[mod] migration of some engine results to the video class (#6743)

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

searx/engines/360search_videos.py

@@ -52,7 +52,7 @@ def response(resp) -> EngineResults:
                 published_date = None
 
         res.add(
-            res.types.LegacyResult(
+            res.types.Video(
                 url=entry["play_url"],
                 title=html_to_text(entry["title"]),
                 content=html_to_text(entry["description"]),

searx/engines/brave.py

@@ -127,7 +127,7 @@ from dateutil import parser
 from searx import locales, logger
 from searx.enginelib.traits import EngineTraits
 from searx.exceptions import SearxEngineResponseException
-from searx.result_types import EngineResults, MainResult
+from searx.result_types import EngineResults, MainResult, Video
 from searx.result_types.image import Image
 from searx.utils import html_to_text, js_obj_str_to_json_str, js_obj_str_to_python
 
@@ -331,8 +331,7 @@ def _parse_video_result(result: dict[str, t.Any]) -> MainResult:
     video: dict[str, t.Any] = result.get("video", {})
     thumbnail: dict[str, t.Any] = result.get("thumbnail", {})
 
-    return MainResult(
+    return Video(
-        template="videos.html",
         title=result.get("title", ""),
         url=result.get("url"),
         content=result.get("description", ""),

searx/engines/dogpile.py

@@ -5,13 +5,13 @@
 """
 
 import typing as t
-from datetime import datetime, timezone
+from datetime import datetime, timezone, timedelta
 import html
 
 from searx.enginelib import EngineCache
 from searx.exceptions import SearxEngineAPIException
 from searx.network import post
-from searx.utils import format_duration, html_to_text, humanize_number
+from searx.utils import html_to_text, humanize_number
 from searx.result_types import EngineResults
 
 if t.TYPE_CHECKING:
@@ -98,14 +98,13 @@ def response(resp: "SXNG_Response"):
             )
         elif dogpile_categ == "videos":
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    template="videos.html",
                     url=result["clickUrl"],
                     title=html_to_text(result["title"]),
                     content=html_to_text(result["description"]),
                     thumbnail=result["thumbnailUrl"],
                     publishedDate=datetime.fromisoformat(result["publishDate"]),
-                    length=format_duration(result["duration"]),
+                    length=timedelta(seconds=result["duration"]),
                     views=humanize_number(result["viewCount"]),
                 )
             )

searx/engines/duckduckgo_extra.py

@@ -10,7 +10,7 @@ from datetime import datetime
 from urllib.parse import urlencode
 from urllib.parse import quote_plus
 
-from searx.result_types import EngineResults, MainResult, LegacyResult, Image
+from searx.result_types import EngineResults, MainResult, Image, Video
 from searx.utils import html_to_text, gen_useragent, extr
 from searx.network import get  # see https://github.com/searxng/searxng/issues/762
 
@@ -162,13 +162,11 @@ def _image_result(result):
 
 
 def _video_result(result):
-    return LegacyResult(
+    return Video(
-        template='videos.html',
         url=result['content'],
         title=result['title'],
         content=result['description'],
         thumbnail=result['images'].get('small') or result['images'].get('medium'),
-        source=result['provider'],
         length=result['duration'],
         metadata=result.get('uploader'),
     )

searx/engines/findfiles.py

@@ -92,8 +92,7 @@ def response(resp: "SXNG_Response") -> EngineResults:
         ):
             video_src = extract_text(eval_xpath(result, ".//video/@src")) or ""
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    template="videos.html",
                     url=video_src,
                     title=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/span")) or "",
                     iframe_src=video_src or "",

searx/engines/fireball.py

@@ -153,16 +153,13 @@ def response(resp: "SXNG_Response") -> EngineResults:
                 length = result["video"].get("duration")
 
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    {
+                    url=result["url"],
-                        "template": "videos.html",
+                    title=html_to_text(result["title"]),
-                        "url": result["url"],
+                    content=html_to_text(result["description"]),
-                        "title": html_to_text(result["title"]),
+                    thumbnail=result.get("thumbnail", {}).get("original"),
-                        "content": html_to_text(result["description"]),
+                    length=length,
-                        "thumbnail": result.get("thumbnail", {}).get("original"),
+                    publishedDate=published_date,
-                        "length": length,
-                        "publishedDate": published_date,
-                    }
                 )
             )
 

searx/engines/google_videos.py

@@ -73,12 +73,11 @@ def response(resp: "SXNG_Response") -> EngineResults:
                 break
 
         results.add(
-            results.types.MainResult(
+            results.types.Video(
                 url=url,
                 title=title,
                 thumbnail=thumbnail,
                 length=length,
-                template="videos.html",
             )
         )
 

searx/engines/kagi.py

@@ -178,17 +178,14 @@ def response(resp: "SXNG_Response") -> EngineResults:
                 length = parse_duration_string(result["props"]["duration"])
 
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    {
+                    url=result["url"],
-                        "template": "videos.html",
+                    title=html_to_text(result.get("title", "no title available")),
-                        "url": result["url"],
+                    content=html_to_text(result.get("snippet", "")),
-                        "title": html_to_text(result.get("title", "no title available")),
+                    thumbnail=result.get("image", {}).get("url"),
-                        "content": html_to_text(result.get("snippet", "")),
+                    publishedDate=published_date,
-                        "thumbnail": result.get("image", {}).get("url"),
+                    author=result.get("props", {}).get("creator_name"),
-                        "publishedDate": published_date,
+                    length=length,
-                        "author": result.get("props", {}).get("creator_name"),
-                        "length": length,
-                    }
                 )
             )
 

searx/engines/luxxle.py

@@ -157,8 +157,7 @@ def _news_results(doc: ElementType, res: EngineResults):
 def _video_results(doc: ElementType, res: EngineResults):
     for result in eval_xpath_list(doc, "//div[@id='mainResults']/div[contains(@class, 'mediaResult')]"):
         res.add(
-            res.types.MainResult(
+            res.types.Video(
-                template="videos.html",
                 url=extract_text(eval_xpath(result, "./@data-url")) or "",
                 title=extract_text(eval_xpath(result, ".//div[contains(@class, 'mediaResultTitleVideo')]/a")) or "",
                 content=extract_text(eval_xpath(result, ".//div[contains(@class, 'mediaResultDescription')]")) or "",

searx/engines/mediathekviewweb.py

@@ -5,6 +5,7 @@ import typing as t
 import datetime
 
 from searx.result_types import EngineResults
+from searx.utils import parse_duration_string
 
 if t.TYPE_CHECKING:
     from searx.extended_types import SXNG_Response
@@ -67,13 +68,12 @@ def response(resp: "SXNG_Response") -> EngineResults:
             continue
 
         res.add(
-            res.types.LegacyResult(
+            res.types.Video(
                 url=video_url,
                 title="%(channel)s: %(title)s (%(hms)s)" % item,
-                length=item["hms"],
+                length=parse_duration_string(item["hms"]),
                 content="%(description)s" % item,
                 iframe_src=video_url,
-                template="videos.html",
             )
         )
 

searx/engines/naver.py

@@ -201,7 +201,7 @@ def parse_videos(data):
     for item in eval_xpath_list(dom, "//li[contains(@class, 'video_item')]"):
         url = eval_xpath_getindex(item, ".//a[contains(@class, 'info_title')]/@href", 0)
 
-        thumbnail = None
+        thumbnail = ""
         try:
             thumbnail = eval_xpath_getindex(item, ".//img[contains(@class, 'thumb')]/@src", 0)
         except (ValueError, TypeError, SearxEngineXPathException):
@@ -214,9 +214,8 @@ def parse_videos(data):
             pass
 
         res.add(
-            res.types.LegacyResult(
+            res.types.Video(
-                template="videos.html",
+                title=extract_text(eval_xpath(item, ".//a[contains(@class, 'info_title')]")) or "",
-                title=extract_text(eval_xpath(item, ".//a[contains(@class, 'info_title')]")),
                 url=url,
                 thumbnail=thumbnail,
                 length=length,

searx/engines/odysee.py

@@ -4,14 +4,19 @@
 .. _Odysee: https://github.com/OdyseeTeam/odysee-frontend
 """
 
-from datetime import datetime
+import typing as t
+from datetime import datetime, timedelta
 from urllib.parse import urlencode
 
 import babel
 
 from searx.enginelib.traits import EngineTraits
 from searx.locales import language_tag
-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
 
 # Engine metadata
 about = {
@@ -34,7 +39,7 @@ categories = ["videos"]
 base_url = "https://lighthouse.odysee.tv/search"
 
 
-def request(query, params):
+def request(query: str, params: "OnlineParams"):
     time_range_dict = {
         "day": "today",
         "week": "thisweek",
@@ -59,22 +64,17 @@ def request(query, params):
         query_params["time_filter"] = time_range_dict[params["time_range"]]
 
     params["url"] = f"{base_url}?{urlencode(query_params)}"
-    return params
 
 
-def response(resp):
+def response(resp: "SXNG_Response") -> EngineResults:
+    res = EngineResults()
     data = resp.json()
-    results = []
 
     for item in data:
         name = item["name"]
         claim_id = item["claimId"]
-        title = item["title"]
         thumbnail_url = item["thumbnail_url"]
-        description = item["description"] or ""
-        channel = item["channel"]
         release_time = item["release_time"]
-        duration = item["duration"]
 
         release_date = datetime.fromisoformat(release_time.split("T")[0])
         formatted_date = datetime.fromtimestamp(release_date.timestamp())
@@ -82,23 +82,21 @@ def response(resp):
         url = f"https://odysee.com/{name}:{claim_id}"
         iframe_url = f"https://odysee.com/$/embed/{name}:{claim_id}"
         odysee_thumbnail = f"https://thumbnails.odycdn.com/optimize/s:390:0/quality:85/plain/{thumbnail_url}"
-        formatted_duration = format_duration(duration)
 
-        results.append(
+        res.add(
-            {
+            res.types.Video(
-                "title": title,
+                title=item["title"],
-                "url": url,
+                url=url,
-                "content": description,
+                content=item["description"] or "",
-                "author": channel,
+                author=item["channel"],
-                "publishedDate": formatted_date,
+                publishedDate=formatted_date,
-                "length": formatted_duration,
+                length=timedelta(seconds=item["duration"]),
-                "thumbnail": odysee_thumbnail,
+                thumbnail=odysee_thumbnail,
-                "iframe_src": iframe_url,
+                iframe_src=iframe_url,
-                "template": "videos.html",
+            )
-            }
         )
 
-    return results
+    return res
 
 
 def fetch_traits(engine_traits: EngineTraits):

searx/engines/privacywall.py

@@ -146,8 +146,7 @@ def _video_results(doc: "ElementBase") -> EngineResults:
             thumbnail = _extract_thumbnail_url(extr(thumbnail_style, ":url(", ")"))
 
         res.add(
-            res.types.LegacyResult(
+            res.types.Video(
-                template="videos.html",
                 url=url,
                 title=extract_text(eval_xpath(result, ".//h2[contains(@class, 'video-card-title')]")) or "",
                 content=extract_text(eval_xpath(result, ".//p")) or "",

searx/engines/qwant.py

@@ -292,7 +292,7 @@ def response(resp: "SXNG_Response") -> EngineResults:
                 thumbnail = thumbnail.replace("https://s2.qwant.com", "https://s1.qwant.com", 1)
 
                 res.add(
-                    res.types.LegacyResult(
+                    res.types.Video(
                         title=title,
                         url=res_url,
                         content=content,

searx/engines/startpagina.py

@@ -7,10 +7,9 @@ results from Google.
 
 import typing as t
 from urllib.parse import urlencode
-
+import datetime
 from dateutil import parser
 
-from searx.utils import format_duration
 from searx.result_types import EngineResults
 
 if t.TYPE_CHECKING:
@@ -49,7 +48,7 @@ def request(query: str, params: "OnlineParams") -> None:
     params["url"] = f"{api_url}/api/v2/search/{startpagina_categ}/?{urlencode(args)}"
 
 
-def response(resp: "SXNG_Response"):
+def response(resp: "SXNG_Response") -> EngineResults:
     res = EngineResults()
 
     json_resp = resp.json()
@@ -81,13 +80,12 @@ def response(resp: "SXNG_Response"):
             )
         elif startpagina_categ == "videos":
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    template="videos.html",
                     url=result["original_url"],
                     title=result["title"],
                     content=result["description"],
                     thumbnail=result["video"]["thumbnail_url"],
-                    length=format_duration(result["video"]["duration"]),
+                    length=datetime.timedelta(seconds=result["video"]["duration"]),
                 )
             )
         elif startpagina_categ == "images":

searx/engines/swisscows.py

@@ -16,7 +16,7 @@ from urllib.parse import urlencode
 
 from babel.core import get_global
 
-from searx.result_types import EngineResults, LegacyResult  # pyright: ignore[reportPrivateLocalImportUsage]
+from searx.result_types import EngineResults, Video
 from searx.utils import humanize_number, html_to_text
 
 if t.TYPE_CHECKING:
@@ -215,28 +215,24 @@ def request(query: str, params: "OnlineParams") -> None:
     params["url"] = f"{base_url}{base_path}?{urlencode(args)}"
 
 
-def _video_result(result: dict[str, str]) -> LegacyResult:
+def _video_result(result: dict[str, str]) -> Video:
     published_date = None
     if result.get("datePublished"):
         published_date = datetime.fromisoformat(result["datePublished"])
 
-    view_count = None
+    view_count = ""
     if result.get("viewCount"):
         view_count = humanize_number(result["viewCount"])  # pyright: ignore[reportArgumentType]
 
-    return LegacyResult(
+    return Video(
-        {
+        url=result["url"],
-            "template": "videos.html",
+        title=html_to_text(result.get("title") or result["name"]),
-            "url": result["url"],
+        content=result["description"],
-            "title": html_to_text(result.get("title") or result["name"]),
+        thumbnail=result.get("thumbnailUrl") or result.get("thumbnail", {}).get("url"),
-            "content": result["description"],
+        length=result.get("duration"),
-            "thumbnail": result.get("thumbnailUrl")
+        iframe_src=result.get("embedUrl"),
-            or result.get("thumbnail", {}).get("url"),  # pyright: ignore[reportAttributeAccessIssue]
+        publishedDate=published_date,
-            "length": result.get("duration"),
+        views=view_count,
-            "iframe_src": result.get("embedUrl"),
-            "publishedDate": published_date,
-            "views": view_count,
-        }
     )
 
 
@@ -272,14 +268,11 @@ def response(resp: "SXNG_Response") -> EngineResults:
                 res.add(_video_result(video))
         elif result["type"] == "ImageObject":
             res.add(
-                res.types.LegacyResult(
+                res.types.Image(
-                    {
+                    url=result["url"],
-                        "template": "images.html",
+                    thumbnail_src=result["thumbnail"]["url"],
-                        "url": result["url"],
+                    img_src=result["contentUrl"],
-                        "thumbnail_src": result["thumbnail"]["url"],
+                    title=result["name"],
-                        "img_src": result["contentUrl"],
-                        "title": result["name"],
-                    }
                 )
             )
         elif result["type"] == "video":

searx/engines/tagesschau.py

@@ -20,6 +20,8 @@ from datetime import datetime
 from urllib.parse import urlencode
 import re
 
+from searx.result_types import Video
+
 about = {
     'website': "https://tagesschau.de",
     'wikidata_id': "Q703907",
@@ -87,7 +89,7 @@ def _story(item):
     }
 
 
-def _video(item):
+def _video(item) -> Video:
     streams = item['streams']
     video_url = streams.get('h264s') or streams.get('h264m') or streams.get('h264l') or streams.get('h264xl')
     title = item['title']
@@ -99,12 +101,11 @@ def _video(item):
     # sometimes, only adaptive m3u8 streams are available, so video_url is None
     url = video_url or f"{base_url}/multimedia/video/{item['sophoraId']}.html"
 
-    return {
+    return Video(
-        'template': 'videos.html',
+        title=title,
-        'title': title,
+        thumbnail=item.get('teaserImage', {}).get('imageVariants', {}).get('16x9-256'),
-        'thumbnail': item.get('teaserImage', {}).get('imageVariants', {}).get('16x9-256'),
+        publishedDate=datetime.fromisoformat(item['date'][:19]),
-        'publishedDate': datetime.fromisoformat(item['date'][:19]),
+        content=item.get('firstSentence', ''),
-        'content': item.get('firstSentence', ''),
+        iframe_src=video_url,
-        'iframe_src': video_url,
+        url=url,
-        'url': url,
+    )
-    }

searx/engines/tonline.py

@@ -119,8 +119,7 @@ def _video_results(doc: ElementType, res: EngineResults):
             continue
         title_parts: list[ElementType] = eval_xpath(result, ".//a[starts-with(@class, 'tMMReshl')]")
         res.add(
-            res.types.LegacyResult(
+            res.types.Video(
-                template="videos.html",
                 url=url,
                 title=" - ".join(extract_text(part) or "" for part in title_parts),
                 thumbnail=extract_text(eval_xpath(result, ".//img/@src") or "") or "",

searx/engines/tusksearch.py

@@ -143,8 +143,7 @@ def response(resp: "SXNG_Response"):
                 pass
 
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    template="videos.html",
                     url=result["url"],
                     title=html_to_text(result["title"]),
                     content=html_to_text(result["description"]),

searx/engines/vuhuv.py

@@ -85,8 +85,7 @@ def _video_results(doc: "ElementBase") -> EngineResults:
     for result in eval_xpath_list(doc, "//div[contains(@class, 'item video')]"):
         (
             res.add(
-                res.types.MainResult(
+                res.types.Video(
-                    template="videos.html",
                     url=extract_text(eval_xpath(result, "./a/@href")) or "",
                     title=extract_text(eval_xpath(result, "./a/@title")) or "",
                     content=extract_text(eval_xpath(result, ".//div[contains(@class, 'abaslik')]")) or "",

searx/engines/wikicommons.py

@@ -181,8 +181,7 @@ def response(resp: "SXNG_Response") -> EngineResults:
 
         if wc_search_type == "video":
             res.add(
-                res.types.LegacyResult(
+                res.types.Video(
-                    template="videos.html",
                     title=title,
                     url=url,
                     content=content,

searx/utils.py

@@ -2,7 +2,6 @@
 """Utility functions for the engines"""
 
 from hashlib import pbkdf2_hmac
-import time
 
 import re
 import importlib
@@ -816,15 +815,6 @@ def parse_duration_string(duration_str: str) -> timedelta | None:
     return None
 
 
-# Format the video duration
-def format_duration(duration: str | int) -> str:
-    seconds = int(duration)
-    length = time.gmtime(seconds)
-    if length.tm_hour:
-        return time.strftime("%H:%M:%S", length)
-    return time.strftime("%M:%S", length)
-
-
 def _array_startswith(arr: bytes, prefix: bytes) -> bool:
     return arr[: len(prefix)] == prefix