[feat] engines: add support for iseek.com (general)

FindFiles.net is a specialized file search engine designed to help you search
files online with precision. Unlike traditional search engines that mainly index
web pages, FindFiles focuses on finding real files on the internet - including
PDFs, documents, archives, videos, datasets, and more. [1]

[1] https://findfiles.net

client/simple/package-lock.json

@@ -20,7 +20,7 @@
         "browserslist": "^4.28.2",
         "browserslist-to-esbuild": "^2.1.1",
         "edge.js": "^6.5.1",
-        "less": "^4.6.4",
+        "less": "^4.6.6",
         "mathjs": "^15.2.0",
         "sharp": "~0.35.1",
         "sort-package-json": "^4.0.0",
@@ -2890,9 +2890,9 @@
       "license": "Apache-2.0"
     },
     "node_modules/less": {
-      "version": "4.6.4",
-      "resolved": "https://registry.npmjs.org/less/-/less-4.6.4.tgz",
-      "integrity": "sha512-OJmO5+HxZLLw0RLzkqaNHzcgEAQG7C0y3aMbwtCzIUFZsLMNNq/1IdAdHEycQ58CwUO3jPTHmoN+tE5I7FQxNg==",
+      "version": "4.6.6",
+      "resolved": "https://registry.npmjs.org/less/-/less-4.6.6.tgz",
+      "integrity": "sha512-ooPSwQGQ2sVe8Dh1jVsbKKsRR2gd8lFK72BDkeSzjnD1T5aIHL65hCMfO0GVmtriKgDKrQv6xp9UrihUsWuAzA==",
       "dev": true,
       "license": "Apache-2.0",
       "dependencies": {
@@ -2909,7 +2909,7 @@
         "errno": "^0.1.1",
         "graceful-fs": "^4.1.2",
         "image-size": "~0.5.0",
-        "make-dir": "^2.1.0",
+        "make-dir": "^5.1.0",
         "mime": "^1.4.1",
         "needle": "^3.1.0",
         "source-map": "~0.6.0"
@@ -3191,18 +3191,17 @@
       "license": "MIT"
     },
     "node_modules/make-dir": {
-      "version": "2.1.0",
-      "resolved": "https://registry.npmjs.org/make-dir/-/make-dir-2.1.0.tgz",
-      "integrity": "sha512-LS9X+dc8KLxXCb8dni79fLIIUA5VyZoyjSMCwTluaXA0o27cCK0bhXkpgw+sTXVpPy/lSO57ilRixqk0vDmtRA==",
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/make-dir/-/make-dir-5.1.0.tgz",
+      "integrity": "sha512-IfpFq6UM39dUNiphpA6uDezNx/AvWyhwfICWPR3t1VspkgkMZrL+Rk1RbN1bx+aeNYwOrqGJgEgV3yotk+ZUVw==",
       "dev": true,
       "license": "MIT",
       "optional": true,
-      "dependencies": {
-        "pify": "^4.0.1",
-        "semver": "^5.6.0"
-      },
       "engines": {
-        "node": ">=6"
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
       }
     },
     "node_modules/mathjs": {
@@ -3491,17 +3490,6 @@
         "url": "https://github.com/sponsors/jonschlinkert"
       }
     },
-    "node_modules/pify": {
-      "version": "4.0.1",
-      "resolved": "https://registry.npmjs.org/pify/-/pify-4.0.1.tgz",
-      "integrity": "sha512-uB80kBFb/tfd68bVleG9T5GGsGPjJrLAUpR5PZIrhBnIaRTQRjqdJSsIKkOP6OAIFbj7GOrcudc5pNjZ+geV2g==",
-      "dev": true,
-      "license": "MIT",
-      "optional": true,
-      "engines": {
-        "node": ">=6"
-      }
-    },
     "node_modules/pluralize": {
       "version": "8.0.0",
       "resolved": "https://registry.npmjs.org/pluralize/-/pluralize-8.0.0.tgz",
@@ -3861,17 +3849,6 @@
       "dev": true,
       "license": "MIT"
     },
-    "node_modules/semver": {
-      "version": "5.7.2",
-      "resolved": "https://registry.npmjs.org/semver/-/semver-5.7.2.tgz",
-      "integrity": "sha512-cBznnQ9KjJqU67B52RMC65CMarK2600WFnbkcaiwWq3xy/5haFJlshgnpjovMVJ+Hff49d8GEn0b87C5pDQ10g==",
-      "dev": true,
-      "license": "ISC",
-      "optional": true,
-      "bin": {
-        "semver": "bin/semver"
-      }
-    },
     "node_modules/sharp": {
       "version": "0.35.1",
       "resolved": "https://registry.npmjs.org/sharp/-/sharp-0.35.1.tgz",

client/simple/package.json

@@ -34,7 +34,7 @@
     "browserslist": "^4.28.2",
     "browserslist-to-esbuild": "^2.1.1",
     "edge.js": "^6.5.1",
-    "less": "^4.6.4",
+    "less": "^4.6.6",
     "mathjs": "^15.2.0",
     "sharp": "~0.35.1",
     "sort-package-json": "^4.0.0",

docs/dev/engines/online/aol.rst

@@ -1,8 +0,0 @@
-.. _aol engine:
-
-===
-AOL
-===
-
-.. automodule:: searx.engines.aol
-   :members:

requirements-dev.txt

@@ -2,16 +2,16 @@ mock==5.2.0
 nose2[coverage_plugin]==0.16.0
 cov-core==1.15.0
 black==25.9.0
-pylint==4.0.5
+pylint==4.0.6
 splinter==0.21.0
-selenium==4.44.0
+selenium==4.45.0
 Sphinx==8.2.3;python_version <= "3.11"
 Sphinx==9.1.0; python_version > "3.11"
 sphinx-issues==6.0.0
 sphinx-jinja==2.0.2
 sphinx-tabs==3.5.0
 furo==2025.12.19
-sphinxcontrib-programoutput==0.19
+sphinxcontrib-programoutput==0.20
 sphinx-autobuild==2025.8.25
 sphinx-notfound-page==1.1.0
 myst-parser==5.0.0
@@ -24,5 +24,5 @@ docutils>=0.21.2;python_version <= "3.11"
 docutils>=0.22.4; python_version > "3.11"
 parameterized==0.9.0
 granian[reload]==2.7.6
-basedpyright==1.39.7
+basedpyright==1.39.8
 types-lxml==2026.2.16

requirements.txt

@@ -1,4 +1,4 @@
-certifi==2026.5.20
+certifi==2026.6.17
 babel==2.18.0
 flask-babel==4.0.0
 flask==3.1.3

searx/engines/aol.py

@@ -1,210 +0,0 @@
-# SPDX-License-Identifier: AGPL-3.0-or-later
-"""AOL supports WEB, image, and video search.  Internally, it uses the Bing
-index.
-
-AOL doesn't seem to support setting the language via request parameters, instead
-the results are based on the URL.  For example, there is
-
-- `search.aol.com <https://search.aol.com>`_ for English results
-- `suche.aol.de <https://suche.aol.de>`_ for German results
-
-However, AOL offers its services only in a few regions:
-
-- en-US: search.aol.com
-- de-DE: suche.aol.de
-- fr-FR: recherche.aol.fr
-- en-GB: search.aol.co.uk
-- en-CA: search.aol.ca
-
-In order to still offer sufficient support for language and region, the `search
-keywords`_ known from Bing, ``language`` and ``loc`` (region), are added to the
-search term (AOL is basically just a proxy for Bing).
-
-.. _search keywords:
-    https://support.microsoft.com/en-us/topic/advanced-search-keywords-ea595928-5d63-4a0b-9c6b-0b769865e78a
-
-"""
-
-from urllib.parse import urlencode, unquote_plus
-import typing as t
-
-from lxml import html
-from dateutil import parser
-
-from searx.result_types import EngineResults
-from searx.utils import eval_xpath_list, eval_xpath, extract_text
-
-if t.TYPE_CHECKING:
-    from searx.extended_types import SXNG_Response
-    from searx.search.processors import OnlineParams
-
-about = {
-    "website": "https://www.aol.com",
-    "wikidata_id": "Q27585",
-    "official_api_documentation": None,
-    "use_official_api": False,
-    "require_api_key": False,
-    "results": "HTML",
-}
-
-categories = ["general"]
-search_type = "search"  # supported: search, image, video
-
-paging = True
-safesearch = True
-time_range_support = True
-results_per_page = 10
-
-
-base_url = "https://search.aol.com"
-time_range_map = {"day": "1d", "week": "1w", "month": "1m", "year": "1y"}
-safesearch_map = {0: "p", 1: "r", 2: "i"}
-
-enable_http2 = False
-
-
-def init(_):
-    if search_type not in ("search", "image", "video"):
-        raise ValueError(f"unsupported search type {search_type}")
-
-
-def request(query: str, params: "OnlineParams") -> None:
-
-    language, region = (params["searxng_locale"].split("-") + [None])[:2]
-    if language and language != "all":
-        query = f"{query} language:{language}"
-    if region:
-        query = f"{query} loc:{region}"
-
-    args: dict[str, str | int | None] = {
-        "q": query,
-        "b": params["pageno"] * results_per_page + 1,  # page is 1-indexed
-        "pz": results_per_page,
-    }
-
-    if params["time_range"]:
-        args["fr2"] = "time"
-        args["age"] = params["time_range"]
-    else:
-        args["fr2"] = "sb-top-search"
-
-    params["cookies"]["sB"] = f"vm={safesearch_map[params['safesearch']]}"
-    params["url"] = f"{base_url}/aol/{search_type}?{urlencode(args)}"
-    logger.debug(params)
-
-
-def _deobfuscate_url(obfuscated_url: str) -> str | None:
-    # URL looks like "https://search.aol.com/click/_ylt=AwjFSDjd;_ylu=JfsdjDFd/RV=2/RE=1774058166/RO=10/RU=https%3a%2f%2fen.wikipedia.org%2fwiki%2fTree/RK=0/RS=BP2CqeMLjscg4n8cTmuddlEQA2I-"  # pylint: disable=line-too-long
-    if not obfuscated_url:
-        return None
-
-    for part in obfuscated_url.split("/"):
-        if part.startswith("RU="):
-            return unquote_plus(part[3:])
-    # pattern for de-obfuscating URL not found, fall back to Yahoo's tracking link
-    return obfuscated_url
-
-
-def _general_results(doc: html.HtmlElement) -> EngineResults:
-    res = EngineResults()
-
-    for result in eval_xpath_list(doc, "//div[@id='web']//ol/li[not(contains(@class, 'first'))]"):
-        obfuscated_url = extract_text(eval_xpath(result, ".//h3/a/@href"))
-        if not obfuscated_url:
-            continue
-
-        url = _deobfuscate_url(obfuscated_url)
-        if not url:
-            continue
-
-        res.add(
-            res.types.MainResult(
-                url=url,
-                title=extract_text(eval_xpath(result, ".//h3/a")) or "",
-                content=extract_text(eval_xpath(result, ".//div[contains(@class, 'compText')]")) or "",
-                thumbnail=extract_text(eval_xpath(result, ".//a[contains(@class, 'thm')]/img/@data-src")) or "",
-            )
-        )
-    return res
-
-
-def _video_results(doc: html.HtmlElement) -> EngineResults:
-    res = EngineResults()
-
-    for result in eval_xpath_list(doc, "//div[contains(@class, 'results')]//ol/li"):
-        obfuscated_url = extract_text(eval_xpath(result, ".//a/@href"))
-        if not obfuscated_url:
-            continue
-
-        url = _deobfuscate_url(obfuscated_url)
-        if not url:
-            continue
-
-        published_date_raw = extract_text(eval_xpath(result, ".//div[contains(@class, 'v-age')]"))
-        try:
-            published_date = parser.parse(published_date_raw or "")
-        except parser.ParserError:
-            published_date = None
-
-        res.add(
-            res.types.LegacyResult(
-                {
-                    "template": "videos.html",
-                    "url": url,
-                    "title": extract_text(eval_xpath(result, ".//h3")),
-                    "content": extract_text(eval_xpath(result, ".//div[contains(@class, 'compText')]")),
-                    "thumbnail": extract_text(eval_xpath(result, ".//img[contains(@class, 'thm')]/@src")),
-                    "length": extract_text(eval_xpath(result, ".//span[contains(@class, 'v-time')]")),
-                    "publishedDate": published_date,
-                }
-            )
-        )
-
-    return res
-
-
-def _image_results(doc: html.HtmlElement) -> EngineResults:
-    res = EngineResults()
-
-    for result in eval_xpath_list(doc, "//section[@id='results']//ul/li"):
-        obfuscated_url = extract_text(eval_xpath(result, "./a/@href"))
-        if not obfuscated_url:
-            continue
-
-        url = _deobfuscate_url(obfuscated_url)
-        if not url:
-            continue
-
-        res.add(
-            res.types.LegacyResult(
-                {
-                    "template": "images.html",
-                    # results don't have an extra URL, only the image source
-                    "url": url,
-                    "title": extract_text(eval_xpath(result, ".//a/@aria-label")),
-                    "thumbnail_src": extract_text(eval_xpath(result, ".//img/@src")),
-                    "img_src": url,
-                }
-            )
-        )
-
-    return res
-
-
-def response(resp: "SXNG_Response") -> EngineResults:
-    doc = html.fromstring(resp.text)
-
-    match search_type:
-        case "search":
-            results = _general_results(doc)
-        case "image":
-            results = _image_results(doc)
-        case "video":
-            results = _video_results(doc)
-        case _:
-            raise ValueError("unsupported search type")
-
-    for suggestion in eval_xpath_list(doc, ".//ol[contains(@class, 'searchRightBottom')]//table//a"):
-        results.add(results.types.LegacyResult({"suggestion": extract_text(suggestion)}))
-
-    return results

searx/engines/findfiles.py

@@ -0,0 +1,118 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""FindFiles.net_ is a Germany-based file search engine.
+
+FindFiles.net_ is a specialized file search engine designed to help you search
+files online with precision. Unlike traditional search engines that mainly index
+web pages, FindFiles focuses on finding real files on the internet - including
+PDFs, documents, archives, videos, datasets, and more.
+
+.. _FindFiles.net: https://findfiles.net
+"""
+
+from os.path import basename
+from urllib.parse import urlencode
+import typing as t
+
+from lxml import html
+
+from searx.result_types import EngineResults
+from searx.utils import extract_text, eval_xpath, eval_xpath_list
+
+if t.TYPE_CHECKING:
+    from extended_types import SXNG_Response
+    from search.processors import OnlineParams
+
+about = {
+    "website": "https://findfiles.net",
+    "wikidata_id": None,
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "HTML",
+}
+
+base_url = "https://findfiles.net"
+categories = ["files"]
+paging = True
+safeserach = True
+
+safesearch_map = {
+    0: "contentguard.off",
+    1: "contentguard.moderate",
+    2: "contentguard.strict",
+}
+
+FindFilesCategory = t.Literal[
+    "all",
+    "document",
+    "text",
+    "image",
+    "audio",
+    "video",
+]
+FINDFILES_CATEGORIES = t.get_args(FindFilesCategory)
+
+findfiles_categ: FindFilesCategory = "all"
+"""Category to search in."""
+
+
+def setup(_: dict[str, t.Any]) -> bool:
+    if findfiles_categ not in FINDFILES_CATEGORIES:
+        raise ValueError("invalid category: %s" % findfiles_categ)
+    return True
+
+
+def request(query: str, params: "OnlineParams") -> None:
+    args = {
+        "query": query,
+        "contentguard": safesearch_map[params["safesearch"]],
+        "page": params["pageno"],
+    }
+    # the language in the path doesn't change anything about the results, it
+    # only changes the UI
+    params["url"] = f"{base_url}/en/serp/{findfiles_categ}/?{urlencode(args)}"
+
+
+def response(resp: "SXNG_Response") -> EngineResults:
+    res = EngineResults()
+
+    dom = html.fromstring(resp.text)
+    if findfiles_categ == "image":
+        for result in eval_xpath_list(
+            dom, "//div[contains(@class, 'image-mosaic')]/div[contains(@class, 'image-item')]"
+        ):
+            res.add(
+                res.types.Image(
+                    url=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/a/@href")) or "",
+                    title=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/a")) or "",
+                    thumbnail_src=extract_text(eval_xpath(result, ".//img/@src")) or "",
+                )
+            )
+    elif findfiles_categ == "video":
+        for result in eval_xpath_list(
+            dom, "//div[contains(@class, 'video-mosaic')]/div[contains(@class, 'video-item')]"
+        ):
+            video_src = extract_text(eval_xpath(result, ".//video/@src")) or ""
+            res.add(
+                res.types.LegacyResult(
+                    template="videos.html",
+                    url=video_src,
+                    title=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/span")) or "",
+                    iframe_src=video_src or "",
+                )
+            )
+    else:
+        for result in eval_xpath_list(dom, "//ol/li[contains(@class, 'result-item')]/article"):
+            filename = basename(extract_text(eval_xpath(result, ".//h3")) or "")
+            res.add(
+                res.types.File(
+                    url=extract_text(eval_xpath(result, ".//h3/a/@href")) or "",
+                    title=filename,
+                    content=" ".join(extract_text(el) or "" for el in eval_xpath_list(result, "./div/span")),
+                    filename=filename,
+                    size=extract_text(eval_xpath(result, "(.//span[@id])[1]")) or "",
+                    embedded=extract_text(eval_xpath(result, ".//audio/@src")) or "",
+                )
+            )
+
+    return res

searx/engines/giphy.py

@@ -0,0 +1,127 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Giphy (images)"""
+
+import random
+from urllib.parse import urlencode
+import re
+
+import typing as t
+
+from lxml import html
+
+from searx.enginelib import EngineCache
+from searx.exceptions import SearxEngineAPIException
+from searx.network import get
+from searx.result_types import EngineResults
+from searx.result_types.image import ImageRef
+from searx.utils import eval_xpath_list, humanize_bytes
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
+
+
+about = {
+    "website": "https://giphy.com",
+    "wikidata_id": "Q17054335",
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+base_url = "https://giphy.com"
+api_url = "https://api.giphy.com"
+
+categories = ["images"]
+paging = True
+page_size = 15
+
+GiphyCategs = t.Literal["gifs", "stickers", "clips"]
+giphy_categ: GiphyCategs = "gifs"
+"""Giphy category to search in."""
+
+CACHE: EngineCache
+"""Cache for storing the extracted api key."""
+
+
+_GIPHY_API_KEY_RE = re.compile(r"[Aa]piKey\s*:\s*\"(\w+)\"")
+
+
+def setup(engine_settings: dict[str, str]) -> bool:
+    if giphy_categ not in t.get_args(GiphyCategs):
+        raise ValueError("invalid category: %s" % giphy_categ)
+
+    global CACHE  # pylint: disable=global-statement
+    CACHE = EngineCache(engine_settings["name"])
+
+    return True
+
+
+def _get_api_key() -> str:
+    """
+    Extract the Giphy API key from the JavaScript code. There are different API keys
+    (e.g. for mobile, desktop, ...), so we just pick a random one of these.
+    """
+    cached = CACHE.get("api_key")
+    if cached:
+        return cached
+
+    homepage_resp = get(base_url)
+    homepage_doc = html.fromstring(homepage_resp.text)
+
+    for script_src in eval_xpath_list(homepage_doc, "//script[contains(@src, 'layout')]/@src"):
+        script_resp = get(base_url + script_src)
+        api_keys = _GIPHY_API_KEY_RE.findall(script_resp.text)
+        if api_keys:
+            api_key = random.choice(api_keys)
+            CACHE.set("api_key", api_key, expire=60 * 60 * 6)  # 6 hours
+            return api_key
+
+    raise SearxEngineAPIException("failed to extract api keys")
+
+
+def request(query: str, params: "OnlineParams") -> None:
+    args = {
+        "q": query,
+        "api_key": _get_api_key(),
+        "limit": page_size,
+        "offset": (params["pageno"] - 1) * page_size,
+        "type": giphy_categ,
+    }
+    params["url"] = f"{api_url}/v1/{giphy_categ}/search?{urlencode(args)}"
+
+
+def response(resp: "SXNG_Response"):
+    res = EngineResults()
+
+    result: dict[str, t.Any]
+    for result in resp.json()["data"]:
+        img = result['images']['original']
+        formats = [
+            ImageRef(url=img["mp4"], subtype="mp4"),  # type: ignore
+            ImageRef(url=img["webp"], subtype="webp"),  # type: ignore
+        ]
+        thumb = (
+            result["images"].get("downsized")
+            or result["images"].get("downsized_medium")
+            or result["images"].get("downsized_small")
+            or result["images"].get("downsized_large")
+        )
+        res.add(
+            res.types.Image(
+                title=result["title"],
+                content=", ".join(result.get("tags", [])),
+                url=result["url"],
+                thumbnail_src=thumb.get("url") or img["url"],
+                img_src=img["url"],
+                resolution=f"{img['width']}x{img['height']}",
+                img_format="GIF",
+                formats=formats,
+                author=result["username"],
+                filesize=humanize_bytes(int(img["size"])),
+                source=result.get("source_tld") or "",
+            )
+        )
+
+    return res

searx/engines/iseek.py

@@ -0,0 +1,88 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""iseek_ is a search engine by the AI company Vantage Labs LLC,
+that focuses on medical and educational applicances.
+Although it's an AI company, it doesn't include any AI stuff in its results.
+
+.. _iseek : https://www.iseek.ai/
+"""
+
+import base64
+from hashlib import sha256
+import typing as t
+from urllib.parse import urlencode
+
+from searx.result_types import EngineResults
+
+if t.TYPE_CHECKING:
+    from searx.search.processors import OnlineParams
+    from searx.extended_types import SXNG_Response
+
+
+about = {
+    "website": 'https://www.iseek.com',
+    "wikidata_id": None,
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+categories = ["general"]
+paging = True
+
+base_url = "https://api.iseek.com"
+page_size = 10
+
+
+def _get_new_token(query: str, pageno: int) -> str:
+    """Create a new ``qToken``. This reduced the time for fetching subsequent pages
+    from 4 seconds to 200ms when testing."""
+    # The website uses a random value as qToken for the first page. For our use case,
+    # it's easier if the qToken can be deterministically re-calculated based on the search query,
+    # so that we can the same result when calling _get_new_token for the second, third, ... page
+    #
+    # var qToken = Math.ceil(Math.random() * parseInt("ZZZZ", 36)).toString(36);
+    # while (qToken.length < 4) qToken = '0' + qToken;
+    # qToken = qToken + "_" + pageno
+    query_hash = sha256(query.encode()).digest()
+    hash_start = base64.b64encode(query_hash).decode()[0:4]
+    return f"{hash_start}_{pageno}"
+
+
+def request(query: str, params: "OnlineParams"):
+    offset = (params["pageno"] - 1) * page_size
+
+    # always seems to find 20 results max
+    if offset >= 20:
+        params["url"] = None
+        return
+
+    args = {
+        "q": query,
+        "key": "core-web",
+        "num": str(page_size),
+        "off": offset,
+        "rSort": "__metasearch_score_d:desc",
+        # it supports many more fields, but none of them are really relevant
+        "names": "title_t,content_txt,url_s",
+        "qNames": "title_t",
+        "qToken": _get_new_token(query, params["pageno"]),
+    }
+    params["url"] = f"{base_url}/search?{urlencode(args)}"
+
+
+def response(resp: "SXNG_Response") -> EngineResults:
+    res = EngineResults()
+
+    for group in resp.json()["data"]:
+        group: dict[str, t.Any]
+        for result in group["doclist"]["docs"]:
+            result: dict[str, str]
+            res.add(
+                res.types.MainResult(
+                    url=result["url_s"],
+                    title=result["title_t"],
+                    content="".join(result["content_txt"]),
+                )
+            )
+
+    return res

searx/engines/marginalia.py

@@ -44,7 +44,7 @@ about = {
 
 base_url = "https://api2.marginalia-search.com"
 safesearch = True
-categories = ["general"]
+categories = ["general", "blogs"]
 paging = True
 results_per_page = 20
 api_key = None

searx/engines/startpagina.py

@@ -0,0 +1,107 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Startpagina is a Netherlands search engine by `Kompas`_. It takes all its
+results from Google.
+
+.. _Kompas: https://www.kompaspublishing.nl/
+"""
+
+import typing as t
+from urllib.parse import urlencode
+
+from dateutil import parser
+
+from searx.utils import format_duration
+from searx.result_types import EngineResults
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
+
+about = {
+    "website": "https://startpagina.nl",
+    "wikidata_id": None,
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+language = "ne"
+paging = True
+safesearch = True
+
+categories = ["general"]
+startpagina_categ = "web"
+"""Category to search in. Can be either "web", "images", "videos" or "news"."""
+page_size = 10
+
+
+api_url = "https://search.kompas.services"
+
+
+def init(_):
+    if startpagina_categ not in ("web", "images", "videos", "news"):
+        raise ValueError("invalid search type: %s" % startpagina_categ)
+
+
+def request(query: str, params: "OnlineParams") -> None:
+    args = {"q": query, "page_size": page_size, "page": params["pageno"]}
+    params["url"] = f"{api_url}/api/v2/search/{startpagina_categ}/?{urlencode(args)}"
+
+
+def response(resp: "SXNG_Response"):
+    res = EngineResults()
+
+    json_resp = resp.json()
+
+    for result in json_resp["results"]:
+        if startpagina_categ == "web":
+            res.add(
+                res.types.MainResult(
+                    url=result["original_url"],
+                    title=result["title"],
+                    content=result["description"],
+                )
+            )
+        elif startpagina_categ == "news":
+            publishedDate = None
+            try:
+                publishedDate = parser.parse(result["date"])
+            except parser.ParserError:
+                pass
+
+            res.add(
+                res.types.MainResult(
+                    url=result["original_url"],
+                    title=result["title"],
+                    content=result["description"],
+                    thumbnail=result["image"]["thumbnail_url"],
+                    publishedDate=publishedDate,
+                )
+            )
+        elif startpagina_categ == "videos":
+            res.add(
+                res.types.LegacyResult(
+                    template="videos.html",
+                    url=result["original_url"],
+                    title=result["title"],
+                    content=result["description"],
+                    thumbnail=result["video"]["thumbnail_url"],
+                    length=format_duration(result["video"]["duration"]),
+                )
+            )
+        elif startpagina_categ == "images":
+            res.add(
+                res.types.Image(
+                    url=result["original_url"],
+                    title=result["title"],
+                    content=result["description"],
+                    thumbnail_src=result["image"]["thumbnail_url"],
+                    resolution=f"{result['image']['width']}x{result['image']['height']}",
+                )
+            )
+
+    for related in json_resp["related_searches"]:
+        res.add(res.types.LegacyResult(suggestion=related["query"]))
+
+    return res

searx/engines/tusksearch.py

@@ -0,0 +1,162 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Tusksearch_ is an American search engine that claims to fight censorship.
+Its search results are (at least partially) from Brave.
+
+.. _Tusksearch: https://tusksearch.com/about
+"""
+
+from json import loads
+import random
+import typing as t
+from urllib.parse import urlencode
+from dateutil import parser
+
+from searx.exceptions import SearxEngineAPIException
+from searx.network import get
+from searx.utils import html_to_text
+from searx.result_types import EngineResults
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
+
+about = {
+    "website": "https://tusksearch.com",
+    "wikidata_id": None,
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "JSON",
+}
+
+paging = True
+
+categories = ["general"]
+tusk_categ = "web"
+"""Category to search in. Can be either "web", "images", "videos" or "news"."""
+
+
+api_url = "https://api.tusksearch.com"
+
+
+def init(_):
+    if tusk_categ not in ("web", "images", "videos", "news"):
+        raise ValueError("invalid search type: %s" % tusk_categ)
+
+
+def _obtain_x_sid() -> tuple[str, str]:
+    """
+    The session ID ("sid") is encoded as a byte array in ``embed.js``.
+    It is only valid for exactly one request, so we can't cache it.
+
+    The header key is usually called `x-sid-{UUIDv4}`, and the value is
+    usually a plain UUIDv4 (but a different one than in the header key).
+    """
+    resp = get(f"{api_url}/revcontent/embed.js")
+    if not resp.ok:
+        raise SearxEngineAPIException("failed to obtain request x-sid token")
+
+    # data is prefixed by 'var x='
+    data_array = loads(resp.text[6:])
+
+    def _byte_array_to_ascii(text: list[int]) -> str:
+        """
+        Converts a byte array (e.g. [81, 101, 97, 114, 88, 78, 71]) to the ASCII
+        string representation (e.g. "SearXNG").
+        """
+        return "".join([chr(x) for x in text])
+
+    x_sid_header = _byte_array_to_ascii(data_array[3])
+    x_sid_value = _byte_array_to_ascii(data_array[4])
+    return x_sid_header, x_sid_value
+
+
+def request(query: str, params: "OnlineParams") -> None:
+    # images don't support pagination, news and videos only support two pages
+    if tusk_categ == "images" and params["pageno"] > 1 or tusk_categ in ("news", "videos") and params["pageno"] > 2:
+        params["url"] = None
+        return
+
+    args = {
+        "q": query,
+        "p": params["pageno"],
+        "l": "center",  # political direction: "left", "center" or "right"
+    }
+    if tusk_categ == "images":
+        params["url"] = f"{api_url}/Search/Image?{urlencode(args)}"
+    else:
+        # web response also contains news and videos
+        params["url"] = f"{api_url}/Search/Web?{urlencode(args)}"
+
+    x_sid_header, x_sid_value = _obtain_x_sid()
+    params["headers"] = {
+        x_sid_header: x_sid_value,
+        # required - we send a random longitude and latitude instead of the actual user location
+        'x-lon': str(random.random() * 90),
+        'x-lat': str(random.random() * 90),
+    }
+
+
+def response(resp: "SXNG_Response"):
+    res = EngineResults()
+
+    json_resp = resp.json()["results"]
+
+    if tusk_categ == "web":
+        for result in (json_resp.get("web") or {}).get("results", []):
+            res.add(
+                res.types.MainResult(
+                    url=result["url"],
+                    title=html_to_text(result["title"]),
+                    content=html_to_text(result["description"]),
+                    thumbnail=(result["thumbnail"] or {}).get("src") or "",
+                )
+            )
+    elif tusk_categ == "news":
+        for result in (json_resp.get("news") or {}).get("results", []):
+            publishedDate = None
+            try:
+                publishedDate = parser.parse(result["age"])
+            except parser.ParserError:
+                pass
+
+            res.add(
+                res.types.MainResult(
+                    url=result["url"],
+                    title=html_to_text(result["title"]),
+                    content=html_to_text(result["description"]),
+                    thumbnail=result["thumbnail"]["src"],
+                    publishedDate=publishedDate,
+                )
+            )
+    elif tusk_categ == "videos":
+        for result in (json_resp.get("videos") or {}).get("results", []):
+            publishedDate = None
+            try:
+                publishedDate = parser.parse(result["age"])
+            except parser.ParserError:
+                pass
+
+            res.add(
+                res.types.LegacyResult(
+                    template="videos.html",
+                    url=result["url"],
+                    title=html_to_text(result["title"]),
+                    content=html_to_text(result["description"]),
+                    thumbnail=result["thumbnail"]["src"],
+                    publishedDate=publishedDate,
+                    length=result["video"].get("duration"),
+                )
+            )
+    elif tusk_categ == "images":
+        for result in json_resp:
+            res.add(
+                res.types.Image(
+                    url=result["url"],
+                    title=html_to_text(result["title"]),
+                    img_src=result["properties"]["url"],
+                    thumbnail_src=result["thumbnail"]["src"],
+                )
+            )
+
+    return res

searx/settings.yml

@@ -444,27 +444,6 @@ engines:
     shortcut: conda
     disabled: true
 
-  - name: aol
-    engine: aol
-    search_type: search
-    categories: [general]
-    shortcut: aol
-    disabled: true
-
-  - name: aol images
-    engine: aol
-    search_type: image
-    categories: [images]
-    shortcut: aoli
-    disabled: true
-
-  - name: aol videos
-    engine: aol
-    search_type: video
-    categories: [videos]
-    shortcut: aolv
-    disabled: true
-
   - name: arch linux wiki
     engine: archlinux
     shortcut: al
@@ -688,6 +667,26 @@ engines:
     disabled: true
     inactive: true
 
+  - name: cl0q
+    engine: json_engine
+    shortcut: cl
+    categories: general
+    paging: true
+    first_page_num: 0
+    page_size: 20
+    search_url: https://cl0q.com/search?q={query}&limit=20&offset={pageno}
+    results_query: results
+    url_query: domain
+    url_prefix: https://
+    title_query: title
+    content_query: description
+    disabled: true
+    inactive: true
+    about:
+      website: https://cl0q.com
+      description: "Open source network for searching domains"
+      results: JSON
+
   - name: cloudflareai
     engine: cloudflareai
     shortcut: cfai
@@ -978,6 +977,34 @@ engines:
     shortcut: fd
     disabled: true
 
+  - name: findfiles
+    engine: findfiles
+    findfiles_categ: all
+    categories: files
+    shortcut: fif
+    disabled: true
+
+  - name: findfiles images
+    engine: findfiles
+    findfiles_categ: image
+    categories: images
+    shortcut: fifi
+    disabled: true
+
+  - name: findfiles videos
+    engine: findfiles
+    findfiles_categ: video
+    categories: videos
+    shortcut: fifv
+    disabled: true
+
+  - name: findfiles music
+    engine: findfiles
+    findfiles_categ: audio
+    categories: music
+    shortcut: fifm
+    disabled: true
+
   - name: findthatmeme
     engine: findthatmeme
     shortcut: ftm
@@ -1114,6 +1141,11 @@ engines:
     search_type: text
     timeout: 10
 
+  - name: giphy
+    engine: giphy
+    shortcut: gif
+    disabled: true
+
   - name: gitlab
     engine: gitlab
     base_url: https://gitlab.com
@@ -1295,6 +1327,13 @@ engines:
       require_api_key: false
       results: JSON
 
+  - name: iseek
+    engine: iseek
+    shortcut: isk
+    timeout: 4
+    disabled: true
+    inactive: true
+
   - name: il post
     engine: il_post
     shortcut: pst
@@ -1383,6 +1422,23 @@ engines:
   #   api_key: ""  # required
   #   kagi_categ: videos
 
+  - name: kozmonavt
+    engine: xpath
+    paging: true
+    # pageno query param key is U+1D5D1 (looks like an "x")
+    search_url: https://kozmonavt.su/s?q={query}&%F0%9D%97%91={pageno}
+    shortcut: koz
+    disabled: true
+    # inactive: true
+    results_xpath: //div[contains(@class, 'list')]/section
+    url_xpath: concat('https://', substring-after(.//a/@href, '?q='))
+    title_xpath: .//a
+    content_xpath: .//div[contains(@class, 'snip')]
+    about:
+      website: https://kozmonavt.su
+      description: "Web site directory with its own index"
+      results: HTML
+
   - name: jisho
     engine: jisho
     shortcut: js
@@ -1400,6 +1456,22 @@ engines:
     shortcut: kc
     timeout: 4.0
 
+  - name: kukei
+    engine: xpath
+    categories: [general, blogs]
+    search_url: https://kukei.eu/?q={query}
+    shortcut: kuk
+    disabled: true
+    inactive: true
+    results_xpath: //ul/li[contains(@class, "result-item-first-level")]
+    url_xpath: .//a/@href
+    title_xpath: .//h3
+    content_xpath: .//p
+    about:
+      website: https://kukei.eu
+      description: "Curated search for the small web"
+      results: HTML
+
   - name: lemmy communities
     engine: lemmy
     lemmy_type: Communities
@@ -2033,7 +2105,7 @@ engines:
   - name: rawweb
     engine: json_engine
     shortcut: rw
-    categories: general
+    categories: [general, blogs]
     paging: true
     search_url: 'https://api.rawweb.org/api/search?keyword={query}&page={pageno}&lang=*'
     results_query: data
@@ -2088,7 +2160,7 @@ engines:
   - name: searchmysite
     engine: xpath
     shortcut: sms
-    categories: general
+    categories: [general, blogs]
     paging: true
     search_url: https://searchmysite.net/search/?q={query}&page={pageno}
     results_xpath: //div[contains(@class,'search-result')]
@@ -2404,6 +2476,35 @@ engines:
       - 5000
     inactive: true
 
+  - name: tusksearch
+    engine: tusksearch
+    shortcut: tu
+    tusk_categ: web
+    categories: general
+    disabled: true
+
+  - name: tusksearch images
+    engine: tusksearch
+    shortcut: tui
+    paging: false
+    tusk_categ: images
+    categories: images
+    disabled: true
+
+  - name: tusksearch videos
+    engine: tusksearch
+    shortcut: tuv
+    tusk_categ: videos
+    categories: videos
+    disabled: true
+
+  - name: tusksearch news
+    engine: tusksearch
+    shortcut: tun
+    tusk_categ: news
+    categories: news
+    disabled: true
+
   # tmp suspended - too slow, too many errors
   #  - name: urbandictionary
   #    engine      : xpath
@@ -2417,6 +2518,24 @@ engines:
     engine: unsplash
     shortcut: us
 
+  - name: unobtanium
+    engine: xpath
+    shortcut: uno
+    categories: [general, blogs]
+    paging: true
+    first_page_num: 0
+    search_url: https://unobtanium.rocks/search?search={query}&page={pageno}
+    results_xpath: //ul[contains(@class, 'search-results')]/li
+    url_xpath: .//a/@href
+    title_xpath: .//a/b
+    content_xpath: .//p[2]
+    disabled: true
+    inactive: true
+    about:
+      website: https://unobtanium.rocks
+      description: "Personal websites focused search engine."
+      results: HTML
+
   - name: yandex
     engine: yandex
     categories: general
@@ -2476,7 +2595,7 @@ engines:
     url_query: URL
     title_query: Title
     content_query: Snippet
-    categories: [general, web]
+    categories: [general, blogs]
     shortcut: wib
     disabled: true
     about:
@@ -2901,6 +3020,38 @@ engines:
     disabled: true
     inactive: true
 
+  - name: startpagina
+    engine: startpagina
+    shortcut: spnl
+    startpagina_categ: web
+    categories: general
+    disabled: true
+    inactive: true
+
+  - name: startpagina images
+    engine: startpagina
+    shortcut: spnli
+    startpagina_categ: images
+    categories: images
+    disabled: true
+    inactive: true
+
+  - name: startpagina videos
+    engine: startpagina
+    shortcut: spnlv
+    startpagina_categ: videos
+    categories: videos
+    disabled: true
+    inactive: true
+
+  - name: startpagina news
+    engine: startpagina
+    shortcut: spnln
+    startpagina_categ: news
+    categories: news
+    disabled: true
+    inactive: true
+
   - name: swisscows
     engine: swisscows
     categories: general
@@ -3023,6 +3174,22 @@ engines:
     shortcut: wttr
     timeout: 9.0
 
+  - name: xonaly
+    engine: xpath
+    search_url: https://xonaly.com/?query={query}
+    categories: general
+    shortcut: xo
+    disabled: true
+    inactive: true
+    results_xpath: //div[contains(@class, 'results-block')]/ul/li
+    url_xpath: ./div[contains(@class, 'result-title')]/a/@href
+    title_xpath: ./div[contains(@class, 'result-title')]/a
+    content_xpath: ./p[contains(@class, 'excerpt')]
+    about:
+      website: https://xonaly.com
+      description: "Canadian search engine with its own index"
+      results: HTML
+
   - name: zapmeta
     engine: xpath
     shortcut: zpm