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 )
   9
  10
  11 class InstagramIE(InfoExtractor):
  12     _VALID_URL = r'http://instagram\.com/p/(?P<id>.*?)/'
  13     _TEST = {
  14         'url': 'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
  15         'md5': '0d2da106a9d2631273e192b372806516',
  16         'info_dict': {
  17             'id': 'aye83DjauH',
  18             'ext': 'mp4',
  19             'uploader_id': 'naomipq',
  20             'title': 'Video by naomipq',
  21             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         video_id = mobj.group('id')
  28         webpage = self._download_webpage(url, video_id)
  29         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
  30                                          webpage, 'uploader id', fatal=False)
  31         desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
  32                                   fatal=False)
  33
  34         return {
  35             'id': video_id,
  36             'url': self._og_search_video_url(webpage, secure=False),
  37             'ext': 'mp4',
  38             'title': 'Video by %s' % uploader_id,
  39             'thumbnail': self._og_search_thumbnail(webpage),
  40             'uploader_id': uploader_id,
  41             'description': desc,
  42         }
  43
  44
  45 class InstagramUserIE(InfoExtractor):
  46     _VALID_URL = r'http://instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
  47     IE_DESC = 'Instagram user profile'
  48     IE_NAME = 'instagram:user'
  49     _TEST = {
  50         'url': 'http://instagram.com/porsche',
  51         'info_dict': {
  52             'id': 'porsche',
  53             'title': 'porsche',
  54         },
  55         'playlist_mincount': 2,
  56         'playlist': [{
  57             'info_dict': {
  58                 'id': '614605558512799803_462752227',
  59                 'ext': 'mp4',
  60                 'title': '#Porsche Intelligent Performance.',
  61                 'thumbnail': 're:^https?://.*\.jpg',
  62                 'uploader': 'Porsche',
  63                 'uploader_id': 'porsche',
  64                 'timestamp': 1387486713,
  65                 'upload_date': '20131219',
  66             },
  67         }],
  68         'params': {
  69             'extract_flat': True,
  70             'skip_download': True,
  71         }
  72     }
  73
  74     def _real_extract(self, url):
  75         mobj = re.match(self._VALID_URL, url)
  76         uploader_id = mobj.group('username')
  77
  78         entries = []
  79         page_count = 0
  80         media_url = 'http://instagram.com/%s/media' % uploader_id
  81         while True:
  82             page = self._download_json(
  83                 media_url, uploader_id,
  84                 note='Downloading page %d ' % (page_count + 1),
  85             )
  86             page_count += 1
  87
  88             for it in page['items']:
  89                 if it.get('type') != 'video':
  90                     continue
  91                 like_count = int_or_none(it.get('likes', {}).get('count'))
  92                 user = it.get('user', {})
  93
  94                 formats = [{
  95                     'format_id': k,
  96                     'height': v.get('height'),
  97                     'width': v.get('width'),
  98                     'url': v['url'],
  99                 } for k, v in it['videos'].items()]
 100                 self._sort_formats(formats)
 101
 102                 thumbnails_el = it.get('images', {})
 103                 thumbnail = thumbnails_el.get('thumbnail', {}).get('url')
 104
 105                 title = it.get('caption', {}).get('text', it['id'])
 106
 107                 entries.append({
 108                     'id': it['id'],
 109                     'title': title,
 110                     'formats': formats,
 111                     'thumbnail': thumbnail,
 112                     'webpage_url': it.get('link'),
 113                     'uploader': user.get('full_name'),
 114                     'uploader_id': user.get('username'),
 115                     'like_count': like_count,
 116                     'timestamp': int_or_none(it.get('created_time')),
 117                 })
 118
 119             if not page['items']:
 120                 break
 121             max_id = page['items'][-1]['id']
 122             media_url = (
 123                 'http://instagram.com/%s/media?max_id=%s' % (
 124                     uploader_id, max_id))
 125
 126         return {
 127             '_type': 'playlist',
 128             'entries': entries,
 129             'id': uploader_id,
 130             'title': uploader_id,
 131         }