diff options
Diffstat (limited to 'hypervideo_dl/extractor/gettr.py')
-rw-r--r-- | hypervideo_dl/extractor/gettr.py | 159 |
1 files changed, 131 insertions, 28 deletions
diff --git a/hypervideo_dl/extractor/gettr.py b/hypervideo_dl/extractor/gettr.py index aa50b2f..327a4d0 100644 --- a/hypervideo_dl/extractor/gettr.py +++ b/hypervideo_dl/extractor/gettr.py @@ -3,22 +3,30 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( + bool_or_none, ExtractorError, dict_get, float_or_none, int_or_none, - remove_end, str_or_none, + traverse_obj, try_get, url_or_none, urljoin, ) -class GettrIE(InfoExtractor): - _VALID_URL = r'https?://(www\.)?gettr\.com/post/(?P<id>[a-z0-9]+)' +class GettrBaseIE(InfoExtractor): + _BASE_REGEX = r'https?://(www\.)?gettr\.com/' _MEDIA_BASE_URL = 'https://media.gettr.com/' + def _call_api(self, path, video_id, *args, **kwargs): + return self._download_json(urljoin('https://api.gettr.com/u/', path), video_id, *args, **kwargs)['result'] + + +class GettrIE(GettrBaseIE): + _VALID_URL = GettrBaseIE._BASE_REGEX + r'post/(?P<id>[a-z0-9]+)' + _TESTS = [{ 'url': 'https://www.gettr.com/post/pcf6uv838f', 'info_dict': { @@ -28,9 +36,11 @@ class GettrIE(InfoExtractor): 'ext': 'mp4', 'uploader': 'EpochTV', 'uploader_id': 'epochtv', + 'upload_date': '20210927', 'thumbnail': r're:^https?://.+/out\.jpg', - 'timestamp': 1632782451058, + 'timestamp': 1632782451.058, 'duration': 58.5585, + 'tags': ['hornofafrica', 'explorations'], } }, { 'url': 'https://gettr.com/post/p4iahp', @@ -41,43 +51,69 @@ class GettrIE(InfoExtractor): 'ext': 'mp4', 'uploader': 'Neues Forum Freiheit', 'uploader_id': 'nf_freiheit', + 'upload_date': '20210718', 'thumbnail': r're:^https?://.+/out\.jpg', - 'timestamp': 1626594455017, + 'timestamp': 1626594455.017, 'duration': 23, + 'tags': 'count:12', } + }, { + # quote post + 'url': 'https://gettr.com/post/pxn5b743a9', + 'only_matching': True, + }, { + # quote with video + 'url': 'https://gettr.com/post/pxtiiz5ca2', + 'only_matching': True, + }, { + # streaming embed + 'url': 'https://gettr.com/post/pxlu8p3b13', + 'only_matching': True, + }, { + # youtube embed + 'url': 'https://gettr.com/post/pv6wp9e24c', + 'only_matching': True, + 'add_ie': ['Youtube'], }] def _real_extract(self, url): post_id = self._match_id(url) webpage = self._download_webpage(url, post_id) + api_data = self._call_api('post/%s?incl="poststats|userinfo"' % post_id, post_id) + + post_data = api_data.get('data') + user_data = try_get(api_data, lambda x: x['aux']['uinf'][post_data['uid']], dict) or {} + + vid = post_data.get('vid') + ovid = post_data.get('ovid') - api_data = self._download_json( - 'https://api.gettr.com/u/post/%s?incl="poststats|userinfo"' % post_id, post_id) + if post_data.get('p_type') == 'stream': + return self.url_result(f'https://gettr.com/streaming/{post_id}', ie='GettrStreaming', video_id=post_id) - post_data = try_get(api_data, lambda x: x['result']['data']) - user_data = try_get(api_data, lambda x: x['result']['aux']['uinf'][post_data['uid']]) or {} + if not (ovid or vid): + embed_url = url_or_none(post_data.get('prevsrc')) + shared_post_id = traverse_obj(api_data, ('aux', 'shrdpst', '_id'), ('data', 'rpstIds', 0), expected_type=str) - if post_data.get('nfound'): - raise ExtractorError(post_data.get('txt'), expected=True) + if embed_url: + return self.url_result(embed_url) + elif shared_post_id: + return self.url_result(f'https://gettr.com/post/{shared_post_id}', ie='Gettr', video_id=shared_post_id) + else: + raise ExtractorError('There\'s no video in this post.') title = description = str_or_none( post_data.get('txt') or self._og_search_description(webpage)) uploader = str_or_none( user_data.get('nickname') - or remove_end(self._og_search_title(webpage), ' on GETTR')) + or self._search_regex(r'^(.+?) on GETTR', self._og_search_title(webpage, default=''), 'uploader', fatal=False)) + if uploader: title = '%s - %s' % (uploader, title) - if not dict_get(post_data, ['vid', 'ovid']): - raise ExtractorError('There\'s no video in this post.') - - vid = post_data.get('vid') - ovid = post_data.get('ovid') - - formats = self._extract_m3u8_formats( + formats, subtitles = self._extract_m3u8_formats_and_subtitles( urljoin(self._MEDIA_BASE_URL, vid), post_id, 'mp4', - entry_protocol='m3u8_native', m3u8_id='hls') if vid else [] + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False) if vid else ([], {}) if ovid: formats.append({ @@ -86,8 +122,6 @@ class GettrIE(InfoExtractor): 'ext': 'mp4', 'width': int_or_none(post_data.get('vid_wid')), 'height': int_or_none(post_data.get('vid_hgt')), - 'source_preference': 1, - 'quality': 1, }) self._sort_formats(formats) @@ -96,15 +130,84 @@ class GettrIE(InfoExtractor): 'id': post_id, 'title': title, 'description': description, - 'thumbnail': url_or_none( - urljoin(self._MEDIA_BASE_URL, post_data.get('main')) - or self._og_search_thumbnail(webpage)), - 'timestamp': int_or_none(post_data.get('cdate')), + 'formats': formats, + 'subtitles': subtitles, + 'uploader': uploader, 'uploader_id': str_or_none( dict_get(user_data, ['_id', 'username']) or post_data.get('uid')), - 'uploader': uploader, - 'formats': formats, + 'thumbnail': url_or_none( + urljoin(self._MEDIA_BASE_URL, post_data.get('main')) + or self._html_search_meta(['og:image', 'image'], webpage, 'thumbnail', fatal=False)), + 'timestamp': float_or_none(dict_get(post_data, ['cdate', 'udate']), scale=1000), 'duration': float_or_none(post_data.get('vid_dur')), 'tags': post_data.get('htgs'), } + + +class GettrStreamingIE(GettrBaseIE): + _VALID_URL = GettrBaseIE._BASE_REGEX + r'streaming/(?P<id>[a-z0-9]+)' + + _TESTS = [{ + 'url': 'https://gettr.com/streaming/psoiulc122', + 'info_dict': { + 'id': 'psoiulc122', + 'ext': 'mp4', + 'description': 'md5:56bca4b8f48f1743d9fd03d49c723017', + 'view_count': int, + 'uploader': 'Corona Investigative Committee', + 'uploader_id': 'coronacommittee', + 'duration': 5180.184, + 'thumbnail': r're:^https?://.+', + 'title': 'Day 1: Opening Session of the Grand Jury Proceeding', + 'timestamp': 1644080997.164, + 'upload_date': '20220205', + } + }, { + 'url': 'https://gettr.com/streaming/psfmeefcc1', + 'info_dict': { + 'id': 'psfmeefcc1', + 'ext': 'mp4', + 'title': 'Session 90: "The Virus Of Power"', + 'view_count': int, + 'uploader_id': 'coronacommittee', + 'description': 'md5:98986acdf656aa836bf36f9c9704c65b', + 'uploader': 'Corona Investigative Committee', + 'thumbnail': r're:^https?://.+', + 'duration': 21872.507, + 'timestamp': 1643976662.858, + 'upload_date': '20220204', + } + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + video_info = self._call_api('live/join/%s' % video_id, video_id, data={}) + + live_info = video_info['broadcast'] + live_url = url_or_none(live_info.get('url')) + + formats, subtitles = self._extract_m3u8_formats_and_subtitles( + live_url, video_id, ext='mp4', + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False) if live_url else ([], {}) + + thumbnails = [{ + 'url': urljoin(self._MEDIA_BASE_URL, thumbnail), + } for thumbnail in try_get(video_info, lambda x: x['postData']['imgs'], list) or []] + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': try_get(video_info, lambda x: x['postData']['ttl'], str), + 'description': try_get(video_info, lambda x: x['postData']['dsc'], str), + 'formats': formats, + 'subtitles': subtitles, + 'thumbnails': thumbnails, + 'uploader': try_get(video_info, lambda x: x['liveHostInfo']['nickname'], str), + 'uploader_id': try_get(video_info, lambda x: x['liveHostInfo']['_id'], str), + 'view_count': int_or_none(live_info.get('viewsCount')), + 'timestamp': float_or_none(live_info.get('startAt'), scale=1000), + 'duration': float_or_none(live_info.get('duration'), scale=1000), + 'is_live': bool_or_none(live_info.get('isLive')), + } |