From 031747f29ed06c1a67d75538662eae72f199e35c Mon Sep 17 00:00:00 2001
From: Bnyro <bnyro@tutanota.com>
Date: Sat, 13 Jun 2026 13:52:01 +0200
Subject: [PATCH] [feat] engines: add chatnoir general engine (#6183)

Chatnoir is an open source search engine developed by universities, based on
CommonCrawl (and others).  It's uncommented by default - we don't want to
overload the universities with bot traffic that targets SearXNG (sad truth why
we can't have nice things anymore)
---
 searx/engines/chatnoir.py | 134 ++++++++++++++++++++++++++++++++++++++
 searx/settings.yml        |   6 ++
 2 files changed, 140 insertions(+)
 create mode 100644 searx/engines/chatnoir.py

diff --git a/searx/engines/chatnoir.py b/searx/engines/chatnoir.py
new file mode 100644
index 000000000..838bdb725
--- /dev/null
+++ b/searx/engines/chatnoir.py
@@ -0,0 +1,134 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Chatnoir is an open source search engine developed by Webis, a network of
+researchers from the universities of Weimar, Halle and Leipzig. It supports
+different different text corpora as indexes, e.g. CommonCrawl. See its
+`announcement`_ for more information.
+
+.. _announcement : https://groups.google.com/g/common-crawl/c/3o2dOHpeRxo/m/H2Osqz9dAAAJ
+"""
+
+import typing as t
+
+from searx.exceptions import SearxEngineAPIException
+from searx.extended_types import SXNG_Response
+from searx.network import get, post
+from searx.result_types import EngineResults
+from searx.utils import html_to_text
+from searx.enginelib import EngineCache
+
+if t.TYPE_CHECKING:
+    from searx.search.processors import OnlineParams
+
+about = {
+    "website": "https://www.chatnoir.eu",
+    "official_api_documentation": "https://www.chatnoir.eu/docs/api-general",
+    "use_official_api": True,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+base_url = "https://www.chatnoir.eu"
+categories = ["general"]
+
+paging = True
+page_size = 10
+
+api_key = ""
+"""You can optionally provide your own API key here. This one will then be used
+instead of scraping an API key."""
+
+search_index = "cw22"
+"""Search index to browse in. See `the API documentation
+<https://www.chatnoir.eu/docs/api-general>`_ for a full list."""
+
+
+CACHE: EngineCache
+"""Cache to store session info (i.e. api key, csrf token, session id)."""
+
+
+def setup(engine_settings: dict[str, t.Any]) -> bool:
+    global CACHE  # pylint: disable=global-statement
+    CACHE = EngineCache(engine_settings["name"])
+    return True
+
+
+def _obtain_api_key() -> tuple[str, str, str]:
+    cached_session = CACHE.get("session")
+    if cached_session:
+        return tuple(cached_session.split("|"))
+
+    home_resp = get(base_url)
+    if not home_resp.ok:
+        raise SearxEngineAPIException("failed to obtain api key")
+    csrf_token = home_resp.cookies["csrftoken"]
+
+    token_resp = post(
+        "https://www.chatnoir.eu/?init",
+        headers={
+            "Referer": f"{base_url}/",
+            "X-Requested-With": "XMLHttpRequest",
+            "X-Csrf-Token": csrf_token,
+        },
+        cookies=home_resp.cookies,
+    )
+    if not token_resp.ok:
+        raise SearxEngineAPIException("failed to obtain api key")
+    session_id = token_resp.cookies["sessionid"]
+    scraped_api_key = token_resp.json()["token"]["token"]
+
+    # session keys seem to become rate-limited very fast, so only remembering
+    # for 1 minute here
+    CACHE.set("session", f"{csrf_token}|{session_id}|{scraped_api_key}", expire=60)
+
+    return csrf_token, session_id, scraped_api_key
+
+
+def request(query: str, params: "OnlineParams"):
+    if api_key:
+        # use user-provided API key instead of scraping one
+        headers = {
+            "Authorization": f"Bearer {api_key}",
+        }
+
+        params["headers"].update(headers)
+    else:
+        csrf_token, session_id, scraped_api_key = _obtain_api_key()
+
+        headers = {
+            "Authorization": f"Bearer {scraped_api_key}",
+            "X-Csrf-Token": csrf_token,
+        }
+
+        params["headers"].update(headers)
+        params["cookies"] = {"csrftoken": session_id, "sessionid": session_id}
+
+    params["url"] = f"{base_url}/api/v1/_search"
+    params["method"] = "POST"
+
+    json_data = {
+        "query": query,
+        "index": [
+            search_index,
+        ],
+        "from": (params["pageno"] - 1) * page_size,
+        "size": page_size,
+        "_extended_meta": True,
+    }
+    params["json"] = json_data
+
+
+def response(resp: "SXNG_Response") -> EngineResults:
+    res = EngineResults()
+
+    results = resp.json()["results"]
+
+    for result in results:
+        res.add(
+            res.types.MainResult(
+                url=result["target_uri"],
+                title=html_to_text(result["title"]),
+                content=html_to_text(result["snippet"]),
+            )
+        )
+
+    return res
diff --git a/searx/settings.yml b/searx/settings.yml
index 97573f241..9e2fd9804 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -609,6 +609,12 @@ engines:
     shortcut: ca
     disabled: true
 
+  # - name: chatnoir
+  #   engine: chatnoir
+  #   shortcut: cha
+  #   search_index: cw22
+  #   disabled: true
+
   - name: chefkoch
     engine: chefkoch
     shortcut: chef