMATS
:
82 if format_id
== 'm3u8':
83 formats
.extend(self
._extract
_m
3u8_formats
(
84 format_url
, media_id
, 'mp4', 'm3u8_native',
85 m3u8_id
='hls', fatal
=False))
86 elif format_id
== 'smil':
87 smil_formats
= self
._extract
_smil
_formats
(
88 format_url
, media_id
, transform_source
=lambda s
: s
.replace(
89 'rtmp://flash.npr.org/ondemand/', 'https://ondemand.npr.org/'))
90 self
._check
_formats
(smil_formats
, media_id
)
91 formats
.extend(smil_formats
)
95 'format_id': format_id
,
96 'quality': quality(format_id
),
98 self
._sort
_formats
(formats
)
102 'title': media
.get('title', {}).get('$text') or playlist_title
,
103 'thumbnail': media
.get('altImageUrl', {}).get('$text'),
104 'duration': int_or_none(media
.get('duration', {}).get('$text')),
108 return self
.playlist_result(entries
, playlist_id
, playlist_title
)