[fix] Installation Script install fails (msgspec) (#5346)

Since #5280 has been merged, msgspec, like yaml, is a fixed part of the SearXNG
*settings framework* and therefore, like yaml, must be installed in the virtual
environment before installing SearXNG (``searx``).

The actual reason is that in SearXNG we store settings in the configuration that
are required for the installation of the ``searx`` package.  This means that
these settings (from settings.yml) are read in during the installation, and all the
necessary tools for this (pyyaml, msgspec, setuptools, etc.) must be installed
beforehand (chicken or the egg dilemma).

Related:

- https://github.com/searxng/searxng/pull/5280
- https://github.com/searxng/searxng/pull/5254

Closes: https://github.com/searxng/searxng/issues/5343

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>

client/simple/src/less/result_types/file.less

@@ -0,0 +1,22 @@
+// SPDX-License-Identifier: AGPL-3.0-or-later
+
+/*
+  Layout of the Files result class
+*/
+
+#main_results .result-file {
+  border: 1px solid var(--color-result-border);
+  margin: 0 @results-tablet-offset 1rem @results-tablet-offset !important;
+  .rounded-corners;
+
+  video {
+    width: 100%;
+    aspect-ratio: 16 / 9;
+    padding: 10px 0 0 0;
+  }
+
+  audio {
+    width: 100%;
+    padding: 10px 0 0 0;
+  }
+}

client/simple/src/less/style.less

@@ -163,7 +163,6 @@ article[data-vim-selected].category-videos,
 article[data-vim-selected].category-news,
 article[data-vim-selected].category-map,
 article[data-vim-selected].category-music,
-article[data-vim-selected].category-files,
 article[data-vim-selected].category-social {
   border: 1px solid var(--color-result-vim-arrow);
   .rounded-corners;
@@ -387,7 +386,6 @@ article[data-vim-selected].category-social {
 .category-news,
 .category-map,
 .category-music,
-.category-files,
 .category-social {
   border: 1px solid var(--color-result-border);
   margin: 0 @results-tablet-offset 1rem @results-tablet-offset !important;
@@ -1168,3 +1166,4 @@ pre code {
 @import "result_types/keyvalue.less";
 @import "result_types/code.less";
 @import "result_types/paper.less";
+@import "result_types/file.less";

docs/dev/result_types/main/file.rst

@@ -0,0 +1,7 @@
+.. _result_types.file:
+
+============
+File Results
+============
+
+.. automodule:: searx.result_types.file

docs/dev/result_types/main_result.rst

@@ -17,6 +17,7 @@ following types have been implemented so far ..
    main/keyvalue
    main/code
    main/paper
+   main/file
 
 The :ref:`LegacyResult <LegacyResult>` is used internally for the results that
 have not yet been typed.  The templates can be used as orientation until the
@@ -28,5 +29,4 @@ final typing is complete.
 - :ref:`template torrent`
 - :ref:`template map`
 - :ref:`template packages`
-- :ref:`template files`
 - :ref:`template products`

docs/dev/templates.rst

@@ -60,7 +60,7 @@ Fields used in the template :origin:`macro result_sub_header
 publishedDate : :py:obj:`datetime.datetime`
   The date on which the object was published.
 
-length: :py:obj:`time.struct_time`
+length: :py:obj:`datetime.timedelta`
   Playing duration in seconds.
 
 views: :py:class:`str`
@@ -469,38 +469,6 @@ links : :py:class:`dict`
   Additional links in the form of ``{'link_name': 'http://example.com'}``
 
 
-.. _template files:
-
-``files.html``
---------------
-
-Displays result fields from:
-
-- :ref:`macro result_header` and
-- :ref:`macro result_sub_header`
-
-Additional fields used in the :origin:`code.html
-<searx/templates/simple/result_templates/files.html>`:
-
-filename, size, time: :py:class:`str`
-  Filename, Filesize and Date of the file.
-
-mtype : ``audio`` | ``video`` | :py:class:`str`
-  Mimetype type of the file.
-
-subtype : :py:class:`str`
-  Mimetype / subtype of the file.
-
-abstract : :py:class:`str`
-  Abstract of the file.
-
-author : :py:class:`str`
-  Name of the author of the file
-
-embedded : :py:class:`str`
-  URL of an embedded media type (``audio`` or ``video``) / is collapsible.
-
-
 .. _template products:
 
 ``products.html``

searx/data/tracker_patterns.py

@@ -1,5 +1,6 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
 """Simple implementation to store TrackerPatterns data in a SQL database."""
+# pylint: disable=too-many-branches
 
 import typing as t
 
@@ -119,6 +120,12 @@ class TrackerPatternsDB:
 
         for rule in self.rules():
 
+            query_str: str = parsed_new_url.query
+            if not query_str:
+                # There are no more query arguments in the parsed_new_url on
+                # which rules can be applied, stop iterating over the rules.
+                break
+
             if not re.match(rule[self.Fields.url_regexp], new_url):
                 # no match / ignore pattern
                 continue
@@ -136,18 +143,32 @@ class TrackerPatternsDB:
                 #    overlapping urlPattern like ".*"
                 continue
 
-            # remove tracker arguments from the url-query part
             query_args: list[tuple[str, str]] = list(parse_qsl(parsed_new_url.query))
+            if query_args:
+                # remove tracker arguments from the url-query part
+                for name, val in query_args.copy():
+                    # remove URL arguments
+                    for pattern in rule[self.Fields.del_args]:
+                        if re.match(pattern, name):
+                            log.debug(
+                                "TRACKER_PATTERNS: %s remove tracker arg: %s='%s'", parsed_new_url.netloc, name, val
+                            )
+                            query_args.remove((name, val))
 
-            for name, val in query_args.copy():
-                # remove URL arguments
+                parsed_new_url = parsed_new_url._replace(query=urlencode(query_args))
+                new_url = urlunparse(parsed_new_url)
+
+            else:
+                # The query argument for URLs like:
+                # - 'http://example.org?q='       --> query_str is 'q=' and query_args is []
+                # - 'http://example.org?/foo/bar' --> query_str is 'foo/bar' and  query_args is []
+                # is a simple string and not a key/value dict.
                 for pattern in rule[self.Fields.del_args]:
-                    if re.match(pattern, name):
-                        log.debug("TRACKER_PATTERNS: %s remove tracker arg: %s='%s'", parsed_new_url.netloc, name, val)
-                        query_args.remove((name, val))
-
-            parsed_new_url = parsed_new_url._replace(query=urlencode(query_args))
-            new_url = urlunparse(parsed_new_url)
+                    if re.match(pattern, query_str):
+                        log.debug("TRACKER_PATTERNS: %s remove tracker arg: '%s'", parsed_new_url.netloc, query_str)
+                        parsed_new_url = parsed_new_url._replace(query="")
+                        new_url = urlunparse(parsed_new_url)
+                        break
 
         if new_url != url:
             return new_url

searx/engines/recoll.py

@@ -13,23 +13,12 @@ Configuration
 
 You must configure the following settings:
 
-``base_url``:
-  Location where recoll-webui can be reached.
+- :py:obj:`base_url`
+- :py:obj:`mount_prefix`
+- :py:obj:`dl_prefix`
+- :py:obj:`search_dir`
 
-``mount_prefix``:
-  Location where the file hierarchy is mounted on your *local* filesystem.
-
-``dl_prefix``:
-  Location where the file hierarchy as indexed by recoll can be reached.
-
-``search_dir``:
-  Part of the indexed file hierarchy to be search, if empty the full domain is
-  searched.
-
-Example
-=======
-
-Scenario:
+Example scenario:
 
 #. Recoll indexes a local filesystem mounted in ``/export/documents/reference``,
 #. the Recoll search interface can be reached at https://recoll.example.org/ and
@@ -37,107 +26,128 @@ Scenario:
 
 .. code:: yaml
 
-   base_url: https://recoll.example.org/
+   base_url: https://recoll.example.org
    mount_prefix: /export/documents
    dl_prefix: https://download.example.org
-   search_dir: ''
+   search_dir: ""
 
 Implementations
 ===============
 
 """
+import typing as t
 
 from datetime import date, timedelta
-from json import loads
 from urllib.parse import urlencode, quote
 
-# about
+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": None,
-    "wikidata_id": 'Q15735774',
-    "official_api_documentation": 'https://www.lesbonscomptes.com/recoll/',
+    "wikidata_id": "Q15735774",
+    "official_api_documentation": "https://www.lesbonscomptes.com/recoll/",
     "use_official_api": True,
     "require_api_key": False,
-    "results": 'JSON',
+    "results": "JSON",
 }
 
-# engine dependent config
 paging = True
 time_range_support = True
 
-# parameters from settings.yml
-base_url = None
-search_dir = ''
-mount_prefix = None
-dl_prefix = None
+base_url: str = ""
+"""Location where recoll-webui can be reached."""
 
-# embedded
-embedded_url = '<{ttype} controls height="166px" ' + 'src="{url}" type="{mtype}"></{ttype}>'
+mount_prefix: str = ""
+"""Location where the file hierarchy is mounted on your *local* filesystem."""
+
+dl_prefix: str = ""
+"""Location where the file hierarchy as indexed by recoll can be reached."""
+
+search_dir: str = ""
+"""Part of the indexed file hierarchy to be search, if empty the full domain is
+searched."""
+
+_s2i: dict[str | None, int] = {"day": 1, "week": 7, "month": 30, "year": 365}
 
 
-# helper functions
-def get_time_range(time_range):
-    sw = {'day': 1, 'week': 7, 'month': 30, 'year': 365}  # pylint: disable=invalid-name
+def setup(engine_settings: dict[str, t.Any]) -> bool:
+    """Initialization of the Recoll engine, checks if the mandatory values are
+    configured.
+    """
+    missing: list[str] = []
+    for cfg_name in ["base_url", "mount_prefix", "dl_prefix"]:
+        if not engine_settings.get(cfg_name):
+            missing.append(cfg_name)
+    if missing:
+        logger.error("missing recoll configuration: %s", missing)
+        return False
 
-    offset = sw.get(time_range, 0)
+    if engine_settings["base_url"].endswith("/"):
+        engine_settings["base_url"] = engine_settings["base_url"][:-1]
+    return True
+
+
+def search_after(time_range: str | None) -> str:
+    offset = _s2i.get(time_range, 0)
     if not offset:
-        return ''
-
+        return ""
     return (date.today() - timedelta(days=offset)).isoformat()
 
 
-# do search-request
-def request(query, params):
-    search_after = get_time_range(params['time_range'])
-    search_url = base_url + 'json?{query}&highlight=0'
-    params['url'] = search_url.format(
-        query=urlencode({'query': query, 'page': params['pageno'], 'after': search_after, 'dir': search_dir})
-    )
-
-    return params
+def request(query: str, params: "OnlineParams") -> None:
+    args = {
+        "query": query,
+        "page": params["pageno"],
+        "after": search_after(params["time_range"]),
+        "dir": search_dir,
+        "highlight": 0,
+    }
+    params["url"] = f"{base_url}/json?{urlencode(args)}"
 
 
-# get response from search-request
-def response(resp):
-    results = []
+def response(resp: "SXNG_Response") -> EngineResults:
 
-    response_json = loads(resp.text)
+    res = EngineResults()
+    json_data = resp.json()
 
-    if not response_json:
-        return []
+    if not json_data:
+        return res
 
-    for result in response_json.get('results', []):
-        title = result['label']
-        url = result['url'].replace('file://' + mount_prefix, dl_prefix)
-        content = '{}'.format(result['snippet'])
+    for result in json_data.get("results", []):
 
-        # append result
-        item = {'url': url, 'title': title, 'content': content, 'template': 'files.html'}
+        url = result.get("url", "").replace("file://" + mount_prefix, dl_prefix)
 
-        if result['size']:
-            item['size'] = int(result['size'])
-
-        for parameter in ['filename', 'abstract', 'author', 'mtype', 'time']:
-            if result[parameter]:
-                item[parameter] = result[parameter]
+        mtype = subtype = result.get("mime", "")
+        if mtype:
+            mtype, subtype = (mtype.split("/", 1) + [""])[:2]
 
         # facilitate preview support for known mime types
-        if 'mtype' in result and '/' in result['mtype']:
-            (mtype, subtype) = result['mtype'].split('/')
-            item['mtype'] = mtype
-            item['subtype'] = subtype
+        thumbnail = embedded = ""
+        if mtype in ["audio", "video"]:
+            embedded_url = '<{ttype} controls height="166px" ' + 'src="{url}" type="{mtype}"></{ttype}>'
+            embedded = embedded_url.format(ttype=mtype, url=quote(url.encode("utf8"), "/:"), mtype=result["mtype"])
+        if mtype in ["image"] and subtype in ["bmp", "gif", "jpeg", "png"]:
+            thumbnail = url
 
-            if mtype in ['audio', 'video']:
-                item['embedded'] = embedded_url.format(
-                    ttype=mtype, url=quote(url.encode('utf8'), '/:'), mtype=result['mtype']
-                )
-
-            if mtype in ['image'] and subtype in ['bmp', 'gif', 'jpeg', 'png']:
-                item['thumbnail'] = url
-
-        results.append(item)
-
-    if 'nres' in response_json:
-        results.append({'number_of_results': response_json['nres']})
-
-    return results
+        res.add(
+            res.types.File(
+                title=result.get("label", ""),
+                url=url,
+                content=result.get("snippet", ""),
+                size=result.get("size", ""),
+                filename=result.get("filename", ""),
+                abstract=result.get("abstract", ""),
+                author=result.get("author", ""),
+                mtype=mtype,
+                subtype=subtype,
+                time=result.get("time", ""),
+                embedded=embedded,
+                thumbnail=thumbnail,
+            )
+        )
+    return res

searx/engines/wikicommons.py

@@ -1,102 +1,208 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
-"""Wikimedia Commons (images)"""
+"""`Wikimedia Commons`_ is a collection of more than 120 millions freely usable
+media files to which anyone can contribute.
+
+This engine uses the `MediaWiki query API`_, with which engines can be configured
+for searching images, videos, audio, and other files in the Wikimedia.
+
+.. _MediaWiki query API: https://commons.wikimedia.org/w/api.php?action=help&modules=query
+.. _Wikimedia Commons: https://commons.wikimedia.org/
+
+
+Configuration
+=============
+
+The engine has the following additional settings:
+
+.. code:: yaml
+
+   - name: wikicommons.images
+     engine: wikicommons
+     wc_search_type: image
+
+   - name: wikicommons.videos
+     engine: wikicommons
+     wc_search_type: video
+
+   - name: wikicommons.audio
+     engine: wikicommons
+     wc_search_type: audio
+
+   - name: wikicommons.files
+     engine: wikicommons
+     wc_search_type: file
+
+
+Implementations
+===============
+
+"""
+
+import typing as t
 
 import datetime
-
-from urllib.parse import urlencode
+import pathlib
+from urllib.parse import urlencode, unquote
 
 from searx.utils import html_to_text, humanize_bytes
+from searx.result_types import EngineResults
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
 
-# about
 about = {
-    "website": 'https://commons.wikimedia.org/',
-    "wikidata_id": 'Q565',
-    "official_api_documentation": 'https://commons.wikimedia.org/w/api.php',
+    "website": "https://commons.wikimedia.org/",
+    "wikidata_id": "Q565",
+    "official_api_documentation": "https://commons.wikimedia.org/w/api.php",
     "use_official_api": True,
     "require_api_key": False,
-    "results": 'JSON',
+    "results": "JSON",
 }
-categories = ['images']
-search_type = 'images'
 
-base_url = "https://commons.wikimedia.org"
-search_prefix = (
-    '?action=query'
-    '&format=json'
-    '&generator=search'
-    '&gsrnamespace=6'
-    '&gsrprop=snippet'
-    '&prop=info|imageinfo'
-    '&iiprop=url|size|mime'
-    '&iiurlheight=180'  # needed for the thumb url
-)
+categories: list[str] = []
 paging = True
 number_of_results = 10
 
-search_types = {
-    'images': 'bitmap|drawing',
-    'videos': 'video',
-    'audio': 'audio',
-    'files': 'multimedia|office|archive|3d',
+wc_api_url = "https://commons.wikimedia.org/w/api.php"
+wc_search_type: str = ""
+
+SEARCH_TYPES: dict[str, str] = {
+    "image": "bitmap|drawing",
+    "video": "video",
+    "audio": "audio",
+    "file": "multimedia|office|archive|3d",
 }
+# FileType = t.Literal["bitmap", "drawing", "video", "audio", "multimedia", "office", "archive", "3d"]
+# FILE_TYPES = list(t.get_args(FileType))
 
 
-def request(query, params):
-    language = 'en'
-    if params['language'] != 'all':
-        language = params['language'].split('-')[0]
+def setup(engine_settings: dict[str, t.Any]) -> bool:
+    """Initialization of the Wikimedia engine, checks if the value configured in
+    :py:obj:`wc_search_type` is valid."""
 
-    if search_type not in search_types:
-        raise ValueError(f"Unsupported search type: {search_type}")
+    if engine_settings.get("wc_search_type") not in SEARCH_TYPES:
+        logger.error(
+            "wc_search_type: %s isn't a valid file type (%s)",
+            engine_settings.get("wc_search_type"),
+            ",".join(SEARCH_TYPES.keys()),
+        )
+        return False
+    return True
 
-    filetype = search_types[search_type]
 
+def request(query: str, params: "OnlineParams") -> None:
+    uselang: str = "en"
+    if params["searxng_locale"] != "all":
+        uselang = params["searxng_locale"].split("-")[0]
+    filetype = SEARCH_TYPES[wc_search_type]
     args = {
-        'uselang': language,
-        'gsrlimit': number_of_results,
-        'gsroffset': number_of_results * (params["pageno"] - 1),
-        'gsrsearch': f"filetype:{filetype} {query}",
+        # https://commons.wikimedia.org/w/api.php
+        "format": "json",
+        "uselang": uselang,
+        "action": "query",
+        # https://commons.wikimedia.org/w/api.php?action=help&modules=query
+        "prop": "info|imageinfo",
+        # generator (gsr optins) https://commons.wikimedia.org/w/api.php?action=help&modules=query%2Bsearch
+        "generator": "search",
+        "gsrnamespace": "6",  # https://www.mediawiki.org/wiki/Help:Namespaces#Renaming_namespaces
+        "gsrprop": "snippet",
+        "gsrlimit": number_of_results,
+        "gsroffset": number_of_results * (params["pageno"] - 1),
+        "gsrsearch": f"filetype:{filetype} {query}",
+        # imageinfo: https://commons.wikimedia.org/w/api.php?action=help&modules=query%2Bimageinfo
+        "iiprop": "url|size|mime",
+        "iiurlheight": "180",  # needed for the thumb url
     }
-
-    params["url"] = f"{base_url}/w/api.php{search_prefix}&{urlencode(args, safe=':|')}"
-    return params
+    params["url"] = f"{wc_api_url}?{urlencode(args, safe=':|')}"
 
 
-def response(resp):
-    results = []
-    json = resp.json()
+def response(resp: "SXNG_Response") -> EngineResults:
 
-    if not json.get("query", {}).get("pages"):
-        return results
-    for item in json["query"]["pages"].values():
+    res = EngineResults()
+    json_data = resp.json()
+    pages = json_data.get("query", {}).get("pages", {}).values()
+
+    for item in pages:
+
+        if not item.get("imageinfo", []):
+            continue
         imageinfo = item["imageinfo"][0]
-        title = item["title"].replace("File:", "").rsplit('.', 1)[0]
-        result = {
-            'url': imageinfo["descriptionurl"],
-            'title': title,
-            'content': html_to_text(item["snippet"]),
-        }
 
-        if search_type == "images":
-            result['template'] = 'images.html'
-            result['img_src'] = imageinfo["url"]
-            result['thumbnail_src'] = imageinfo["thumburl"]
-            result['resolution'] = f'{imageinfo["width"]} x {imageinfo["height"]}'
-        else:
-            result['thumbnail'] = imageinfo["thumburl"]
+        title: str = item["title"].replace("File:", "").rsplit(".", 1)[0]
+        content = html_to_text(item["snippet"])
 
-        if search_type == "videos":
-            result['template'] = 'videos.html'
-            if imageinfo.get('duration'):
-                result['length'] = datetime.timedelta(seconds=int(imageinfo['duration']))
-            result['iframe_src'] = imageinfo['url']
-        elif search_type == "files":
-            result['template'] = 'files.html'
-            result['metadata'] = imageinfo['mime']
-            result['size'] = humanize_bytes(imageinfo['size'])
-        elif search_type == "audio":
-            result['iframe_src'] = imageinfo['url']
+        url: str = imageinfo["descriptionurl"]
+        media_url: str = imageinfo["url"]
+        mimetype: str = imageinfo["mime"]
+        thumbnail: str = imageinfo["thumburl"]
+        size = imageinfo.get("size")
+        if size:
+            size = humanize_bytes(size)
 
-        results.append(result)
+        duration = None
+        seconds: str = imageinfo.get("duration")
+        if seconds:
+            try:
+                duration = datetime.timedelta(seconds=int(seconds))
+            except OverflowError:
+                pass
 
-    return results
+        if wc_search_type == "file":
+            res.add(
+                res.types.File(
+                    title=title,
+                    url=url,
+                    content=content,
+                    size=size,
+                    mimetype=mimetype,
+                    filename=unquote(pathlib.Path(media_url).name),
+                    embedded=media_url,
+                    thumbnail=thumbnail,
+                )
+            )
+            continue
+
+        if wc_search_type == "image":
+            res.add(
+                res.types.LegacyResult(
+                    template="images.html",
+                    title=title,
+                    url=url,
+                    content=content,
+                    img_src=imageinfo["url"],
+                    thumbnail_src=thumbnail,
+                    resolution=f"{imageinfo['width']} x {imageinfo['height']}",
+                    img_format=imageinfo["mime"],
+                    filesize=size,
+                )
+            )
+            continue
+
+        if wc_search_type == "video":
+            res.add(
+                res.types.LegacyResult(
+                    template="videos.html",
+                    title=title,
+                    url=url,
+                    content=content,
+                    iframe_src=media_url,
+                    length=duration,
+                )
+            )
+            continue
+
+        if wc_search_type == "audio":
+            res.add(
+                res.types.MainResult(
+                    template="default.html",
+                    title=title,
+                    url=url,
+                    content=content,
+                    audio_src=media_url,
+                    length=duration,
+                )
+            )
+            continue
+
+    return res

searx/result_types/__init__.py

@@ -23,6 +23,7 @@ __all__ = [
     "WeatherAnswer",
     "Code",
     "Paper",
+    "File",
 ]
 
 import typing as t
@@ -33,6 +34,7 @@ from .answer import AnswerSet, Answer, Translations, WeatherAnswer
 from .keyvalue import KeyValue
 from .code import Code
 from .paper import Paper
+from .file import File
 
 
 class ResultList(list[Result | LegacyResult], abc.ABC):
@@ -47,6 +49,7 @@ class ResultList(list[Result | LegacyResult], abc.ABC):
         KeyValue = KeyValue
         Code = Code
         Paper = Paper
+        File = File
         MainResult = MainResult
         Result = Result
         Translations = Translations

searx/result_types/_base.py

@@ -27,7 +27,6 @@ import typing as t
 import re
 import urllib.parse
 import warnings
-import time
 import datetime
 
 from collections.abc import Callable
@@ -236,13 +235,6 @@ class Result(msgspec.Struct, kw_only=True):
     url: str | None = None
     """A link related to this *result*"""
 
-    template: str = "default.html"
-    """Name of the template used to render the result.
-
-    By default :origin:`result_templates/default.html
-    <searx/templates/simple/result_templates/default.html>` is used.
-    """
-
     engine: str | None = ""
     """Name of the engine *this* result comes from.  In case of *plugins* a
     prefix ``plugin:`` is set, in case of *answerer* prefix ``answerer:`` is
@@ -350,6 +342,13 @@ class Result(msgspec.Struct, kw_only=True):
 class MainResult(Result):  # pylint: disable=missing-class-docstring
     """Base class of all result types displayed in :ref:`area main results`."""
 
+    template: str = "default.html"
+    """Name of the template used to render the result.
+
+    By default :origin:`result_templates/default.html
+    <searx/templates/simple/result_templates/default.html>` is used.
+    """
+
     title: str = ""
     """Link title of the result item."""
 
@@ -359,6 +358,12 @@ class MainResult(Result):  # pylint: disable=missing-class-docstring
     img_src: str = ""
     """URL of a image that is displayed in the result item."""
 
+    iframe_src: str = ""
+    """URL of an embedded ``<iframe>`` / the frame is collapsible."""
+
+    audio_src: str = ""
+    """URL of an embedded ``<audio controls>``."""
+
     thumbnail: str = ""
     """URL of a thumbnail that is displayed in the result item."""
 
@@ -372,7 +377,7 @@ class MainResult(Result):  # pylint: disable=missing-class-docstring
     completely eliminated.
     """
 
-    length: time.struct_time | None = None
+    length: datetime.timedelta | None = None
     """Playing duration in seconds."""
 
     views: str = ""

searx/result_types/file.py

@@ -0,0 +1,94 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""
+Typification of the *file* results.  Results of this type are rendered in
+the :origin:`file.html <searx/templates/simple/result_templates/file.html>`
+template.
+
+----
+
+.. autoclass:: File
+   :members:
+   :show-inheritance:
+
+"""
+# pylint: disable=too-few-public-methods
+
+
+__all__ = ["File"]
+
+import typing as t
+import mimetypes
+
+from ._base import MainResult
+
+
+@t.final
+class File(MainResult, kw_only=True):
+    """Class for results of type *file*"""
+
+    template: str = "file.html"
+
+    filename: str = ""
+    """Name of the file."""
+
+    size: str = ""
+    """Size of bytes in human readable notation (``MB`` for 1024 * 1024 Bytes
+    file size.)"""
+
+    time: str = ""
+    """Indication of a time, such as the date of the last modification or the
+    date of creation. This is a simple string, the *date* of which can be freely
+    chosen according to the context."""
+
+    mimetype: str = ""
+    """Mimetype/Subtype of the file.  For ``audio`` and ``video``, a URL can be
+    passed in the :py:obj:`File.embedded` field to embed the referenced media in
+    the result.  If no value is specified, the MIME type is determined from
+    ``self.filename`` or, alternatively, from ``self.embedded`` (if either of
+    the two values is set)."""
+
+    abstract: str = ""
+    """Abstract of the file."""
+
+    author: str = ""
+    """Author of the file."""
+
+    embedded: str = ""
+    """URL of an embedded media type (audio or video) / is collapsible."""
+
+    mtype: str = ""
+    """Used for displaying :py:obj:`File.embedded`.  Its value is automatically
+    populated from the base type of :py:obj:`File.mimetype`, and can be
+    explicitly set to enforce e.g. ``audio`` or ``video`` when mimetype is
+    something like "application/ogg" but its know the content is for example a
+    video."""
+
+    subtype: str = ""
+    """Used for displaying :py:obj:`File.embedded`.  Its value is automatically
+    populated from the subtype type of :py:obj:`File.mimetype`, and can be
+    explicitly set to enforce a subtype for the :py:obj:`File.embedded`
+    element."""
+
+    def __post_init__(self):
+        super().__post_init__()
+
+        if not self.mtype or not self.subtype:
+
+            fn = self.filename or self.embedded
+            if not self.mimetype and fn:
+                self.mimetype = mimetypes.guess_type(fn, strict=False)[0] or ""
+
+            mtype, subtype = (self.mimetype.split("/", 1) + [""])[:2]
+
+            if not self.mtype:
+                # I don't know why, but the ogg video stream is not displayed,
+                # may https://github.com/videojs/video.js can help?
+                if self.embedded.endswith(".ogv"):
+                    self.mtype = "video"
+                elif self.embedded.endswith(".oga"):
+                    self.mtype = "audio"
+                else:
+                    self.mtype = mtype
+
+            if not self.subtype:
+                self.subtype = subtype

searx/settings.yml

@@ -2298,31 +2298,27 @@ engines:
 
   - name: wikicommons.images
     engine: wikicommons
-    shortcut: wc
+    shortcut: wci
     categories: images
-    search_type: images
-    number_of_results: 10
+    wc_search_type: image
 
   - name: wikicommons.videos
     engine: wikicommons
     shortcut: wcv
     categories: videos
-    search_type: videos
-    number_of_results: 10
+    wc_search_type: video
 
   - name: wikicommons.audio
     engine: wikicommons
     shortcut: wca
     categories: music
-    search_type: audio
-    number_of_results: 10
+    wc_search_type: audio
 
   - name: wikicommons.files
     engine: wikicommons
     shortcut: wcf
     categories: files
-    search_type: files
-    number_of_results: 10
+    wc_search_type: file
 
   - name: wolframalpha
     shortcut: wa

searx/templates/simple/result_templates/default.html

@@ -3,7 +3,7 @@
 {{ result_header(result, favicons, image_proxify) -}}
 {{- result_sub_header(result) -}}
 {% if result.iframe_src -%}
-<p class="altlink"><a class="btn-collapse collapsed media-loader disabled_if_nojs" data-target="#result-media-{{ index }}" data-btn-text-collapsed="{{ _('show media') }}" data-btn-text-not-collapsed="{{ _('hide media') }}">{{ icon('music-note') }} {{ _('show media') }}</a></p>
+<p class="altlink"><a class="btn-collapse collapsed media-loader disabled_if_nojs" data-target="#result-media-{{ index }}" data-btn-text-collapsed="{{ _('show media') }}" data-btn-text-not-collapsed="{{ _('hide media') }}">{{ icon('play') }} {{ _('show media') }}</a></p>
 {%- endif %}
 {%- if result.content %}
   <p class="content">

searx/templates/simple/result_templates/file.html

@@ -0,0 +1,74 @@
+{% from "simple/macros.html" import result_header, result_sub_header, result_sub_footer, result_footer, result_link with context %}
+{% from "simple/icons.html" import icon_small %}
+
+{{ result_header(result, favicons, image_proxify) }}
+{{ result_sub_header(result) }}
+
+{% if result.abstract %}
+  <p class="abstract">{{ result.abstract|safe }}</p>
+{% endif -%}
+
+{%- if result.content %}
+  <p class="content">{{ result.content|safe }}</p>
+{% endif -%}
+
+<div class="attributes">
+  {% if result.author %}
+  <div>
+    <span>{{ _("Author") }}:</span>
+    <span>{{ result.author }}</span>
+  </div>
+  {% endif %}
+  {% if result.filename %}
+    <div>
+      <span>{{ _("Filename") }}:</span>
+      <span>{{ result.filename }}</span>
+    </div>
+  {% endif %}
+  {% if result.size %}
+    <div>
+      <span>{{ _("Filesize") }}:</span>
+      <span>{{ result.size }}</span>
+    </div>
+  {% endif %}
+  {% if result.time %}
+    <div>
+      <span>{{ _("Date") }}:</span>
+      <span>{{ result.time }}</span>
+    </div>
+  {% endif %}
+  {% if result.mimetype %}
+    <div>
+      <span>{{ _("Type") }}:</span>
+      <span>{{ result.mimetype }}</span>
+    </div>
+  {% endif %}
+</div>
+
+{% if result.embedded %}
+  {% if result.mtype in ("audio", "video") %}
+    <p class="altlink">
+      <a class="btn-collapse collapsed media-loader disabled_if_nojs"
+        data-target="#result-media-{{ index }}"
+        data-btn-text-collapsed="{{ _("show media") }}"
+        data-btn-text-not-collapsed="{{ _("hide media") }}"
+      >
+        {{ _("show media") }}
+      </a>
+    </p>
+    <div id="result-media-{{ index }}" class="embedded-{{ result.mtype }} invisible">
+      <{{ result.mtype }} controls preload="metadata" {% if result.thumbnail %}poster="{{ result.thumbnail }}" {%  endif %}>
+      <source src="{{result.embedded}}" type="{{ result.mtype }}/{{ result.subtype }}">
+      </{{ result.mtype }}>
+    </div>
+  {% else %}
+    <p class="altlink">
+      <a href="{{result.embedded }}" target="_blank" rel="noopener noreferrer" download>
+        {{  _("Download") }}
+      </a>
+    </p>
+  {% endif %}
+{% endif %}
+
+{{ result_sub_footer(result) }}
+{{ result_footer(result) }}

searx/templates/simple/result_templates/files.html

@@ -1,45 +0,0 @@
-{% from 'simple/macros.html' import result_header, result_sub_header, result_sub_footer, result_footer, result_link with context %}
-{% from 'simple/icons.html' import icon_small %}
-
-{{- result_header(result, favicons, image_proxify) -}}
-{{- result_sub_header(result) -}}
-
-{%- if result.embedded -%}
-    <small> &bull; <a class="text-info btn-collapse collapsed cursor-pointer media-loader disabled_if_nojs" data-toggle="collapse" data-target="#result-media-{{ index }}" data-btn-text-collapsed="{{ _('show media') }}" data-btn-text-not-collapsed="{{ _('hide media') }}">
-    {%- if result.mtype == 'audio' %}{{ icon_small('musical-notes') -}}
-    {%- elif result.mtype == 'video' %} {{ icon_small('play') -}}
-    {%- endif %} {{ _('show media') }}</a></small>
-{%- endif -%}
-
-{%- if result.embedded -%}
-<div id="result-media-{{ index }}" class="collapse invisible">
-   {{- result.embedded|safe -}}
-</div>
-{%- endif -%}
-
-{%- if result.abstract %}<p class="result-content result-abstract">{{ result.abstract|safe }}</p>{% endif -%}
-
-{%- if result.img_src -%}
-<div class="container-fluid">
-    <div class="row">
-<img src="{{ image_proxify(result.img_src) }}" alt="{{ result.title|striptags }}" title="{{ result.title|striptags }}" style="width: auto; max-height: 60px; min-height: 60px;" class="col-xs-2 col-sm-4 col-md-4 result-content">
-{%- if result.content %}<p class="result-content col-xs-8 col-sm-8 col-md-8">{{ result.content|safe }}</p>{% endif -%}
-    </div>
-</div>
-{%- else -%}
-{%- if result.content %}<p class="result-content">{{ result.content|safe }}</p>{% endif -%}
-{%- endif -%}
-
-<table class="result-metadata result-content">
-{%- if result.author %}<tr><td>{{ _('Author') }}</td><td>{{ result.author|safe }}</td></tr>{% endif -%}
-
-{%- if result.filename %}<tr><td>{{ _('Filename') }}</td><td>{{ result.filename|safe }}</td></tr>{% endif -%}
-
-{%- if result.size %}<tr><td>{{ _('Filesize') }}</td><td>{{ result.size|safe }}</td></tr>{%- endif -%}
-
-{%- if result.time %}<tr><td>{{ _('Date') }}</td><td>{{ result.time|safe }}</td></tr>{% endif -%}
-
-{%- if result.mtype %}<tr><td>{{ _('Type') }}</td><td>{{ result.mtype|safe }}/{{ result.subtype|safe }}</td></tr>{% endif -%}
-</table>
-
-{{ result_footer(result) }}

searx/webapp.py

@@ -356,6 +356,12 @@ def get_pretty_url(parsed_url: urllib.parse.ParseResult):
     path = parsed_url.path
     path = path[:-1] if len(path) > 0 and path[-1] == '/' else path
     path = unquote(path.replace("/", " › "))
+
+    # Keep the query argument for URLs like:
+    # - 'http://example.org?/foo/bar' --> parsed_url.query is 'foo/bar'
+    query_args: list[tuple[str, str]] = list(urllib.parse.parse_qsl(parsed_url.query))
+    if not query_args and parsed_url.query:
+        path += (" › .." if len(parsed_url.query) > 24 else " › ") + parsed_url.query[-24:]
     return [parsed_url.scheme + "://" + parsed_url.netloc, path]
 
 

utils/searxng.sh

@@ -492,6 +492,7 @@ pip install -U pip
 pip install -U setuptools
 pip install -U wheel
 pip install -U pyyaml
+pip install -U msgspec
 cd ${SEARXNG_SRC}
 pip install --use-pep517 --no-build-isolation -e .
 EOF