searxngRebrandZaclys/searx/engines/deviantart.py

74 lines
2.1 KiB
Python
Raw Normal View History

2014-09-02 14:48:18 +00:00
## Deviantart (Images)
#
2014-09-02 14:48:18 +00:00
# @website https://www.deviantart.com/
# @provide-api yes (https://www.deviantart.com/developers/) (RSS)
#
2014-09-02 14:48:18 +00:00
# @using-api no (TODO, rewrite to api)
# @results HTML
# @stable no (HTML can change)
# @parse url, title, thumbnail_src, img_src
2014-09-02 14:48:18 +00:00
#
# @todo rewrite to api
2013-10-20 09:12:10 +00:00
from urllib import urlencode
from urlparse import urljoin
2014-02-05 19:24:31 +00:00
from lxml import html
import re
2015-01-29 00:13:33 +00:00
from searx.engines.xpath import extract_text
2013-10-20 09:12:10 +00:00
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
2013-10-20 09:12:10 +00:00
2014-09-02 14:48:18 +00:00
# search-url
2013-10-20 09:12:10 +00:00
base_url = 'https://www.deviantart.com/'
search_url = base_url+'browse/all/?offset={offset}&{query}'
2014-01-29 23:09:47 +00:00
2014-01-20 01:31:20 +00:00
2014-09-02 14:48:18 +00:00
# do search-request
2013-10-20 09:12:10 +00:00
def request(query, params):
2014-01-29 23:09:47 +00:00
offset = (params['pageno'] - 1) * 24
2014-09-02 14:48:18 +00:00
2014-01-29 23:09:47 +00:00
params['url'] = search_url.format(offset=offset,
query=urlencode({'q': query}))
2014-09-02 14:48:18 +00:00
2013-10-20 09:12:10 +00:00
return params
2014-09-02 14:48:18 +00:00
# get response from search-request
2013-10-20 09:12:10 +00:00
def response(resp):
results = []
2014-09-02 14:48:18 +00:00
# return empty array if a redirection code is returned
2013-10-20 09:12:10 +00:00
if resp.status_code == 302:
2014-09-02 14:48:18 +00:00
return []
2013-10-20 09:12:10 +00:00
dom = html.fromstring(resp.text)
2015-01-17 18:24:35 +00:00
regex = re.compile('\/200H\/')
2014-09-02 14:48:18 +00:00
# parse results
2013-10-20 09:12:10 +00:00
for result in dom.xpath('//div[contains(@class, "tt-a tt-fh")]'):
link = result.xpath('.//a[contains(@class, "thumb")]')[0]
url = urljoin(base_url, link.attrib.get('href'))
2015-01-29 00:13:33 +00:00
title_links = result.xpath('.//span[@class="details"]//a[contains(@class, "t")]')
title = extract_text(title_links[0])
thumbnail_src = link.xpath('.//img')[0].attrib.get('src')
img_src = regex.sub('/', thumbnail_src)
2014-09-02 14:48:18 +00:00
# http to https, remove domain sharding
thumbnail_src = re.sub(r"https?://(th|fc)\d+.", "https://th01.", thumbnail_src)
thumbnail_src = re.sub(r"http://", "https://", thumbnail_src)
url = re.sub(r"http://(.*)\.deviantart\.com/", "https://\\1.deviantart.com/", url)
2014-09-02 14:48:18 +00:00
# append result
2014-01-20 01:31:20 +00:00
results.append({'url': url,
'title': title,
'img_src': img_src,
'thumbnail_src': thumbnail_src,
2014-01-20 01:31:20 +00:00
'template': 'images.html'})
2014-09-02 14:48:18 +00:00
# return results
2013-10-20 09:12:10 +00:00
return results