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