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
  50     def _real_extract(self, url):
  51         mobj = re.match(self._VALID_URL, url)
  52         uploader_id = mobj.group('username')
  53
  54         entries = []
  55         page_count = 0
  56         media_url = 'http://instagram.com/%s/media' % uploader_id
  57         while True:
  58             page = self._download_json(
  59                 media_url, uploader_id,
  60                 note='Downloading page %d ' % (page_count + 1),
  61             )
  62             page_count += 1
  63
  64             for it in page['items']:
  65                 if it.get('type') != 'video':
  66                     continue
  67                 like_count = int_or_none(it.get('likes', {}).get('count'))
  68                 user = it.get('user', {})
  69
  70                 formats = [{
  71                     'format_id': k,
  72                     'height': v.get('height'),
  73                     'width': v.get('width'),
  74                     'url': v['url'],
  75                 } for k, v in it['videos'].items()]
  76                 self._sort_formats(formats)
  77
  78                 thumbnails_el = it.get('images', {})
  79                 thumbnail = thumbnails_el.get('thumbnail', {}).get('url')
  80
  81                 title = it.get('caption', {}).get('text', it['id'])
  82
  83                 entries.append({
  84                     'id': it['id'],
  85                     'title': title,
  86                     'formats': formats,
  87                     'thumbnail': thumbnail,
  88                     'webpage_url': it.get('link'),
  89                     'uploader': user.get('full_name'),
  90                     'uploader_id': user.get('username'),
  91                     'like_count': like_count,
  92                     'timestamp': int_or_none(it.get('created_time')),
  93                 })
  94
  95             if not page['items']:
  96                 break
  97             max_id = page['items'][-1]['id']
  98             media_url = (
  99                 'http://instagram.com/%s/media?max_id=%s' % (
 100                     uploader_id, max_id))
 101
 102         return {
 103             '_type': 'playlist',
 104             'entries': entries,
 105             'id': uploader_id,
 106             'title': uploader_id,
 107         }