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

yt-dlp proxy handling update

Browse Source
This commit is contained in:
Barbara Miller 2025-01-23 10:58:32 -08:00
parent eb922f5155
commit 170377fe89
3 changed files with 23 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
brozzler/cli.py
View File

@ -2,7 +2,7 @@
"""
brozzler/cli.py - brozzler command line executables
Copyright (C) 2014-2024 Internet Archive
Copyright (C) 2014-2025 Internet Archive
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
@ -611,14 +611,29 @@ def brozzler_worker(argv=None):
logging.info("running with empty skip_av_seeds")
return skip_av_seeds
def get_proxy_endpoints():
PROXY_ENDPOINTS_FILE = "/opt/local/brozzler/proxy_endpoints.txt"
try:
# make list from file
with open(PROXY_ENDPOINTS_FILE) as endpoints:
proxy_endpoints = [l for l in endpoints.readlines()]
if proxy_endpoints:
logging.info("running with proxy endpoints file %s" % PROXY_ENDPOINTS_FILE)
except Exception as e:
proxy_endpoints = []
logging.info("running with empty proxy endpoints file")
return proxy_endpoints
rr = rethinker(args)
frontier = brozzler.RethinkDbFrontier(rr)
service_registry = doublethink.ServiceRegistry(rr)
skip_av_seeds_from_file = get_skip_av_seeds()
proxy_endpoints_from_file = get_proxy_endpoints()
worker = brozzler.worker.BrozzlerWorker(
frontier,
service_registry,
skip_av_seeds=skip_av_seeds_from_file,
proxy_endpoints=proxy_endpoints_from_file,
max_browsers=int(args.max_browsers),
chrome_exe=args.chrome_exe,
proxy=args.proxy,

4
brozzler/worker.py
View File

@ -57,6 +57,7 @@ class BrozzlerWorker:
frontier,
service_registry=None,
skip_av_seeds=None,
proxy_endpoints=None,
max_browsers=1,
chrome_exe="chromium-browser",
warcprox_auto=False,
@ -81,6 +82,7 @@ class BrozzlerWorker:
self._frontier = frontier
self._service_registry = service_registry
self._skip_av_seeds = skip_av_seeds
self._proxy_endpoints = proxy_endpoints
self._max_browsers = max_browsers
self._warcprox_auto = warcprox_auto
@ -287,7 +289,7 @@ class BrozzlerWorker:
self.logger.info("page interstitial shown (http auth): %s", page)
if enable_youtube_dl and ydl.should_ytdlp(
site, page, status_code, self._skip_av_seeds
site, page, status_code, self._skip_av_seeds, self._proxy_endpoints
):
try:
ydl_outlinks = ydl.do_youtube_dl(self, site, page)

9
brozzler/ydl.py
View File

@ -34,12 +34,11 @@ import time
thread_local = threading.local()
YTDLP_PROXY = ""
PROXY_ATTEMPTS = 4
YTDLP_WAIT = 10
def should_ytdlp(site, page, page_status, skip_av_seeds):
def should_ytdlp(site, page, page_status, skip_av_seeds, proxy_endpoints):
# called only after we've passed needs_browsing() check
if page_status != 200:
@ -285,11 +284,11 @@ def _build_youtube_dl(worker, destdir, site, page):
ytdlp_url = page.redirect_url if page.redirect_url else page.url
is_youtube_host = isyoutubehost(ytdlp_url)
if is_youtube_host and YTDLP_PROXY:
ydl_opts["proxy"] = YTDLP_PROXY
if is_youtube_host and proxy_endpoints:
ydl_opts["proxy"] = random.choice(proxy_endpoints)
# don't log proxy value secrets
ytdlp_proxy_for_logs = (
YTDLP_PROXY.split("@")[1] if "@" in YTDLP_PROXY else "@@@"
ydl_opts["proxy"].split("@")[1] if "@" in ydl_opts["proxy"] else "@@@"
)
logging.info("using yt-dlp proxy ... %s", ytdlp_proxy_for_logs)