[mod] brave-based engines: reuse result parsing methods (#6688)

Many engines use the JSON results from Brave (e.g. Brave, BraveAPI, Tusksearch
and more), so the result parsing logic is always the same. This PR deduplicates
this code.
This commit is contained in:
Bnyro
2026-09-10 12:32:22 +02:00
committed by Markus Heiser
parent 2ed96e6fcf
commit 4d9c982cda
3 changed files with 36 additions and 107 deletions

View File

@@ -261,18 +261,18 @@ def response(resp: "SXNG_Response") -> EngineResults:
match brave_category: match brave_category:
case "search" | "goggles": case "search" | "goggles":
return _parse_results(_parse_search_result, resp) return _parse_results(parse_search_result, resp)
case "news": case "news":
return _parse_results(_parse_news_result, resp) return _parse_results(parse_news_result, resp)
case "images": case "images":
return _parse_results(_parse_image_result, resp) return _parse_results(parse_image_result, resp)
case "videos": case "videos":
return _parse_results(_parse_video_result, resp) return _parse_results(parse_video_result, resp)
case _: case _:
raise ValueError(f"Unsupported brave category: {brave_category}") # pyright: ignore[reportUnreachable] raise ValueError(f"Unsupported brave category: {brave_category}") # pyright: ignore[reportUnreachable]
def _parse_search_result(result: dict[str, t.Any]) -> MainResult: def parse_search_result(result: dict[str, t.Any]) -> MainResult:
thumbnail: dict[str, t.Any] = result.get("thumbnail", {}) thumbnail: dict[str, t.Any] = result.get("thumbnail", {})
return MainResult( return MainResult(
template="default.html", template="default.html",
@@ -291,7 +291,7 @@ def _parse_secondary_items(json_data: dict[str, t.Any], results: EngineResults):
videos_resp: dict[str, t.Any] = body_resp.get("videos", {}) videos_resp: dict[str, t.Any] = body_resp.get("videos", {})
if videos_resp and "results" in videos_resp: if videos_resp and "results" in videos_resp:
for result in videos_resp.get("results", []): for result in videos_resp.get("results", []):
results.add(_parse_video_result(result)) results.add(parse_video_result(result))
# related queries -> suggestion # related queries -> suggestion
query: dict[str, t.Any] = body_resp.get("query", {}) query: dict[str, t.Any] = body_resp.get("query", {})
if query and "related_queries" in query: if query and "related_queries" in query:
@@ -300,7 +300,7 @@ def _parse_secondary_items(json_data: dict[str, t.Any], results: EngineResults):
results.add(results.types.LegacyResult(suggestion=suggestion)) results.add(results.types.LegacyResult(suggestion=suggestion))
def _parse_news_result(result: dict[str, t.Any]) -> MainResult: def parse_news_result(result: dict[str, t.Any]) -> MainResult:
thumbnail: dict[str, t.Any] = result.get("thumbnail", {}) thumbnail: dict[str, t.Any] = result.get("thumbnail", {})
return MainResult( return MainResult(
title=result.get("title", ""), title=result.get("title", ""),
@@ -312,7 +312,7 @@ def _parse_news_result(result: dict[str, t.Any]) -> MainResult:
) )
def _parse_image_result(result: dict[str, t.Any]) -> Image: def parse_image_result(result: dict[str, t.Any]) -> Image:
properties: dict[str, t.Any] = result.get("properties", {}) properties: dict[str, t.Any] = result.get("properties", {})
thumbnail: dict[str, t.Any] = result.get("thumbnail", {}) thumbnail: dict[str, t.Any] = result.get("thumbnail", {})
width, height = properties.get("width"), properties.get("height") width, height = properties.get("width"), properties.get("height")
@@ -327,7 +327,7 @@ def _parse_image_result(result: dict[str, t.Any]) -> Image:
) )
def _parse_video_result(result: dict[str, t.Any]) -> MainResult: def parse_video_result(result: dict[str, t.Any]) -> Video:
video: dict[str, t.Any] = result.get("video", {}) video: dict[str, t.Any] = result.get("video", {})
thumbnail: dict[str, t.Any] = result.get("thumbnail", {}) thumbnail: dict[str, t.Any] = result.get("thumbnail", {})

View File

@@ -27,11 +27,10 @@ The API supports paging and time filters.
import typing as t import typing as t
from urllib.parse import urlencode from urllib.parse import urlencode
from dateutil import parser
from searx.engines.brave import parse_video_result
from searx.exceptions import SearxEngineAPIException from searx.exceptions import SearxEngineAPIException
from searx.result_types import EngineResults from searx.result_types import EngineResults
from searx.utils import html_to_text
if t.TYPE_CHECKING: if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response from searx.extended_types import SXNG_Response
@@ -94,43 +93,13 @@ def request(query: str, params: "OnlineParams") -> None:
params["headers"]["Accept"] = "application/json" params["headers"]["Accept"] = "application/json"
def _extract_published_date(published_date_raw: str):
"""Extract and parse the published date from the API response.
Args:
published_date_raw: Raw date string from the API
Returns:
Parsed datetime object or None if parsing fails
"""
if not published_date_raw:
return None
try:
return parser.parse(published_date_raw)
except parser.ParserError:
return None
def response(resp: "SXNG_Response") -> EngineResults: def response(resp: "SXNG_Response") -> EngineResults:
"""Process the API response and return results.""" """Process the API response and return results."""
res = EngineResults()
data = resp.json() data = resp.json()
for result in (data.get("web") or {}).get("results", []): res = EngineResults()
thumbnail_obj = result.get("thumbnail") results_json = (data.get("web") or {}).get("results", [])
thumbnail = "" for result in results_json:
if thumbnail_obj and not thumbnail_obj.get("logo", False): res.add(parse_video_result(result))
thumbnail = thumbnail_obj.get("src") or ""
res.add(
res.types.MainResult(
url=result["url"],
title=html_to_text(result["title"]),
content=html_to_text(result.get("description", "")),
publishedDate=_extract_published_date(result.get("age")),
thumbnail=thumbnail,
),
)
return res return res

View File

@@ -9,12 +9,11 @@ from json import loads
import random import random
import typing as t import typing as t
from urllib.parse import urlencode from urllib.parse import urlencode
from dateutil import parser
from searx.result_types import EngineResults
from searx.engines.brave import parse_image_result, parse_news_result, parse_search_result, parse_video_result
from searx.exceptions import SearxEngineAPIException from searx.exceptions import SearxEngineAPIException
from searx.network import get from searx.network import get
from searx.utils import html_to_text
from searx.result_types import EngineResults
if t.TYPE_CHECKING: if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response from searx.extended_types import SXNG_Response
@@ -32,7 +31,8 @@ about = {
paging = True paging = True
categories = ["general"] categories = ["general"]
tusk_categ = "web" TuskCategType = t.Literal["web", "images", "videos", "news"]
tusk_categ: TuskCategType = "web"
"""Category to search in. Can be either "web", "images", "videos" or "news".""" """Category to search in. Can be either "web", "images", "videos" or "news"."""
@@ -40,7 +40,7 @@ api_url = "https://api.tusksearch.com"
def setup(_: dict[str, t.Any]) -> bool | None: def setup(_: dict[str, t.Any]) -> bool | None:
if tusk_categ not in ("web", "images", "videos", "news"): if tusk_categ not in t.get_args(TuskCategType):
raise ValueError("invalid search type: %s" % tusk_categ) raise ValueError("invalid search type: %s" % tusk_categ)
@@ -103,64 +103,24 @@ def request(query: str, params: "OnlineParams") -> None:
def response(resp: "SXNG_Response"): def response(resp: "SXNG_Response"):
res = EngineResults()
json_resp = resp.json()["results"] json_resp = resp.json()["results"]
if tusk_categ == "web": res = EngineResults()
for result in (json_resp.get("web") or {}).get("results", []): match tusk_categ:
res.add( case "web":
res.types.MainResult( results = (json_resp.get("web") or {}).get("results", [])
url=result["url"], for result in results:
title=html_to_text(result["title"]), res.add(parse_search_result(result))
content=html_to_text(result["description"]), case "news":
thumbnail=(result["thumbnail"] or {}).get("src") or "", results = (json_resp.get("news") or {}).get("results", [])
) for result in results:
) res.add(parse_news_result(result))
elif tusk_categ == "news": case "videos":
for result in (json_resp.get("news") or {}).get("results", []): results = (json_resp.get("videos") or {}).get("results", [])
publishedDate = None for result in results:
try: res.add(parse_video_result(result))
publishedDate = parser.parse(result["age"]) case "images":
except parser.ParserError: for result in json_resp:
pass res.add(parse_image_result(result))
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.Video(
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 return res