From 70d6ced2cb7ce846178e9f91ae2b9838e322383f Mon Sep 17 00:00:00 2001 From: ThePhaseless Date: Thu, 8 Oct 2026 00:03:59 +0200 Subject: [PATCH] fix: resolve the browser timezone once per process instead of per request Every request builds a new InvisiblePlaywright, and its __aenter__ calls prepare_session_geo(): with the default timezone="" and no proxy that discovers the egress IP through api.ipify.org, icanhazip.com and checkip.amazonaws.com, then calls ensure_geoip_mmdb(), which HEADs GitHub for a newer GeoIP release, three times per session. On a network where those hosts are slow or blocked every request paid for it before the browser even started. Measured in the app image: ipify black-holed adds ~5s per request, all three echo hosts black-holed ~12.5s, GitHub black-holed ~75s. Setting BROWSER_LOCALE alone did not help, because the timezone was still "auto". Without a proxy the egress is the host's own, so the answer does not change between requests. Resolve it on the first request, off the event loop and under a lock so concurrent first requests share one lookup, then hand the zone and the locale's primary tag to InvisiblePlaywright explicitly; an explicit zone and tag make prepare_session_geo touch no network. The resulting browser is the same as with "auto": passing the primary tag back reproduces the decided language list for every Firefox build the locale table can pick, and without a proxy the egress IP, coordinates and srflx are not used by InvisiblePlaywright anyway. When the lookup fails upstream falls back to the host timezone by passing "", but passing "" again would repeat the lookup on every request, so the host zone is named explicitly (TZ, else the /etc/localtime link, else UTC) and the lookup is retried after ten minutes. Sessions behind a proxy (PROXY_SERVER or X-Proxy-Server) still resolve per request through the proxy: each proxy has its own egress, and upstream deliberately fails the launch when it cannot be resolved. BROWSER_TIMEZONE mirrors BROWSER_LOCALE and wins also behind a proxy; with both set the lookup is skipped entirely. Fixes #416 Co-Authored-By: Claude Opus 5.5 --- README.md | 9 ++++ src/consts.py | 2 + src/utils.py | 60 ++++++++++++++++++++- tests/utils_test.py | 126 ++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 196 insertions(+), 1 deletion(-) create mode 100644 tests/utils_test.py diff --git a/README.md b/README.md index f374e7d..3305473 100644 --- a/README.md +++ b/README.md @@ -18,6 +18,7 @@ | `PROXY_PASSWORD` | None | Password for proxy authentication. | | `OWUI_API_KEY` | None | Bearer token for `/load` endpoint authentication. Must match `EXTERNAL_WEB_LOADER_API_KEY` in Open WebUI. | | `BROWSER_LOCALE` | None | Override the browser's language with a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) tag, e.g. `en-US`, `de-DE`, `fr-FR`. When unset, the locale is derived from the egress country. | +| `BROWSER_TIMEZONE` | None | Override the browser's timezone with an [IANA](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones) name, e.g. `Europe/Berlin`, `Asia/Shanghai`. When unset, the timezone is derived from the egress IP. | #### Browser language @@ -25,6 +26,14 @@ Set `BROWSER_LOCALE` to a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) Valid tags are maintained in the [IANA Language Subtag Registry](https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry). For a friendlier list, see [List of ISO 639-1 codes](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (language) combined with an [ISO 3166-1 alpha-2](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2) region code for the full tag, e.g. `pt-BR`. +#### Browser timezone + +Set `BROWSER_TIMEZONE` to an [IANA timezone name](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones) like `Europe/Berlin` or `Asia/Shanghai` to fix the browser's timezone. An explicit value is used as is, also behind a proxy. + +When either `BROWSER_TIMEZONE` or `BROWSER_LOCALE` is unset, Byparr looks up its public IP through `api.ipify.org`, `icanhazip.com` or `checkip.amazonaws.com` and maps it to a timezone and language with a GeoIP database downloaded from GitHub. Without a proxy this happens once, on the first request, and the result is reused until Byparr restarts (restart it if its public IP moves to another country). If that lookup fails, the host timezone (`TZ`, otherwise `/etc/localtime`, otherwise UTC) and `en-US` are used, and the lookup is retried after 10 minutes. With a proxy (`PROXY_SERVER` or the `X-Proxy-Server` header) it runs for every request, through the proxy. + +If those hosts are slow or blocked on your network, set both `BROWSER_TIMEZONE` and `BROWSER_LOCALE`: the lookup is then skipped entirely and no request is made before the browser starts. + ## Proxy Recommendation Recently I've partnered with a _new in town_ proxy service - ProxyBase - to offer affordable proxy services that seems to work seamlessly with Byparr! Using my affiliate code `byparr` (case sensitive!) when signing up will not only get you access to their cost-effective (**$0.69/GB with occasional promotions** _at the time of writing_) proxy network but will also help support the continued development of this project. ProxyBase's proxies can significantly improve your success rate when bypassing anti-bot challenges. [Check out ProxyBase](https://client.proxybase.org/signup?ref=byparr) and enhance your Byparr experience! diff --git a/src/consts.py b/src/consts.py index d2f2a34..d59a7bb 100644 --- a/src/consts.py +++ b/src/consts.py @@ -20,6 +20,7 @@ class Settings(BaseSettings): return_only_cookies: bool = False owui_api_key: str | None = None browser_locale: str | None = None + browser_timezone: str | None = None settings = Settings() @@ -39,3 +40,4 @@ RETURN_ONLY_COOKIES = settings.return_only_cookies OWUI_API_KEY = settings.owui_api_key BROWSER_LOCALE = settings.browser_locale +BROWSER_TIMEZONE = settings.browser_timezone diff --git a/src/utils.py b/src/utils.py index 5d7a94e..1272075 100644 --- a/src/utils.py +++ b/src/utils.py @@ -1,15 +1,20 @@ +import asyncio import logging +import math +import os import time from collections.abc import AsyncGenerator from typing import Annotated, NamedTuple, cast from fastapi import Header +from invisible_core import prepare_session_geo from invisible_playwright.async_api import InvisiblePlaywright from playwright.async_api import Browser, BrowserContext, Page from pydantic import BaseModel, Field from src.consts import ( BROWSER_LOCALE, + BROWSER_TIMEZONE, LOG_LEVEL, PROXY_PASSWORD, PROXY_SERVER, @@ -52,6 +57,54 @@ class BrowserDepClass(NamedTuple): context: BrowserContext +GEO_RETRY_SECONDS = 600 + + +class BrowserGeo(NamedTuple): + timezone: str + locale: str + expires_at: float = math.inf + + +_geo_lock = asyncio.Lock() +_geo: BrowserGeo | None = None + + +def host_timezone() -> str: + """IANA zone of the host, the one the browser falls back to.""" + if tz := os.environ.get("TZ", "").removeprefix(":"): + return tz + _, found, zone = os.path.realpath("/etc/localtime").partition("/zoneinfo/") + return zone if found else "UTC" + + +def resolve_browser_geo() -> BrowserGeo: + """Resolve the timezone and locale of a session without a proxy.""" + geo = prepare_session_geo(BROWSER_TIMEZONE or "", None, BROWSER_LOCALE or "auto") + locale = BROWSER_LOCALE or (geo.locale.primary if geo.locale else "en-US") + if geo.timezone: + logger.info("Browser timezone %s, locale %s", geo.timezone, locale) + return BrowserGeo(geo.timezone, locale) + timezone = host_timezone() + logger.warning( + "Could not resolve the timezone from the egress IP, using %s and locale %s " + "for %d seconds. Set BROWSER_TIMEZONE and BROWSER_LOCALE to skip the lookup.", + timezone, + locale, + GEO_RETRY_SECONDS, + ) + return BrowserGeo(timezone, locale, time.monotonic() + GEO_RETRY_SECONDS) + + +async def get_browser_geo() -> BrowserGeo: + """Per-process cache of resolve_browser_geo().""" + global _geo # noqa: PLW0603 + async with _geo_lock: + if _geo is None or time.monotonic() >= _geo.expires_at: + _geo = await asyncio.to_thread(resolve_browser_geo) + return _geo + + async def get_browser( x_proxy_server: Annotated[ str | None, @@ -79,6 +132,8 @@ async def get_browser( header_password = x_proxy_password proxy_config = None + timezone = BROWSER_TIMEZONE or "" + locale = BROWSER_LOCALE or "auto" if header_server: proxy_config = { @@ -92,12 +147,15 @@ async def get_browser( "username": PROXY_USERNAME, "password": PROXY_PASSWORD, } + else: + timezone, locale, _ = await get_browser_geo() async with InvisiblePlaywright( headless=True, proxy=proxy_config, humanize=True, - locale=BROWSER_LOCALE or "auto", + locale=locale, + timezone=timezone, extra_prefs={ "devtools.jsonview.enabled": False, "browser.tabs.remote.useCrossOriginOpenerPolicy": False, diff --git a/tests/utils_test.py b/tests/utils_test.py new file mode 100644 index 0000000..4d88ae3 --- /dev/null +++ b/tests/utils_test.py @@ -0,0 +1,126 @@ +import asyncio +import math +from unittest.mock import AsyncMock, MagicMock, patch + +import pytest +from invisible_core import SessionGeo, SessionLocale + +from src import utils + +BERLIN = SessionGeo( + "Europe/Berlin", + "203.0.113.7", + locale=SessionLocale(languages=("de", "en-US", "en"), region="DE"), +) +FAILED = SessionGeo( + "", None, locale=SessionLocale(languages=("en-US", "en"), region="US") +) + + +@pytest.fixture(autouse=True) +def fresh_geo_cache(monkeypatch: pytest.MonkeyPatch): + """Every test starts with nothing resolved.""" + monkeypatch.setattr(utils, "_geo", None) + monkeypatch.setattr(utils, "_geo_lock", asyncio.Lock()) + + +def test_host_timezone_prefers_tz_env(monkeypatch: pytest.MonkeyPatch): + """TZ wins over /etc/localtime, with or without the leading colon.""" + monkeypatch.setenv("TZ", ":Asia/Shanghai") + assert utils.host_timezone() == "Asia/Shanghai" + + +def test_host_timezone_reads_localtime_symlink(monkeypatch: pytest.MonkeyPatch): + """Without TZ the zone comes from where /etc/localtime points.""" + monkeypatch.delenv("TZ", raising=False) + with patch( + "src.utils.os.path.realpath", return_value="/usr/share/zoneinfo/Europe/Warsaw" + ): + assert utils.host_timezone() == "Europe/Warsaw" + + +def test_host_timezone_defaults_to_utc(monkeypatch: pytest.MonkeyPatch): + """No TZ and no zoneinfo link is what the Docker image has: UTC.""" + monkeypatch.delenv("TZ", raising=False) + with patch("src.utils.os.path.realpath", return_value="/etc/localtime"): + assert utils.host_timezone() == "UTC" + + +@pytest.mark.asyncio +async def test_egress_lookup_runs_once_per_process(): + """Concurrent first requests and every later one share a single lookup.""" + with patch("src.utils.prepare_session_geo", return_value=BERLIN) as lookup: + results = await asyncio.gather(*(utils.get_browser_geo() for _ in range(3))) + results.append(await utils.get_browser_geo()) + + lookup.assert_called_once_with("", None, "auto") + assert set(results) == {utils.BrowserGeo("Europe/Berlin", "de", math.inf)} + + +@pytest.mark.asyncio +async def test_failed_lookup_uses_host_timezone_then_retries( + monkeypatch: pytest.MonkeyPatch, +): + """A failed lookup is not repaid per request, but is retried later.""" + monkeypatch.setenv("TZ", "Asia/Shanghai") + with patch("src.utils.prepare_session_geo", return_value=FAILED) as lookup: + geo = await utils.get_browser_geo() + assert (geo.timezone, geo.locale) == ("Asia/Shanghai", "en-US") + await utils.get_browser_geo() + assert lookup.call_count == 1 + + monkeypatch.setattr(utils, "_geo", geo._replace(expires_at=0)) + await utils.get_browser_geo() + assert lookup.call_count == 2 # noqa: PLR2004 + + +@pytest.mark.asyncio +async def test_explicit_timezone_and_locale_touch_no_network(): + """BROWSER_TIMEZONE with BROWSER_LOCALE needs neither egress IP nor geoip.""" + with ( + patch("src.utils.BROWSER_TIMEZONE", "Asia/Shanghai"), + patch("src.utils.BROWSER_LOCALE", "zh-CN"), + patch("invisible_core._geo.discover_egress_ip") as discover, + patch("invisible_core.download.ensure_geoip_mmdb") as geoip, + ): + geo = await utils.get_browser_geo() + + assert (geo.timezone, geo.locale) == ("Asia/Shanghai", "zh-CN") + discover.assert_not_called() + geoip.assert_not_called() + + +def fake_invisible_playwright() -> MagicMock: + """Stand-in for InvisiblePlaywright that records how it was built.""" + browser = MagicMock() + browser.new_context = AsyncMock(return_value=AsyncMock()) + session = MagicMock() + session.__aenter__ = AsyncMock(return_value=browser) + session.__aexit__ = AsyncMock(return_value=False) + return MagicMock(return_value=session) + + +@pytest.mark.asyncio +@pytest.mark.parametrize( + ("proxy_header", "expected"), + [ + (None, {"timezone": "Europe/Berlin", "locale": "de"}), + ("socks5://proxy.test:1080", {"timezone": "", "locale": "auto"}), + ], +) +async def test_proxy_sessions_resolve_their_own_egress( + proxy_header: str | None, expected: dict[str, str] +): + """The cache only covers direct sessions; a proxy's egress is its own.""" + launcher = fake_invisible_playwright() + with ( + patch("src.utils.PROXY_SERVER", None), + patch("src.utils.InvisiblePlaywright", launcher), + patch("src.utils.prepare_session_geo", return_value=BERLIN) as lookup, + ): + async for _ in utils.get_browser(x_proxy_server=proxy_header): + pass + + kwargs = launcher.call_args.kwargs + assert {"timezone": kwargs["timezone"], "locale": kwargs["locale"]} == expected + assert lookup.called is (proxy_header is None)