mirror of
https://github.com/bytedance/deer-flow.git
synced 2026-11-16 08:20:34 +00:00
* 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.
155 lines
6.7 KiB
Python
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)
|