_ 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 (
   7     int_or_none,
   8     limit_length,
   9 )
  10
  11
  12 class InstagramIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)'
  14     _TESTS = [{
  15         'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
  16         'md5': '0d2da106a9d2631273e192b372806516',
  17         'info_dict': {
  18             'id': 'aye83DjauH',
  19             'ext': 'mp4',
  20             'uploader_id': 'naomipq',
  21             'title': 'Video by naomipq',
  22             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
  23         }
  24     }, {
  25         'url': 'https://instagram.com/p/-Cmh1cukG2/',
  26         'only_matching': True,
  27     }]
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31
  32         webpage = self._download_webpage(url, video_id)
  33         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
  34                                          webpage, 'uploader id', fatal=False)
  35         desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
  36                                   fatal=False)
  37
  38         return {
  39             'id': video_id,
  40             'url': self._og_search_video_url(webpage, secure=False),
  41             'ext': 'mp4',
  42             'title': 'Video by %s' % uploader_id,
  43             'thumbnail': self._og_search_thumbnail(webpage),
  44             'uploader_id': uploader_id,
  45             'description': desc,
  46         }
  47
  48
  49 class InstagramUserIE(InfoExtractor):
  50     _VALID_URL = r'https://instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
  51     IE_DESC = 'Instagram user profile'
  52     IE_NAME = 'instagram:user'
  53     _TEST = {
  54         'url': 'https://instagram.com/porsche',
  55         'info_dict': {
  56             'id': 'porsche',
  57             'title': 'porsche',
  58         },
  59         'playlist_mincount': 2,
  60         'playlist': [{
  61             'info_dict': {
  62                 'id': '614605558512799803_462752227',
  63                 'ext': 'mp4',
  64                 'title': '#Porsche Intelligent Performance.',
  65                 'thumbnail': 're:^https?://.*\.jpg',
  66                 'uploader': 'Porsche',
  67                 'uploader_id': 'porsche',
  68                 'timestamp': 1387486713,
  69                 'upload_date': '20131219',
  70             },
  71         }],
  72         'params': {
  73             'extract_flat': True,
  74             'skip_download': True,
  75         }
  76     }
  77
  78     def _real_extract(self, url):
  79         mobj = re.match(self._VALID_URL, url)
  80         uploader_id = mobj.group('username')
  81
  82         entries = []
  83         page_count = 0
  84         media_url = 'http://instagram.com/%s/media' % uploader_id
  85         while True:
  86             page = self._download_json(
  87                 media_url, uploader_id,
  88                 note='Downloading page %d ' % (page_count + 1),
  89             )
  90             page_count += 1
  91
  92             for it in page['items']:
  93                 if it.get('type') != 'video':
  94                     continue
  95                 like_count = int_or_none(it.get('likes', {}).get('count'))
  96                 user = it.get('user', {})
  97
  98                 formats = [{
  99                     'format_id': k,
 100                     'height': v.get('height'),
 101                     'width': v.get('width'),
 102                     'url': v['url'],
 103                 } for k, v in it['videos'].items()]
 104                 self._sort_formats(formats)
 105
 106                 thumbnails_el = it.get('images', {})
 107                 thumbnail = thumbnails_el.get('thumbnail', {}).get('url')
 108
 109                 # In some cases caption is null, which corresponds to None
 110                 # in python. As a result, it.get('caption', {}) gives None
 111                 title = (it.get('caption') or {}).get('text', it['id'])
 112
 113                 entries.append({
 114                     'id': it['id'],
 115                     'title': limit_length(title, 80),
 116                     'formats': formats,
 117                     'thumbnail': thumbnail,
 118                     'webpage_url': it.get('link'),
 119                     'uploader': user.get('full_name'),
 120                     'uploader_id': user.get('username'),
 121                     'like_count': like_count,
 122                     'timestamp': int_or_none(it.get('created_time')),
 123                 })
 124
 125             if not page['items']:
 126                 break
 127             max_id = page['items'][-1]['id']
 128             media_url = (
 129                 'http://instagram.com/%s/media?max_id=%s' % (
 130                     uploader_id, max_id))
 131
 132         return {
 133             '_type': 'playlist',
 134             'entries': entries,
 135             'id': uploader_id,
 136             'title': uploader_id,
 137         }