searxng/searx/engines/deviantart.py

89 lines
2.5 KiB
Python
Raw Normal View History

# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Deviantart (Images)
"""
2014-09-02 14:48:18 +00:00
import urllib.parse
from lxml import html
2013-10-20 09:12:10 +00:00
from searx.utils import extract_text, eval_xpath, eval_xpath_list
# about
about = {
"website": 'https://www.deviantart.com/',
"wikidata_id": 'Q46523',
"official_api_documentation": 'https://www.deviantart.com/developers/',
"use_official_api": False,
"require_api_key": False,
"results": 'HTML',
}
2014-09-02 14:48:18 +00:00
# engine dependent config
2013-10-20 09:12:10 +00:00
categories = ['images']
2014-09-02 14:48:18 +00:00
paging = True
2014-01-29 23:09:47 +00:00
# search-url
base_url = 'https://www.deviantart.com'
2014-01-20 01:31:20 +00:00
results_xpath = '//div[@class="_2pZkk"]/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="_1OGeq"]/@href)[last()]'
def request(query, params):
# https://www.deviantart.com/search?q=foo
2014-09-02 14:48:18 +00:00
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})}"
2013-10-20 09:12:10 +00:00
return params
2013-10-20 09:12:10 +00:00
2013-10-20 09:12:10 +00:00
def response(resp):
2014-09-02 14:48:18 +00:00
results = []
2013-10-20 09:12:10 +00:00
dom = html.fromstring(resp.text)
2015-01-17 18:24:35 +00:00
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
img_src = extract_text(eval_xpath(result, img_src_xpath))
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)),
}
)
nextpage_url = extract_text(eval_xpath(dom, cursor_xpath))
if nextpage_url:
results.append(
{
'engine_data': nextpage_url.replace("http://", "https://"),
'key': 'nextpage',
}
)
2013-10-20 09:12:10 +00:00
return results