diff options
| author | Remita Amine <remitamine@gmail.com> | 2021-01-13 09:17:43 +0100 |
|---|---|---|
| committer | Remita Amine <remitamine@gmail.com> | 2021-01-13 09:17:43 +0100 |
| commit | 2408e6d26a9a1bd899769903ac28a8deaaa778a3 (patch) | |
| tree | 947cf56bc175ce10e4527dc37ee57ce08a0f302e | |
| parent | cf862771d72b028f8d5335256610c560245dc8c0 (diff) | |
[cspan] improve info extraction(closes #27791)
| -rw-r--r-- | youtube_dl/extractor/cspan.py | 27 |
1 files changed, 26 insertions, 1 deletions
diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py index 766942146..2e01aff48 100644 --- a/youtube_dl/extractor/cspan.py +++ b/youtube_dl/extractor/cspan.py @@ -8,11 +8,14 @@ from ..utils import ( ExtractorError, extract_attributes, find_xpath_attr, + get_element_by_attribute, get_element_by_class, int_or_none, js_to_json, merge_dicts, + parse_iso8601, smuggle_url, + str_to_int, unescapeHTML, ) from .senateisvp import SenateISVPIE @@ -116,8 +119,30 @@ class CSpanIE(InfoExtractor): jwsetup, video_id, require_title=False, m3u8_id='hls', base_url=url) add_referer(info['formats']) + for subtitles in info['subtitles'].values(): + for subtitle in subtitles: + ext = determine_ext(subtitle['url']) + if ext == 'php': + ext = 'vtt' + subtitle['ext'] = ext ld_info = self._search_json_ld(webpage, video_id, default={}) - return merge_dicts(info, ld_info) + title = get_element_by_class('video-page-title', webpage) or \ + self._og_search_title(webpage) + description = get_element_by_attribute('itemprop', 'description', webpage) or \ + self._html_search_meta(['og:description', 'description'], webpage) + return merge_dicts(info, ld_info, { + 'title': title, + 'thumbnail': get_element_by_attribute('itemprop', 'thumbnailUrl', webpage), + 'description': description, + 'timestamp': parse_iso8601(get_element_by_attribute('itemprop', 'uploadDate', webpage)), + 'location': get_element_by_attribute('itemprop', 'contentLocation', webpage), + 'duration': int_or_none(self._search_regex( + r'jwsetup\.seclength\s*=\s*(\d+);', + webpage, 'duration', fatal=False)), + 'view_count': str_to_int(self._search_regex( + r"<span[^>]+class='views'[^>]*>([\d,]+)\s+Views</span>", + webpage, 'views', fatal=False)), + }) # Obsolete # We first look for clipid, because clipprog always appears before |