mirror of
https://github.com/searxng/searxng.git
synced 2026-08-17 20:51:25 +00:00
[fix] engines: deviantart access denied
This commit is contained in:
@@ -1,11 +1,18 @@
|
||||
# SPDX-License-Identifier: AGPL-3.0-or-later
|
||||
"""Deviantart (Images)"""
|
||||
|
||||
import typing as t
|
||||
|
||||
import urllib.parse
|
||||
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 searx.extended_types import SXNG_Response
|
||||
from searx.search.processors import OnlineParams
|
||||
|
||||
# about
|
||||
about = {
|
||||
"website": 'https://www.deviantart.com/',
|
||||
@@ -23,63 +30,62 @@ paging = True
|
||||
# search-url
|
||||
base_url = 'https://www.deviantart.com'
|
||||
|
||||
results_xpath = '//div[@class="V_S0t_"]/div/div/a'
|
||||
url_xpath = './@href'
|
||||
thumbnail_src_xpath = './div/img/@src'
|
||||
img_src_xpath = './div/img/@srcset'
|
||||
title_xpath = './@aria-label'
|
||||
premium_xpath = '../div/div/div/text()'
|
||||
premium_keytext = 'Watch the artist to view this deviation'
|
||||
cursor_xpath = '(//a[@class="vQ2brP"]/@href)[last()]'
|
||||
results_xpath = '//div[@data-testid="content_row"]//a[.//*[@data-testid="thumb"]]'
|
||||
img_src_xpath = './/img/@srcset'
|
||||
thumbnail_src_xpath = './/img/@src'
|
||||
author_xpath = './/*[@property="schema:name"]/@content'
|
||||
cursor_xpath = '//a[contains(@href, "cursor=") and contains(., "Next")]/@href'
|
||||
|
||||
|
||||
def request(query, params):
|
||||
|
||||
def request(query: str, params: "OnlineParams"):
|
||||
# https://www.deviantart.com/search?q=foo
|
||||
|
||||
nextpage_url = params['engine_data'].get('nextpage')
|
||||
# don't use nextpage when user selected to jump back to page 1
|
||||
if params['pageno'] > 1 and nextpage_url is not None:
|
||||
params['url'] = nextpage_url
|
||||
else:
|
||||
params['url'] = f"{base_url}/search?{urllib.parse.urlencode({'q': query})}"
|
||||
args = {'q': query}
|
||||
if params['pageno'] > 1:
|
||||
cursor = params['engine_data'].get('cursor')
|
||||
if cursor:
|
||||
args['cursor'] = cursor
|
||||
|
||||
return params
|
||||
params['url'] = f"{base_url}/search?{urllib.parse.urlencode(args)}"
|
||||
|
||||
|
||||
def response(resp):
|
||||
def response(resp: "SXNG_Response") -> EngineResults:
|
||||
|
||||
results = []
|
||||
res = EngineResults()
|
||||
dom = html.fromstring(resp.text)
|
||||
|
||||
for result in eval_xpath_list(dom, results_xpath):
|
||||
# skip images that are blurred
|
||||
_text = extract_text(eval_xpath(result, premium_xpath))
|
||||
if _text and premium_keytext in _text:
|
||||
continue
|
||||
thumbnail_src = extract_text(eval_xpath(result, thumbnail_src_xpath))
|
||||
img_src = extract_text(eval_xpath(result, img_src_xpath))
|
||||
# mature locked thumbs have blur transform (blur_15, blur_30 etc..)
|
||||
if ',blur_' in f'{thumbnail_src}{img_src}':
|
||||
continue
|
||||
if img_src:
|
||||
img_src = img_src.split(' ')[0]
|
||||
parsed_url = urllib.parse.urlparse(img_src)
|
||||
img_src = parsed_url._replace(path=parsed_url.path.split('/v1')[0]).geturl()
|
||||
|
||||
results.append(
|
||||
{
|
||||
'template': 'images.html',
|
||||
'url': extract_text(eval_xpath(result, url_xpath)),
|
||||
'img_src': img_src,
|
||||
'thumbnail_src': extract_text(eval_xpath(result, thumbnail_src_xpath)),
|
||||
'title': extract_text(eval_xpath(result, title_xpath)),
|
||||
}
|
||||
author = extract_text(eval_xpath(result, author_xpath))
|
||||
|
||||
res.add(
|
||||
res.types.Image(
|
||||
template='images.html',
|
||||
url=result.get('href'),
|
||||
img_src=img_src or "",
|
||||
thumbnail_src=thumbnail_src or "",
|
||||
title=result.get('aria-label'),
|
||||
author=author or "",
|
||||
)
|
||||
)
|
||||
|
||||
nextpage_url = extract_text(eval_xpath(dom, cursor_xpath))
|
||||
if nextpage_url:
|
||||
results.append(
|
||||
{
|
||||
'engine_data': nextpage_url.replace("http://", "https://"),
|
||||
'key': 'nextpage',
|
||||
}
|
||||
cursor = urllib.parse.parse_qs(urllib.parse.urlparse(nextpage_url or '').query).get('cursor', [None])[0]
|
||||
if cursor:
|
||||
res.add(
|
||||
res.types.LegacyResult(
|
||||
engine_data=cursor,
|
||||
key='cursor',
|
||||
)
|
||||
)
|
||||
|
||||
return results
|
||||
return res
|
||||
|
||||
@@ -761,6 +761,7 @@ engines:
|
||||
engine: deviantart
|
||||
shortcut: da
|
||||
timeout: 3.0
|
||||
inactive: true
|
||||
|
||||
- name: devicons
|
||||
engine: devicons
|
||||
|
||||
Reference in New Issue
Block a user