[feat] add 360search engine for searxng

Co-authored-by: Bnyro <bnyro@tutanota.com>

docs/admin/settings/settings_search.rst

@@ -33,6 +33,7 @@
 ``autocomplete``:
   Existing autocomplete backends, leave blank to turn it off.
 
+  - ``360search``
   - ``baidu``
   - ``brave``
   - ``dbpedia``

searx/autocomplete.py

@@ -148,6 +148,21 @@ def mwmbl(query, _lang):
     return [result for result in results if not result.startswith("go: ") and not result.startswith("search: ")]
 
 
+def qihu360search(query, _lang):
+    # 360Search search autocompleter
+    url = f"https://sug.so.360.cn/suggest?{urlencode({'format': 'json', 'word': query})}"
+    response = get(url)
+
+    results = []
+
+    if response.ok:
+        data = response.json()
+        if 'result' in data:
+            for item in data['result']:
+                results.append(item['word'])
+    return results
+
+
 def seznam(query, _lang):
     # seznam search autocompleter
     url = 'https://suggest.seznam.cz/fulltext/cs?{query}'
@@ -246,6 +261,7 @@ def yandex(query, _lang):
 
 
 backends = {
+    '360search': qihu360search,
     'baidu': baidu,
     'brave': brave,
     'dbpedia': dbpedia,

searx/engines/360search.py

@@ -0,0 +1,67 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# pylint: disable=invalid-name
+"""360Search search engine for searxng"""
+
+from urllib.parse import urlencode
+from lxml import html
+
+from searx.utils import extract_text
+
+# Metadata
+about = {
+    "website": "https://www.so.com/",
+    "wikidata_id": "Q10846064",
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "HTML",
+}
+
+# Engine Configuration
+categories = ["general"]
+paging = True
+time_range_support = True
+
+time_range_dict = {'day': 'd', 'week': 'w', 'month': 'm', 'year': 'y'}
+
+# Base URL
+base_url = "https://www.so.com"
+
+
+def request(query, params):
+    query_params = {
+        "pn": params["pageno"],
+        "q": query,
+    }
+
+    if time_range_dict.get(params['time_range']):
+        query_params["adv_t"] = time_range_dict.get(params['time_range'])
+
+    params["url"] = f"{base_url}/s?{urlencode(query_params)}"
+    return params
+
+
+def response(resp):
+    dom = html.fromstring(resp.text)
+    results = []
+
+    for item in dom.xpath('//li[contains(@class, "res-list")]'):
+        title = extract_text(item.xpath('.//h3[contains(@class, "res-title")]/a'))
+
+        url = extract_text(item.xpath('.//h3[contains(@class, "res-title")]/a/@data-mdurl'))
+        if not url:
+            url = extract_text(item.xpath('.//h3[contains(@class, "res-title")]/a/@href'))
+
+        content = extract_text(item.xpath('.//p[@class="res-desc"]'))
+        if not content:
+            content = extract_text(item.xpath('.//span[@class="res-list-summary"]'))
+
+        if title and url:
+            results.append(
+                {
+                    "title": title,
+                    "url": url,
+                    "content": content,
+                }
+            )
+
+    return results

searx/engines/360search_videos.py

@@ -0,0 +1,64 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# pylint: disable=invalid-name
+"""360Search-Videos: A search engine for retrieving videos from 360Search."""
+
+from urllib.parse import urlencode
+from datetime import datetime
+
+from searx.exceptions import SearxEngineAPIException
+from searx.utils import html_to_text
+
+about = {
+    "website": "https://tv.360kan.com/",
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+paging = True
+results_per_page = 10
+categories = ["videos"]
+
+base_url = "https://tv.360kan.com"
+
+
+def request(query, params):
+    query_params = {"count": 10, "q": query, "start": params["pageno"] * 10}
+
+    params["url"] = f"{base_url}/v1/video/list?{urlencode(query_params)}"
+    return params
+
+
+def response(resp):
+    try:
+        data = resp.json()
+    except Exception as e:
+        raise SearxEngineAPIException(f"Invalid response: {e}") from e
+    results = []
+
+    if "data" not in data or "result" not in data["data"]:
+        raise SearxEngineAPIException("Invalid response")
+
+    for entry in data["data"]["result"]:
+        if not entry.get("title") or not entry.get("play_url"):
+            continue
+
+        published_date = None
+        if entry.get("publish_time"):
+            try:
+                published_date = datetime.fromtimestamp(int(entry["publish_time"]))
+            except (ValueError, TypeError):
+                published_date = None
+
+        results.append(
+            {
+                'url': entry["play_url"],
+                'title': html_to_text(entry["title"]),
+                'content': html_to_text(entry["description"]),
+                'template': 'videos.html',
+                'publishedDate': published_date,
+                'thumbnail': entry["cover_img"],
+            }
+        )
+
+    return results

searx/settings.yml

@@ -33,7 +33,7 @@ brand:
 search:
   # Filter results. 0: None, 1: Moderate, 2: Strict
   safe_search: 0
-  # Existing autocomplete backends: "baidu", "brave", "dbpedia", "duckduckgo", "google", "yandex",
+  # Existing autocomplete backends: "360search", "baidu", "brave", "dbpedia", "duckduckgo", "google", "yandex",
   # "mwmbl", "seznam", "stract", "swisscows", "qwant", "wikipedia" -
   # leave blank to turn it off by default.
   autocomplete: ""
@@ -337,6 +337,16 @@ categories_as_tabs:
   social media:
 
 engines:
+  - name: 360search
+    engine: 360search
+    shortcut: 360so
+    disabled: true
+
+  - name: 360search videos
+    engine: 360search_videos
+    shortcut: 360sov
+    disabled: true
+
   - name: 9gag
     engine: 9gag
     shortcut: 9g