summaryrefslogtreecommitdiffhomepage
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
authorRemita Amine <[email protected]>2017-05-02 00:06:19 +0100
committerRemita Amine <[email protected]>2017-05-02 00:06:58 +0100
commit7fc875195fd74c235d6ca86afc642f683a485a4b (patch)
treeb4d930b4fa53ee0bd8b39993a432d9b63a95a943 /youtube_dl/extractor
parentc6fe5a7e12a5ebbe137b889afb8aefc206d132af (diff)
downloadyoutube-dl-7fc875195fd74c235d6ca86afc642f683a485a4b.tar.gz
youtube-dl-7fc875195fd74c235d6ca86afc642f683a485a4b.zip
[amp] imporove thumbnail and subtitle extraction
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/amp.py18
1 files changed, 13 insertions, 5 deletions
diff --git a/youtube_dl/extractor/amp.py b/youtube_dl/extractor/amp.py
index 98f8e69cd..fde1a8ff7 100644
--- a/youtube_dl/extractor/amp.py
+++ b/youtube_dl/extractor/amp.py
@@ -34,9 +34,12 @@ class AMPIE(InfoExtractor):
if isinstance(media_thumbnail, dict):
media_thumbnail = [media_thumbnail]
for thumbnail_data in media_thumbnail:
- thumbnail = thumbnail_data['@attributes']
+ thumbnail = thumbnail_data.get('@attributes', {})
+ thumbnail_url = thumbnail.get('url')
+ if not thumbnail_url:
+ continue
thumbnails.append({
- 'url': self._proto_relative_url(thumbnail['url'], 'http:'),
+ 'url': self._proto_relative_url(thumbnail_url, 'http:'),
'width': int_or_none(thumbnail.get('width')),
'height': int_or_none(thumbnail.get('height')),
})
@@ -47,9 +50,14 @@ class AMPIE(InfoExtractor):
if isinstance(media_subtitle, dict):
media_subtitle = [media_subtitle]
for subtitle_data in media_subtitle:
- subtitle = subtitle_data['@attributes']
- lang = subtitle.get('lang') or 'en'
- subtitles[lang] = [{'url': subtitle['href']}]
+ subtitle = subtitle_data.get('@attributes', {})
+ subtitle_href = subtitle.get('href')
+ if not subtitle_href:
+ continue
+ subtitles.setdefault(subtitle.get('lang') or 'en', []).append({
+ 'url': subtitle_href,
+ 'ext': mimetype2ext(subtitle.get('type')) or determine_ext(subtitle_href),
+ })
formats = []
media_content = get_media_node('content')