Git-Mirrors/brozzler
1
0
Fork 0
mirror of https://github.com/internetarchive/brozzler.git synced 2025-04-20 23:56:34 -04:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Merge branch 'adam/patch-yt-dlp-infinite-loop-bug' into qa

Browse Source
This commit is contained in:
Adam Miller 2025-01-06 11:38:44 -08:00
commit 6abc23d441
1 changed files with 23 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
brozzler/ydl.py
View File

@ -41,6 +41,7 @@ thread_local = threading.local()
YTDLP_PROXY = ""
PROXY_ATTEMPTS = 4
YTDLP_WAIT = 10
YTDLP_MAX_REDIRECTS = 5
def _timestamp4datetime(timestamp):
@ -121,24 +122,34 @@ def _build_youtube_dl(worker, destdir, site, page):
"""Custom Generic Information Extractor to detect redirect loops."""
logger = logging.getLogger(__module__ + "." + __qualname__)
shared_visited_urls = set() # Shared state for all instances
visited_redirect_urls = set()
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.logger.info("[CustomGenericIE] Initialized")
def _real_extract(self, url):
# Check for redirect loops in the shared state
if url in self.shared_visited_urls:
self.logger.error("Redirect loop detected for URL: {url}")
raise ExtractorError(
f"Redirect loop detected for URL: {url}",
expected=True, # Marks the error as non-fatal
)
self.shared_visited_urls.add(url)
# self.visited_redirect_urls.clear()
self.logger.info(f"[CustomGenericIE] Extracting URL: {url}")
return super()._real_extract(url)
def report_following_redirect(self, new_url):
self.logger.info(
f"[CustomGenericIE] Following redirect URL: {new_url} "
f"redirect_count: {len(self.visited_redirect_urls)}"
)
if new_url in self.visited_redirect_urls:
raise ExtractorError(
f"Redirect loop detected for URL: {new_url}",
expected=True,
)
if len(self.visited_redirect_urls) > YTDLP_MAX_REDIRECTS:
raise ExtractorError(
f"Too many redirects for URL: {new_url}",
expected=True,
)
self.visited_redirect_urls.add(new_url)
return super().report_following_redirect(new_url)
yt_dlp.extractor.generic.GenericIE = CustomGenericIE
class _YoutubeDL(yt_dlp.YoutubeDL):
@ -392,9 +403,8 @@ def _try_youtube_dl(worker, ydl, site, page):
and e.exc_info[1].code == 420
):
raise brozzler.ReachedLimit(e.exc_info[1])
elif (
isinstance(e, yt_dlp.utils.DownloadError)
and "Redirect loop detected" in e.msg
elif isinstance(e, yt_dlp.utils.DownloadError) and (
"Redirect loop detected" in e.msg or "Too many redirects" in e.msg
):
raise brozzler.VideoExtractorError(e.msg)
else: