maryjane
/
gpodder
1
0
Fork 0
Code Issues Pull Requests Projects Releases 1 Wiki Activity

Improve Youtube initial player response regular expression. 

The text may contain "};" patterns inside the initial player response.
And the non-greedy regex would fail to match the entire IPR, causing it
to fail during JSON decoding.
This commit is contained in:
auouymous 2021-10-13 21:58:24 -06:00
parent 8d0bfce460
commit 6ced881d74
1 changed files with 20 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
src/gpodder/youtube.py
View File

@ -147,8 +147,21 @@ hls_formats = [
]
hls_formats_dict = dict(hls_formats)
V3_API_ENDPOINT = 'https://www.googleapis.com/youtube/v3'
CHANNEL_VIDEOS_XML = 'https://www.youtube.com/feeds/videos.xml'
WATCH_ENDPOINT = 'https://www.youtube.com/watch?bpctr=9999999999&has_verified=1&v='
# The page may contain "};" sequences inside the initial player response.
# Use a greedy match with script end tag, and fallback to a non-greedy match without.
INITIAL_PLAYER_RESPONSE_RE1 = r'ytInitialPlayerResponse\s*=\s*({.+})\s*;\s*</script'
INITIAL_PLAYER_RESPONSE_RE2 = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
def get_ipr(page):
for regex in (INITIAL_PLAYER_RESPONSE_RE1, INITIAL_PLAYER_RESPONSE_RE2):
ipr = re.search(regex, page)
if ipr is not None:
return ipr
return None
class YouTubeError(Exception):
@ -198,12 +211,12 @@ def youtube_get_old_endpoint(vid):
def youtube_get_new_endpoint(vid):
url = 'https://www.youtube.com/watch?bpctr=9999999999&has_verified=1&v=' + vid
url = WATCH_ENDPOINT + vid
r = util.urlopen(url)
if not r.ok:
raise YouTubeError('Youtube "%s": %d %s' % (url, r.status_code, r.reason))
ipr = re.search(r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;', r.text)
ipr = get_ipr(r.text)
if ipr is None:
try:
url = get_gdpr_consent_url(r.text)
@ -213,7 +226,7 @@ def youtube_get_new_endpoint(vid):
if not r.ok:
raise YouTubeError('Youtube "%s": %d %s' % (url, r.status_code, r.reason))
ipr = re.search(r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;', r.text)
ipr = get_ipr(r.text)
if ipr is None:
raise YouTubeError('Youtube "%s": No ytInitialPlayerResponse found' % url)
@ -226,19 +239,19 @@ def get_total_time(episode):
if vid is None:
return 0
url = 'https://www.youtube.com/watch?v=' + vid
url = WATCH_ENDPOINT + vid
r = util.urlopen(url)
if not r.ok:
return 0
ipr = re.search(r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;', r.text)
ipr = get_ipr(r.text)
if ipr is None:
url = get_gdpr_consent_url(r.text)
r = util.urlopen(url)
if not r.ok:
return 0
ipr = re.search(r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;', r.text)
ipr = get_ipr(r.text)
if ipr is None:
return 0