kongxin cee54cdd49
fix(web-fetch): stop sending .env.example placeholder keys to providers (#8206)
* fix(web-fetch): stop sending .env.example placeholder keys to providers

make config and make setup copy .env.example into .env, and the template
set nine search/fetch provider keys to example values such as
JINA_API_KEY=your-jina-api-key. The default keyless Jina Reader sent
that text as a Bearer key, so every web_fetch on a fresh install failed
with 401 while make doctor reported web fetch as configured.

- .env.example comments those nine lines out, like every other credential
  in the template already is.
- provider_credentials.resolve_api_key reads a your-...-key/-token/-secret
  value as unset, so existing .env files that still hold the placeholders
  recover without editing, for every provider that resolves keys there.

* chore: leave the changelog entry to merge time

Per CONTRIBUTING.md (#8202), maintainers add changelog entries when they
merge a pull request.

* fix(web-search): route every provider key and doctor through one placeholder rule

Review follow-up for #8206:

- provider_credentials exposes is_template_placeholder(), the single
  definition of an .env.example example value. key/token/secret must be
  a whole hyphen-separated word, so a real value such as your-monkey or
  your-passkey is no longer dropped.
- Serper, Serply, Sofya, SpecStream, Unbrowse, Webz, Brave, Tencent WSA
  and You.com resolve config and environment keys through
  resolve_api_key instead of private strip-and-send copies, so an old
  .env that still holds the template value no longer sends it.
- make doctor asks the same predicate: a placeholder key is reported as
  still holding the .env.example value instead of "set", and a literal
  placeholder in config.yaml falls back to the environment key as the
  runtime does.
- A census test keeps credential environment reads under community/
  wrapped in resolve_api_key.

* fix(web-search): match shipped placeholders exactly and mirror Browserless precedence in doctor

Second review follow-up for #8206:

- is_template_placeholder() matches exactly the your-... values that
  .env.example ships instead of a pattern, so a token an operator chose
  (a self-hosted Browserless TOKEN such as your-review-team-token) is
  sent again. A test keeps the set equal to the template.
- make doctor mirrors Browserless: a configured token field is
  authoritative, blank or not, and BROWSERLESS_TOKEN is read only when
  the field is omitted. A blank or placeholder token now warns instead
  of reporting the variable, cross-checked against the client.
- The keyless Jina and You.com providers stay ok but name JINA_API_KEY
  or YDC_API_KEY when a placeholder there is ignored, and a literal
  placeholder api_key in config.yaml is named when no env key backs it.

* fix(doctor): match each web provider's key precedence; drop the dead Jina api_key from docs

Found while re-checking #8206 against every provider's runtime:

- Browserless used as web_fetch had no doctor key check; it now gets the
  same authoritative token handling as web_capture.
- InfoQuest reads only INFOQUEST_API_KEY, so doctor no longer counts an
  api_key field in its tool config as configured.
- GroundRoute needs a key for every call but had no doctor entry, so a
  missing key reported ok.
- You.com names a placeholder api_key in config that it ignores.
- The Jina web_fetch docs snippet carried api_key: $JINA_API_KEY, which
  the tool never reads. With JINA_API_KEY now commented out in
  .env.example, that line made config loading fail on a fresh install.
- The provider_credentials docstring now scopes doctor's use of the
  predicate to web provider keys, and the scripts guide no longer says
  .env.example ships live placeholder keys.
2026-11-13 23:23:29 +08:00

155 lines
6.7 KiB
Python

"""Async adapter for the Webz.io News Search context endpoint."""
import asyncio
import json
import logging
import os
from datetime import UTC, date, datetime, timedelta
from typing import Literal
import httpx
from langchain.tools import tool
from deerflow.community.provider_credentials import resolve_api_key
from deerflow.community.search_time_range import SearchTimeRange
from deerflow.community.url_safety import safe_public_result_url
from deerflow.config import get_app_config
logger = logging.getLogger(__name__)
def _options() -> dict:
config = get_app_config().get_tool_config("web_search")
return dict(config.model_extra or {}) if config else {}
def _count(value: object) -> int:
try:
if isinstance(value, bool) or (isinstance(value, float) and not value.is_integer()):
raise ValueError("Expected an integer")
return max(1, min(100, int(value)))
except (TypeError, ValueError, OverflowError):
logger.warning("Invalid Webz max_results=%r; using default 5", value)
return 5
def _normalize(payload: object, count: int) -> list[dict]:
if not isinstance(payload, dict) or not isinstance(payload.get("results"), list):
raise ValueError("Invalid results")
results = []
for index, item in enumerate(payload["results"][:count]):
try:
if not isinstance(item, dict) or not isinstance(item.get("article"), dict):
raise ValueError("Invalid article")
article = item["article"]
chunk, metadata = item.get("chunk", {}), item.get("metadata", {})
if not isinstance(chunk, dict) or not isinstance(metadata, dict):
raise ValueError("Invalid metadata")
result = {
"title": article.get("title", ""),
"url": article.get("url", ""),
"content": chunk.get("text") or article.get("summary") or "",
"published_at": article.get("published_at", ""),
}
if any(not isinstance(value, str) for value in result.values()):
raise ValueError("Invalid article fields")
result["url"] = safe_public_result_url(result["url"])
except ValueError:
logger.warning("Skipping malformed Webz result at index %d", index)
continue
results.append({**result, "source": metadata})
if payload["results"] and not results:
raise ValueError("No valid articles")
return results
@tool("web_search", parse_docstring=True)
async def web_search_tool(
query: str,
max_results: int | None = None,
time_range: SearchTimeRange | None = None,
published_from: str | None = None,
published_to: str | None = None,
language: list[str] | None = None,
country: list[str] | None = None,
source: list[str] | None = None,
sentiment: list[Literal["positive", "negative", "neutral"]] | None = None,
category: list[str] | None = None,
) -> str:
"""Search recent news with Webz.io and return matching passages and source metadata.
This provider searches news, not the general web. Date filters do not extend
the provider's available news coverage.
Args:
query: News search query, at most 750 characters and 100 words.
max_results: Maximum results, clamped to 1-100; uses configuration or 5 when omitted.
time_range: Optional recency window: day, week, month, or year (UTC).
published_from: Inclusive lower publication date in YYYY-MM-DD; overrides time_range.
published_to: Inclusive upper publication date in YYYY-MM-DD.
language: Optional language names, for example ["english"].
country: Optional country codes, for example ["US"].
source: Optional source domains, for example ["cnn.com"].
sentiment: Optional positive, negative, or neutral sentiments.
category: Optional news categories.
Returns:
JSON containing query, returned_results (the returned page size) and results, or an error.
"""
query = query.strip()
if not query or len(query) > 750 or len(query.split()) > 100:
return json.dumps({"error": "Webz query must contain 1-750 characters and at most 100 words"})
if published_from is None and time_range:
days = {"day": 1, "week": 7, "month": 30, "year": 365}[time_range]
published_from = (datetime.now(UTC).date() - timedelta(days=days)).isoformat()
try:
for value in (published_from, published_to):
if value is not None and date.fromisoformat(value).isoformat() != value:
raise ValueError("Non-canonical date")
if published_from and published_to and published_from > published_to:
raise ValueError("Reversed dates")
except ValueError:
return json.dumps({"error": "Webz dates must use YYYY-MM-DD with published_from <= published_to"})
# Lazy configuration resolution can read files; keep it off the event loop.
options = await asyncio.to_thread(_options)
key = resolve_api_key(options.get("api_key")) or resolve_api_key(os.getenv("WEBZ_API_KEY"))
if not key:
return json.dumps({"error": "Webz API key is missing; configure api_key or WEBZ_API_KEY"})
count = _count(max_results if max_results is not None else options.get("max_results", 5))
body = {"query": query, "k": count}
filters = {
name: value
for name, value in {
"published_from": published_from,
"published_to": published_to,
"language": language,
"country": country,
"domain": source,
"sentiment": sentiment,
"category": category,
}.items()
if value is not None
}
if filters:
body["filters"] = filters
try:
async with await asyncio.to_thread(httpx.AsyncClient, timeout=30.0, follow_redirects=False) as client:
response = await client.post(
"https://api.webz.io/api/news/context",
headers={"Authorization": f"Bearer {key}", "Accept": "application/json"},
json=body,
)
response.raise_for_status()
results = _normalize(response.json(), count)
except httpx.HTTPStatusError as exc:
return json.dumps({"error": f"Webz request failed (HTTP {exc.response.status_code})"})
except httpx.TimeoutException:
return json.dumps({"error": "Webz request timed out"})
except httpx.RequestError:
return json.dumps({"error": "Webz request failed"})
except ValueError:
logger.warning("Webz returned an unexpected response")
return json.dumps({"error": "Webz returned an unexpected response"})
return json.dumps({"query": query, "returned_results": len(results), "results": results}, ensure_ascii=False)