_ Git - youtube-dl/blob - youtube_dl/extractor/instagram.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import int_or_none
   7
   8
   9 class InstagramIE(InfoExtractor):
  10     _VALID_URL = r'https://instagram\.com/p/(?P<id>[\da-zA-Z]+)'
  11     _TEST = {
  12         'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
  13         'md5': '0d2da106a9d2631273e192b372806516',
  14         'info_dict': {
  15             'id': 'aye83DjauH',
  16             'ext': 'mp4',
  17             'uploader_id': 'naomipq',
  18             'title': 'Video by naomipq',
  19             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
  20         }
  21     }
  22
  23     def _real_extract(self, url):
  24         video_id = self._match_id(url)
  25
  26         webpage = self._download_webpage(url, video_id)
  27         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
  28                                          webpage, 'uploader id', fatal=False)
  29         desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
  30                                   fatal=False)
  31
  32         return {
  33             'id': video_id,
  34             'url': self._og_search_video_url(webpage, secure=False),
  35             'ext': 'mp4',
  36             'title': 'Video by %s' % uploader_id,
  37             'thumbnail': self._og_search_thumbnail(webpage),
  38             'uploader_id': uploader_id,
  39             'description': desc,
  40         }
  41
  42
  43 class InstagramUserIE(InfoExtractor):
  44     _VALID_URL = r'https://instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
  45     IE_DESC = 'Instagram user profile'
  46     IE_NAME = 'instagram:user'
  47     _TEST = {
  48         'url': 'https://instagram.com/porsche',
  49         'info_dict': {
  50             'id': 'porsche',
  51             'title': 'porsche',
  52         },
  53         'playlist_mincount': 2,
  54         'playlist': [{
  55             'info_dict': {
  56                 'id': '614605558512799803_462752227',
  57                 'ext': 'mp4',
  58                 'title': '#Porsche Intelligent Performance.',
  59                 'thumbnail': 're:^https?://.*\.jpg',
  60                 'uploader': 'Porsche',
  61                 'uploader_id': 'porsche',
  62                 'timestamp': 1387486713,
  63                 'upload_date': '20131219',
  64             },
  65         }],
  66         'params': {
  67             'extract_flat': True,
  68             'skip_download': True,
  69         }
  70     }
  71
  72     def _real_extract(self, url):
  73         mobj = re.match(self._VALID_URL, url)
  74         uploader_id = mobj.group('username')
  75
  76         entries = []
  77         page_count = 0
  78         media_url = 'http://instagram.com/%s/media' % uploader_id
  79         while True:
  80             page = self._download_json(
  81                 media_url, uploader_id,
  82                 note='Downloading page %d ' % (page_count + 1),
  83             )
  84             page_count += 1
  85
  86             for it in page['items']:
  87                 if it.get('type') != 'video':
  88                     continue
  89                 like_count = int_or_none(it.get('likes', {}).get('count'))
  90                 user = it.get('user', {})
  91
  92                 formats = [{
  93                     'format_id': k,
  94                     'height': v.get('height'),
  95                     'width': v.get('width'),
  96                     'url': v['url'],
  97                 } for k, v in it['videos'].items()]
  98                 self._sort_formats(formats)
  99
 100                 thumbnails_el = it.get('images', {})
 101                 thumbnail = thumbnails_el.get('thumbnail', {}).get('url')
 102
 103                 title = it.get('caption', {}).get('text', it['id'])
 104
 105                 entries.append({
 106                     'id': it['id'],
 107                     'title': title,
 108                     'formats': formats,
 109                     'thumbnail': thumbnail,
 110                     'webpage_url': it.get('link'),
 111                     'uploader': user.get('full_name'),
 112                     'uploader_id': user.get('username'),
 113                     'like_count': like_count,
 114                     'timestamp': int_or_none(it.get('created_time')),
 115                 })
 116
 117             if not page['items']:
 118                 break
 119             max_id = page['items'][-1]['id']
 120             media_url = (
 121                 'http://instagram.com/%s/media?max_id=%s' % (
 122                     uploader_id, max_id))
 123
 124         return {
 125             '_type': 'playlist',
 126             'entries': entries,
 127             'id': uploader_id,
 128             'title': uploader_id,
 129         }