X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/5ea61d104de9ca8a7d19d63d83173eb7391081d2..4121346dfbb2b660ded98e98069857c7a50e381f:/youtube_dl/extractor/nextmedia.py diff --git a/youtube_dl/extractor/nextmedia.py b/youtube_dl/extractor/nextmedia.py index d1b7cff..7bd1290 100644 --- a/youtube_dl/extractor/nextmedia.py +++ b/youtube_dl/extractor/nextmedia.py @@ -2,11 +2,20 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import parse_iso8601 +from ..compat import compat_urlparse +from ..utils import ( + clean_html, + get_element_by_class, + int_or_none, + parse_iso8601, + remove_start, + unified_timestamp, +) class NextMediaIE(InfoExtractor): - _VALID_URL = r'http://hk.apple.nextmedia.com/[^/]+/[^/]+/(?P\d+)/(?P\d+)' + IE_DESC = '蘋果日報' + _VALID_URL = r'https?://hk\.apple\.nextmedia\.com/[^/]+/[^/]+/(?P\d+)/(?P\d+)' _TESTS = [{ 'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199', 'md5': 'dff9fad7009311c421176d1ac90bfe4f', @@ -14,7 +23,7 @@ class NextMediaIE(InfoExtractor): 'id': '53109199', 'ext': 'mp4', 'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'md5:28222b9912b6665a21011b034c70fcc7', 'timestamp': 1415456273, 'upload_date': '20141108', @@ -29,6 +38,12 @@ class NextMediaIE(InfoExtractor): return self._extract_from_nextmedia_page(news_id, url, page) def _extract_from_nextmedia_page(self, news_id, url, page): + redirection_url = self._search_regex( + r'window\.location\.href\s*=\s*([\'"])(?P(?!\1).+)\1', + page, 'redirection URL', default=None, group='url') + if redirection_url: + return self.url_result(compat_urlparse.urljoin(url, redirection_url)) + title = self._fetch_title(page) video_url = self._search_regex(self._URL_PATTERN, page, 'video url') @@ -66,7 +81,8 @@ class NextMediaIE(InfoExtractor): class NextMediaActionNewsIE(NextMediaIE): - _VALID_URL = r'http://hk.dv.nextmedia.com/actionnews/[^/]+/(?P\d+)/(?P\d+)/\d+' + IE_DESC = '蘋果日報 - 動新聞' + _VALID_URL = r'https?://hk\.dv\.nextmedia\.com/actionnews/[^/]+/(?P\d+)/(?P\d+)/\d+' _TESTS = [{ 'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460', 'md5': '05fce8ffeed7a5e00665d4b7cf0f9201', @@ -74,7 +90,7 @@ class NextMediaActionNewsIE(NextMediaIE): 'id': '19009428', 'ext': 'mp4', 'title': '【壹週刊】細10年男友偷食　50歲邵美琪再失戀', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659', 'timestamp': 1421791200, 'upload_date': '20150120', @@ -90,7 +106,8 @@ class NextMediaActionNewsIE(NextMediaIE): class AppleDailyIE(NextMediaIE): - _VALID_URL = r'http://(www|ent).appledaily.com.tw/(?:animation|appledaily|enews|realtimenews)/[^/]+/[^/]+/(?P\d+)/(?P\d+)(/.*)?' + IE_DESC = '臺灣蘋果日報' + _VALID_URL = r'https?://(www|ent)\.appledaily\.com\.tw/[^/]+/[^/]+/[^/]+/(?P\d+)/(?P\d+)(/.*)?' _TESTS = [{ 'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694', 'md5': 'a843ab23d150977cc55ef94f1e2c1e4d', @@ -98,7 +115,7 @@ class AppleDailyIE(NextMediaIE): 'id': '36354694', 'ext': 'mp4', 'title': '周亭羽走過摩鐵陰霾2男陪吃 九把刀孤寒看醫生', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'md5:2acd430e59956dc47cd7f67cb3c003f4', 'upload_date': '20150128', } @@ -109,7 +126,7 @@ class AppleDailyIE(NextMediaIE): 'id': '550549', 'ext': 'mp4', 'title': '不滿被踩腳　山東兩大媽一路打下車', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'md5:175b4260c1d7c085993474217e4ab1b4', 'upload_date': '20150128', } @@ -120,10 +137,11 @@ class AppleDailyIE(NextMediaIE): 'id': '5003671', 'ext': 'mp4', 'title': '20正妹熱舞　《刀龍傳說Online》火辣上市', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd', 'upload_date': '20150128', - } + }, + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { # No thumbnail 'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/', @@ -137,17 +155,33 @@ class AppleDailyIE(NextMediaIE): }, 'expected_warnings': [ 'video thumbnail', - ] + ], + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { 'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/', + 'md5': 'eaa20e6b9df418c912d7f5dec2ba734d', + 'info_dict': { + 'id': '35770334', + 'ext': 'mp4', + 'title': '咖啡占卜測 XU裝熟指數', + 'thumbnail': r're:^https?://.*\.jpg$', + 'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748', + 'upload_date': '20140417', + }, + }, { + 'url': 'http://www.appledaily.com.tw/actionnews/appledaily/7/20161003/960588/', + 'only_matching': True, + }, { + # Redirected from http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694 + 'url': 'http://ent.appledaily.com.tw/section/article/headline/20150128/36354694', 'only_matching': True, }] _URL_PATTERN = r'\{url: \'(.+)\'\}' def _fetch_title(self, page): - return (self._html_search_regex(r'

([^<>]+)

', page, 'news title', default=None) or - self._html_search_meta('description', page, 'news title')) + return (self._html_search_regex(r'

([^<>]+)

', page, 'news title', default=None) + or self._html_search_meta('description', page, 'news title')) def _fetch_thumbnail(self, page): return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False) @@ -157,3 +191,48 @@ class AppleDailyIE(NextMediaIE): def _fetch_description(self, page): return self._html_search_meta('description', page, 'news description') + + +class NextTVIE(InfoExtractor): + IE_DESC = '壹電視' + _VALID_URL = r'https?://(?:www\.)?nexttv\.com\.tw/(?:[^/]+/)+(?P\d+)' + + _TEST = { + 'url': 'http://www.nexttv.com.tw/news/realtime/politics/11779671', + 'info_dict': { + 'id': '11779671', + 'ext': 'mp4', + 'title': '「超收稅」近4千億！　藍議員籲發消費券', + 'thumbnail': r're:^https?://.*\.jpg$', + 'timestamp': 1484825400, + 'upload_date': '20170119', + 'view_count': int, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex( + r']*>([^<]+)', webpage, 'title') + + data = self._hidden_inputs(webpage) + + video_url = data['ntt-vod-src-detailview'] + + date_str = get_element_by_class('date', webpage) + timestamp = unified_timestamp(date_str + '+0800') if date_str else None + + view_count = int_or_none(remove_start( + clean_html(get_element_by_class('click', webpage)), '點閱：')) + + return { + 'id': video_id, + 'title': title, + 'url': video_url, + 'thumbnail': data.get('ntt-vod-img-src'), + 'timestamp': timestamp, + 'view_count': view_count, + }