youtube_dl/extractor/pinkbike.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     remove_end,
  10     remove_start,
  11     str_to_int,
  12     unified_strdate,
  13 )
  14
  15
  16 class PinkbikeIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:(?:www\.)?pinkbike\.com/video/|es\.pinkbike\.org/i/kvid/kvid-y5\.swf\?id=)(?P<id>[0-9]+)'
  18     _TESTS = [{
  19         'url': 'http://www.pinkbike.com/video/402811/',
  20         'md5': '4814b8ca7651034cd87e3361d5c2155a',
  21         'info_dict': {
  22             'id': '402811',
  23             'ext': 'mp4',
  24             'title': 'Brandon Semenuk - RAW 100',
  25             'description': 'Official release: www.redbull.ca/rupertwalker',
  26             'thumbnail': r're:^https?://.*\.jpg$',
  27             'duration': 100,
  28             'upload_date': '20150406',
  29             'uploader': 'revelco',
  30             'location': 'Victoria, British Columbia, Canada',
  31             'view_count': int,
  32             'comment_count': int,
  33         }
  34     }, {
  35         'url': 'http://es.pinkbike.org/i/kvid/kvid-y5.swf?id=406629',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41
  42         webpage = self._download_webpage(
  43             'http://www.pinkbike.com/video/%s' % video_id, video_id)
  44
  45         formats = []
  46         for _, format_id, src in re.findall(
  47                 r'data-quality=((?:\\)?["\'])(.+?)\1[^>]+src=\1(.+?)\1', webpage):
  48             height = int_or_none(self._search_regex(
  49                 r'^(\d+)[pP]$', format_id, 'height', default=None))
  50             formats.append({
  51                 'url': src,
  52                 'format_id': format_id,
  53                 'height': height,
  54             })
  55         self._sort_formats(formats)
  56
  57         title = remove_end(self._og_search_title(webpage), ' Video - Pinkbike')
  58         description = self._html_search_regex(
  59             r'(?s)id="media-description"[^>]*>(.+?)<',
  60             webpage, 'description', default=None) or remove_start(
  61             self._og_search_description(webpage), title + '. ')
  62         thumbnail = self._og_search_thumbnail(webpage)
  63         duration = int_or_none(self._html_search_meta(
  64             'video:duration', webpage, 'duration'))
  65
  66         uploader = self._search_regex(
  67             r'<a[^>]+\brel=["\']author[^>]+>([^<]+)', webpage,
  68             'uploader', fatal=False)
  69         upload_date = unified_strdate(self._search_regex(
  70             r'class="fullTime"[^>]+title="([^"]+)"',
  71             webpage, 'upload date', fatal=False))
  72
  73         location = self._html_search_regex(
  74             r'(?s)<dt>Location</dt>\s*<dd>(.+?)<',
  75             webpage, 'location', fatal=False)
  76
  77         def extract_count(webpage, label):
  78             return str_to_int(self._search_regex(
  79                 r'<span[^>]+class="stat-num"[^>]*>([\d,.]+)</span>\s*<span[^>]+class="stat-label"[^>]*>%s' % label,
  80                 webpage, label, fatal=False))
  81
  82         view_count = extract_count(webpage, 'Views')
  83         comment_count = extract_count(webpage, 'Comments')
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'description': description,
  89             'thumbnail': thumbnail,
  90             'duration': duration,
  91             'upload_date': upload_date,
  92             'uploader': uploader,
  93             'location': location,
  94             'view_count': view_count,
  95             'comment_count': comment_count,
  96             'formats': formats
  97         }