[fix] searchzee: bypass botblocking

requires https://github.com/searxng/searxng/pull/6620
This commit is contained in:
Bnyro
2026-09-02 13:28:52 +02:00
parent 242dc6e398
commit ccffbfc164

View File

@@ -4,13 +4,11 @@ independent search infrastructure."""
import typing as t
from urllib.parse import urlencode
import uuid
from searx.exceptions import SearxEngineAPIException
from searx.extended_types import SXNG_Response
from searx.network import get
from searx.result_types import EngineResults
from searx.utils import extr, html_to_text
from searx.enginelib import EngineCache
from searx.utils import html_to_text
if t.TYPE_CHECKING:
from searx.search.processors import OnlineParams
@@ -34,43 +32,19 @@ SearchzeeCategType = t.Literal["web", "news"]
searchzee_categ: SearchzeeCategType = None # type: ignore[reportAssignmentType]
CACHE: EngineCache
"""Cache for storing the scraped API Token."""
base_url = "https://searchzee.com"
# only supports for news
time_range_map = {"day": "pd", "week": "pw", "month": "pm", "year": "py"}
def setup(engine_settings: dict[str, t.Any]) -> bool:
def setup(_: dict[str, t.Any]):
if searchzee_categ not in t.get_args(SearchzeeCategType):
raise ValueError("invalid category: %s" % searchzee_categ)
global CACHE # pylint: disable=global-statement
CACHE = EngineCache(engine_settings["name"]) # type: ignore[reportAny]
return True
def _obtain_api_token() -> str:
token: str | None = CACHE.get("token") # type: ignore[reportAny]
if token:
return token
token_resp = get(
f"{base_url}/app.js",
)
if not token_resp.ok:
raise SearxEngineAPIException("failed to obtain api key")
token = extr(token_resp.text, "const SEARCHZEE_API_TOKEN = \"", "\";")
CACHE.set("token", token, expire=3600)
return token
def request(query: str, params: "OnlineParams"):
params["headers"]["X-SearchZee-Token"] = _obtain_api_token()
params["cookies"]["szs"] = str(uuid.uuid4())
args = {"q": query, "type": searchzee_categ, "offset": params["pageno"] - 1}
if params["time_range"]: