Raphaël G. Git Repositories - youtubedl/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         # missing description
  26         'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
  27         'info_dict': {
  28             'id': 'BA-pQFBG8HZ',
  29             'ext': 'mp4',
  30             'uploader_id': 'britneyspears',
  31             'title': 'Video by britneyspears',
  32         },
  33         'params': {
  34             'skip_download': True,
  35         },
  36     }, {
  37         'url': 'https://instagram.com/p/-Cmh1cukG2/',
  38         'only_matching': True,
  39     }]
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43
  44         webpage = self._download_webpage(url, video_id)
  45         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
  46                                          webpage, 'uploader id', fatal=False)
  47         desc = self._search_regex(
  48             r'"caption":"(.+?)"', webpage, 'description', default=None)
  49
  50         return {
  51             'id': video_id,
  52             'url': self._og_search_video_url(webpage, secure=False),
  53             'ext': 'mp4',
  54             'title': 'Video by %s' % uploader_id,
  55             'thumbnail': self._og_search_thumbnail(webpage),
  56             'uploader_id': uploader_id,
  57             'description': desc,
  58         }
  59
  60
  61 class InstagramUserIE(InfoExtractor):
  62     _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
  63     IE_DESC = 'Instagram user profile'
  64     IE_NAME = 'instagram:user'
  65     _TEST = {
  66         'url': 'https://instagram.com/porsche',
  67         'info_dict': {
  68             'id': 'porsche',
  69             'title': 'porsche',
  70         },
  71         'playlist_mincount': 2,
  72         'playlist': [{
  73             'info_dict': {
  74                 'id': '614605558512799803_462752227',
  75                 'ext': 'mp4',
  76                 'title': '#Porsche Intelligent Performance.',
  77                 'thumbnail': 're:^https?://.*\.jpg',
  78                 'uploader': 'Porsche',
  79                 'uploader_id': 'porsche',
  80                 'timestamp': 1387486713,
  81                 'upload_date': '20131219',
  82             },
  83         }],
  84         'params': {
  85             'extract_flat': True,
  86             'skip_download': True,
  87         }
  88     }
  89
  90     def _real_extract(self, url):
  91         mobj = re.match(self._VALID_URL, url)
  92         uploader_id = mobj.group('username')
  93
  94         entries = []
  95         page_count = 0
  96         media_url = 'http://instagram.com/%s/media' % uploader_id
  97         while True:
  98             page = self._download_json(
  99                 media_url, uploader_id,
 100                 note='Downloading page %d ' % (page_count + 1),
 101             )
 102             page_count += 1
 103
 104             for it in page['items']:
 105                 if it.get('type') != 'video':
 106                     continue
 107                 like_count = int_or_none(it.get('likes', {}).get('count'))
 108                 user = it.get('user', {})
 109
 110                 formats = [{
 111                     'format_id': k,
 112                     'height': v.get('height'),
 113                     'width': v.get('width'),
 114                     'url': v['url'],
 115                 } for k, v in it['videos'].items()]
 116                 self._sort_formats(formats)
 117
 118                 thumbnails_el = it.get('images', {})
 119                 thumbnail = thumbnails_el.get('thumbnail', {}).get('url')
 120
 121                 # In some cases caption is null, which corresponds to None
 122                 # in python. As a result, it.get('caption', {}) gives None
 123                 title = (it.get('caption') or {}).get('text', it['id'])
 124
 125                 entries.append({
 126                     'id': it['id'],
 127                     'title': limit_length(title, 80),
 128                     'formats': formats,
 129                     'thumbnail': thumbnail,
 130                     'webpage_url': it.get('link'),
 131                     'uploader': user.get('full_name'),
 132                     'uploader_id': user.get('username'),
 133                     'like_count': like_count,
 134                     'timestamp': int_or_none(it.get('created_time')),
 135                 })
 136
 137             if not page['items']:
 138                 break
 139             max_id = page['items'][-1]['id']
 140             media_url = (
 141                 'http://instagram.com/%s/media?max_id=%s' % (
 142                     uploader_id, max_id))
 143
 144         return {
 145             '_type': 'playlist',
 146             'entries': entries,
 147             'id': uploader_id,
 148             'title': uploader_id,
 149         }