aboutsummaryrefslogtreecommitdiff
path: root/components/extractor
diff options
context:
space:
mode:
Diffstat (limited to 'components/extractor')
-rw-r--r--components/extractor/obtain_vid_info.py13
1 files changed, 13 insertions, 0 deletions
diff --git a/components/extractor/obtain_vid_info.py b/components/extractor/obtain_vid_info.py
new file mode 100644
index 0000000..bfbe861
--- /dev/null
+++ b/components/extractor/obtain_vid_info.py
@@ -0,0 +1,13 @@
+from urllib.request import urlopen
+
+from bs4 import BeautifulSoup
+from isodate import parse_duration # type: ignore
+
+def obtain_vid_duration(url: str, html: str = '') -> int:
+ html = html or urlopen(url).read().decode('utf-8')
+ soup = BeautifulSoup(html, 'html.parser')
+
+ duration_meta = soup.find('meta', itemprop='duration')
+ assert duration_meta
+ duration = parse_duration(duration_meta['content'])
+ return int(duration.total_seconds())