From 80cb6404b61c50546381e16db8728d86c3d89167 Mon Sep 17 00:00:00 2001
From: Heo Hao <130041734+Shelden-Hao@users.noreply.github.com>
Date: Wed, 11 Nov 2026 19:43:53 +0800
Subject: [PATCH] fix(browserless): reject timeouts httpx cannot use (#8025)

* fix(browserless): reject timeouts httpx cannot use

`_coerce_timeout` forwarded any int, float or numeric string to
`httpx.AsyncClient(timeout=...)` unchanged, so a configured `timeout_s: 0`
or a negative value failed every fetch and capture with `ConnectTimeout`
against a healthy backend, `nan` raised `ValueError: deadline must not be
NaN`, and `inf` left the request without a deadline so it never completed.
Each of those silently disabled the provider.

Fall back to the documented 30s default with a warning when the number is
not a usable deadline. This matches `crawl4ai._coerce_timeout` after #7481;
`jina_ai._coerce_timeout` still forwards non-positive values.

* fix(browserless): address timeout review and format tests

Explain how HTTPX handles NaN and positive infinity. Document that
Browserless accepts positive finite timeouts without an upper limit,
and pin that behavior with a 1e308 test case.

Crawl4AI and Jina timeout validation are tracked in #7481 and #7823.
Remove the Browserless docstring's claim that Crawl4AI shares its
coercion behavior. Format the test file for the backend lint check.

---------

Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
---
 backend/docs/CONFIGURATION.md                 |  8 ++++
 .../deerflow/community/browserless/tools.py   | 24 ++++++-----
 backend/tests/test_browserless_client.py      | 41 +++++++++++++++++++
 docs/guide/configuration.md                   |  2 +
 4 files changed, 64 insertions(+), 11 deletions(-)

diff --git a/backend/docs/CONFIGURATION.md b/backend/docs/CONFIGURATION.md
index 1fac382419..a795ead967 100644
--- a/backend/docs/CONFIGURATION.md
+++ b/backend/docs/CONFIGURATION.md
@@ -1372,6 +1372,14 @@ shared (`100.64.0.0/10`, used by CGNAT and Tailscale), other non-global, or
 cloud-metadata addresses; set `allow_private_addresses: true` only when you
 intentionally point the tool at an internal target.
 
+Each Browserless tool entry reads its request timeout from `timeout_s`, with
+`timeout` as an alias when `timeout_s` is absent. The default is 30 seconds.
+Zero, negative, and non-finite values fall back to the default with a warning.
+Strings that cannot be parsed as numbers also trigger this fallback. HTTPX can
+time out immediately for non-positive values. It raises an error for NaN and
+applies no deadline for positive infinity. Browserless preserves positive finite
+values in seconds, including very large values.
+
 Both `web_fetch` (Browserless provider) and `web_capture` need a running
 Browserless instance. You can point `base_url` at [Browserless Cloud](https://www.browserless.io/)
 (set `BROWSERLESS_TOKEN`) or run one locally with Docker:
diff --git a/backend/packages/harness/deerflow/community/browserless/tools.py b/backend/packages/harness/deerflow/community/browserless/tools.py
index 3184395840..16042fe731 100644
--- a/backend/packages/harness/deerflow/community/browserless/tools.py
+++ b/backend/packages/harness/deerflow/community/browserless/tools.py
@@ -1,5 +1,6 @@
 import asyncio
 import logging
+import math
 import os
 import re
 from datetime import UTC, datetime
@@ -48,22 +49,23 @@ def _get_tool_config(tool_name: str) -> dict | None:
 
 
 def _coerce_timeout(value: object, default: float) -> float:
-    """Coerce a config timeout into seconds, falling back to ``default`` on bad input.
+    """Parse Browserless timeout seconds, falling back to ``default`` on bad input.
 
-    Mirrors ``crawl4ai._coerce_timeout`` / ``jina_ai._coerce_timeout``: booleans and
-    non-numeric strings fall back to the default, so ``timeout_s: off`` (YAML ``False``)
-    does not become ``0.0`` and time out every request against a healthy server, and a
-    typo'd value does not raise out of tool construction.
+    Reject booleans before conversion so YAML ``timeout_s: off`` does not
+    become an immediate timeout. HTTPX treats non-positive deadlines as expired,
+    raises for NaN and treats positive infinity as an unlimited deadline.
+    Preserve positive finite values without imposing an upper limit.
     """
     if isinstance(value, bool):
         return default
-    if isinstance(value, (int, float)):
-        return float(value)
-    if isinstance(value, str):
+    if isinstance(value, (int, float, str)):
         try:
-            return float(value)
-        except ValueError:
-            logger.warning("Browserless: invalid timeout %r in config; using %ss", value, default)
+            timeout = float(value)
+        except (TypeError, ValueError, OverflowError):
+            timeout = 0.0
+        if math.isfinite(timeout) and timeout > 0:
+            return timeout
+    logger.warning("Browserless: invalid timeout %r in config; using %ss", value, default)
     return default
 
 
diff --git a/backend/tests/test_browserless_client.py b/backend/tests/test_browserless_client.py
index a402ca2719..8ca535d14f 100644
--- a/backend/tests/test_browserless_client.py
+++ b/backend/tests/test_browserless_client.py
@@ -442,6 +442,47 @@ class TestBrowserlessTools:
         """YAML `timeout_s: off` parses as False; 0.0 would time out every request."""
         assert self._client_with_config({"timeout_s": False}).timeout_s == 30.0
 
+    @pytest.mark.parametrize(
+        "value",
+        [
+            0,
+            0.0,
+            -1,
+            -0.5,
+            float("inf"),
+            float("-inf"),
+            float("nan"),
+            "0",
+            "-1",
+            "inf",
+            "-inf",
+            "nan",
+            "1e400",
+            pytest.param(10**400, id="int-overflow"),
+        ],
+    )
+    async def test_unusable_timeout_falls_back_to_default(self, value):
+        """A value `httpx` cannot use as a deadline must never reach it.
+
+        Measured with httpx 0.28.1 against a backend that accepts the connection
+        and then never answers: `0` and the negatives fail every fetch with
+        `ConnectTimeout`, `nan` raises `ValueError: deadline must not be NaN`,
+        and `inf` leaves the request without a deadline so it never completes.
+        Either way a healthy Browserless backend looks broken.
+        """
+        assert self._client_with_config({"timeout_s": value}).timeout_s == 30.0
+
+    @pytest.mark.parametrize("value", [1, 0.001, 45, 12.5, "1", "45", "12.5", pytest.param(1e308, id="large-finite")])
+    async def test_usable_timeout_is_kept(self, value):
+        """Preserve positive finite timeout values without an upper limit."""
+        assert self._client_with_config({"timeout_s": value}).timeout_s == float(value)
+
+    async def test_unusable_timeout_warns(self, caplog):
+        """Swapping in the default silently would hide the operator's typo."""
+        with caplog.at_level("WARNING", logger="deerflow.community.browserless.tools"):
+            assert self._client_with_config({"timeout_s": 0}).timeout_s == 30.0
+        assert "invalid timeout 0" in caplog.text
+
     async def test_env_token_is_stripped(self):
         """A token read from a file keeps its trailing newline.
 
diff --git a/docs/guide/configuration.md b/docs/guide/configuration.md
index 12065235ed..ed9af164b2 100644
--- a/docs/guide/configuration.md
+++ b/docs/guide/configuration.md
@@ -149,6 +149,8 @@ fences, including unclosed fences, in complete and streamed replies.
 
    Direct Python calls to `web_fetch_tool` return `(content, artifact)` so Browserless can keep the target page's HTTP status out of page prose. Callers that previously expected a string should unpack the tuple; Agent tool calls continue to expose content and artifact through LangChain's tool result.
 
+   Browserless fetch and capture accept positive finite request timeouts in seconds. Invalid values fall back to 30 seconds. See [Browserless configuration](../../backend/docs/CONFIGURATION.md#delegated-fetch-backend-isolation).
+
    Self-hosted Browserless, Crawl4AI, Firecrawl, and fastCRW fetch backends require outbound isolation from private, loopback, link-local, and cloud-metadata networks before setting `network_isolation_confirmed: true` on each tool. DeerFlow rejects private or unverifiable backends by default because those services handle redirects, DNS resolution, and subresources themselves. `allow_private_addresses` permits intentional internal target URLs and does not bypass the backend check. See [delegated fetch deployment](../../backend/docs/CONFIGURATION.md#delegated-fetch-backend-isolation); `make doctor` reports backend configurations that would be refused.
 
    Jina fetches support opt-in bounded retries via `max_retries` (default `0`) and `retry_budget_seconds` (default `30`) in the tool configuration. Valid `Retry-After` hints set a minimum wait for HTTP 429/503; 429 without a valid hint stays terminal. Hints that cannot fit the remaining budget stop retries. Local backoff remains randomized. Retries may increase upstream requests and cost; see [Jina fetch retries](../../backend/docs/CONFIGURATION.md#jina-fetch-retries).