From e25cf07b76f4f5f74d5f8d10548425912499c0b6 Mon Sep 17 00:00:00 2001
From: Bubu <43925055+p3psi-boo@users.noreply.github.com>
Date: Wed, 18 Dec 2024 11:21:40 +0000
Subject: [PATCH] [feat] engines: add baidu (general)

---
 searx/engines/baidu.py | 64 ++++++++++++++++++++++++++++++++++++++++++
 searx/settings.yml     |  5 ++++
 2 files changed, 69 insertions(+)
 create mode 100644 searx/engines/baidu.py

diff --git a/searx/engines/baidu.py b/searx/engines/baidu.py
new file mode 100644
index 000000000..b0d0b5353
--- /dev/null
+++ b/searx/engines/baidu.py
@@ -0,0 +1,64 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Baidu
+
+.. Website: https://www.baidu.com
+"""
+
+from urllib.parse import urlencode
+from datetime import datetime
+
+from searx.exceptions import SearxEngineAPIException
+
+about = {
+    "website": "https://www.baidu.com",
+    "wikidata_id": "Q14772",
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+paging = True
+results_per_page = 10
+categories = ["general"]
+
+base_url = "https://www.baidu.com/s"
+
+
+def request(query, params):
+    keyword = query.strip()
+
+    query_params = {"wd": keyword, "rn": 20, "pn": params["pageno"], "tn": "json"}
+
+    params["url"] = f"{base_url}?{urlencode(query_params)}"
+    return params
+
+
+def response(resp):
+    data = resp.json()
+    results = []
+
+    entries = data.get("feed", {}).get("entry")
+    if not entries:
+        raise SearxEngineAPIException("Invalid response")
+    for entry in entries:
+        if not entry.get("title") or not entry.get("url"):
+            continue
+
+        published_date = None
+        if entry.get("time"):
+            try:
+                published_date = datetime.fromtimestamp(entry["time"])
+            except (ValueError, TypeError):
+                published_date = None
+
+        results.append(
+            {
+                "title": entry["title"],
+                "url": entry["url"],
+                "content": entry.get("abs", ""),
+                "publishedDate": published_date,
+            }
+        )
+
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
index c29994227..5bc6f5f93 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -326,6 +326,11 @@ categories_as_tabs:
   social media:
 
 engines:
+  - name: baidu
+    engine: baidu
+    shortcut: baidu
+    disabled: true
+
   - name: 9gag
     engine: 9gag
     shortcut: 9g