youtube_dl/extractor/newgrounds.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     extract_attributes,
   8     int_or_none,
   9     parse_duration,
  10     parse_filesize,
  11     unified_timestamp,
  12 )
  13
  14
  15 class NewgroundsIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)'
  17     _TESTS = [{
  18         'url': 'https://www.newgrounds.com/audio/listen/549479',
  19         'md5': 'fe6033d297591288fa1c1f780386f07a',
  20         'info_dict': {
  21             'id': '549479',
  22             'ext': 'mp3',
  23             'title': 'B7 - BusMode',
  24             'uploader': 'Burn7',
  25             'timestamp': 1378878540,
  26             'upload_date': '20130911',
  27             'duration': 143,
  28         },
  29     }, {
  30         'url': 'https://www.newgrounds.com/portal/view/673111',
  31         'md5': '3394735822aab2478c31b1004fe5e5bc',
  32         'info_dict': {
  33             'id': '673111',
  34             'ext': 'mp4',
  35             'title': 'Dancin',
  36             'uploader': 'Squirrelman82',
  37             'timestamp': 1460256780,
  38             'upload_date': '20160410',
  39         },
  40     }, {
  41         # source format unavailable, additional mp4 formats
  42         'url': 'http://www.newgrounds.com/portal/view/689400',
  43         'info_dict': {
  44             'id': '689400',
  45             'ext': 'mp4',
  46             'title': 'ZTV News Episode 8',
  47             'uploader': 'BennettTheSage',
  48             'timestamp': 1487965140,
  49             'upload_date': '20170224',
  50         },
  51         'params': {
  52             'skip_download': True,
  53         },
  54     }]
  55
  56     def _real_extract(self, url):
  57         media_id = self._match_id(url)
  58
  59         webpage = self._download_webpage(url, media_id)
  60
  61         title = self._html_search_regex(
  62             r'<title>([^>]+)</title>', webpage, 'title')
  63
  64         media_url = self._parse_json(self._search_regex(
  65             r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
  66
  67         formats = [{
  68             'url': media_url,
  69             'format_id': 'source',
  70             'quality': 1,
  71         }]
  72
  73         max_resolution = int_or_none(self._search_regex(
  74             r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
  75             default=None))
  76         if max_resolution:
  77             url_base = media_url.rpartition('.')[0]
  78             for resolution in (360, 720, 1080):
  79                 if resolution > max_resolution:
  80                     break
  81                 formats.append({
  82                     'url': '%s.%dp.mp4' % (url_base, resolution),
  83                     'format_id': '%dp' % resolution,
  84                     'height': resolution,
  85                 })
  86
  87         self._check_formats(formats, media_id)
  88         self._sort_formats(formats)
  89
  90         uploader = self._html_search_regex(
  91             (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
  92              r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
  93             fatal=False)
  94
  95         timestamp = unified_timestamp(self._html_search_regex(
  96             (r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+</dd>\s*<dd>[^<]+)',
  97              r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+)'), webpage, 'timestamp',
  98             default=None))
  99         duration = parse_duration(self._search_regex(
 100             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
 101             'duration', default=None))
 102
 103         filesize_approx = parse_filesize(self._html_search_regex(
 104             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>(.+?)</dd>', webpage, 'filesize',
 105             default=None))
 106         if len(formats) == 1:
 107             formats[0]['filesize_approx'] = filesize_approx
 108
 109         if '<dd>Song' in webpage:
 110             formats[0]['vcodec'] = 'none'
 111
 112         return {
 113             'id': media_id,
 114             'title': title,
 115             'uploader': uploader,
 116             'timestamp': timestamp,
 117             'duration': duration,
 118             'formats': formats,
 119         }
 120
 121
 122 class NewgroundsPlaylistIE(InfoExtractor):
 123     _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:collection|[^/]+/search/[^/]+)/(?P<id>[^/?#&]+)'
 124     _TESTS = [{
 125         'url': 'https://www.newgrounds.com/collection/cats',
 126         'info_dict': {
 127             'id': 'cats',
 128             'title': 'Cats',
 129         },
 130         'playlist_mincount': 46,
 131     }, {
 132         'url': 'http://www.newgrounds.com/portal/search/author/ZONE-SAMA',
 133         'info_dict': {
 134             'id': 'ZONE-SAMA',
 135             'title': 'Portal Search: ZONE-SAMA',
 136         },
 137         'playlist_mincount': 47,
 138     }, {
 139         'url': 'http://www.newgrounds.com/audio/search/title/cats',
 140         'only_matching': True,
 141     }]
 142
 143     def _real_extract(self, url):
 144         playlist_id = self._match_id(url)
 145
 146         webpage = self._download_webpage(url, playlist_id)
 147
 148         title = self._search_regex(
 149             r'<title>([^>]+)</title>', webpage, 'title', default=None)
 150
 151         # cut left menu
 152         webpage = self._search_regex(
 153             r'(?s)<div[^>]+\bclass=["\']column wide(.+)',
 154             webpage, 'wide column', default=webpage)
 155
 156         entries = []
 157         for a, path, media_id in re.findall(
 158                 r'(<a[^>]+\bhref=["\']/?((?:portal/view|audio/listen)/(\d+))[^>]+>)',
 159                 webpage):
 160             a_class = extract_attributes(a).get('class')
 161             if a_class not in ('item-portalsubmission', 'item-audiosubmission'):
 162                 continue
 163             entries.append(
 164                 self.url_result(
 165                     'https://www.newgrounds.com/%s' % path,
 166                     ie=NewgroundsIE.ie_key(), video_id=media_id))
 167
 168         return self.playlist_result(entries, playlist_id, title)