Merge pull request #2286 from return42/fix-deviantart

[refactor] deviantart - improve results and clean up source code
This commit is contained in:
Alexandre Flament 2020-11-14 17:31:39 +01:00 committed by GitHub
commit 17b48ff6e8
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 38 additions and 42 deletions

View file

@ -217,6 +217,7 @@ test.pylint: pyenvinstall
searx/preferences.py \
searx/testing.py \
searx/engines/gigablast.py \
searx/engines/deviantart.py \
)
# ignored rules:
@ -236,7 +237,7 @@ test.sh:
test.pep8: pyenvinstall
@echo "TEST pycodestyle (formerly pep8)"
$(Q)$(PY_ENV_ACT); pycodestyle --exclude='searx/static, searx/languages.py, searx/engines/gigablast.py' \
$(Q)$(PY_ENV_ACT); pycodestyle --exclude='searx/static, searx/languages.py, searx/engines/gigablast.py, searx/engines/deviantart.py' \
--max-line-length=120 --ignore "E117,E252,E402,E722,E741,W503,W504,W605" searx tests
test.unit: pyenvinstall

View file

@ -7,75 +7,70 @@
@using-api no (TODO, rewrite to api)
@results HTML
@stable no (HTML can change)
@parse url, title, thumbnail_src, img_src
@parse url, title, img_src
@todo rewrite to api
"""
# pylint: disable=missing-function-docstring
from lxml import html
import re
from urllib.parse import urlencode
from lxml import html
# engine dependent config
categories = ['images']
paging = True
time_range_support = True
time_range_dict = {
'day': 'popular-24-hours',
'week': 'popular-1-week',
'month': 'popular-1-month',
'year': 'most-recent',
}
# search-url
base_url = 'https://www.deviantart.com/'
search_url = base_url + 'search?page={page}&{query}'
time_range_url = '&order={range}'
base_url = 'https://www.deviantart.com'
time_range_dict = {'day': 11,
'week': 14,
'month': 15}
# do search-request
def request(query, params):
if params['time_range'] and params['time_range'] not in time_range_dict:
return params
params['url'] = search_url.format(page=params['pageno'],
query=urlencode({'q': query}))
# https://www.deviantart.com/search/deviations?page=5&q=foo
query = {
'page' : params['pageno'],
'q' : query,
}
if params['time_range'] in time_range_dict:
params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
query['order'] = time_range_dict[params['time_range']]
params['url'] = base_url + '/search/deviations?' + urlencode(query)
return params
# get response from search-request
def response(resp):
results = []
# return empty array if a redirection code is returned
if resp.status_code == 302:
return []
results = []
dom = html.fromstring(resp.text)
# parse results
for row in dom.xpath('//div[contains(@data-hook, "content_row")]'):
for result in row.xpath('./div'):
link = result.xpath('.//a[@data-hook="deviation_link"]')[0]
url = link.attrib.get('href')
title = link.attrib.get('title')
thumbnail_src = result.xpath('.//img')[0].attrib.get('src')
img_src = thumbnail_src
# 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)
a_tag = result.xpath('.//a[@data-hook="deviation_link"]')[0]
noscript_tag = a_tag.xpath('.//noscript')
url = re.sub(r"http://(.*)\.deviantart\.com/", "https://\\1.deviantart.com/", url)
if noscript_tag:
img_tag = noscript_tag[0].xpath('.//img')
else:
img_tag = a_tag.xpath('.//img')
if not img_tag:
continue
img_tag = img_tag[0]
# append result
results.append({'url': url,
'title': title,
'img_src': img_src,
'thumbnail_src': thumbnail_src,
'template': 'images.html'})
results.append({
'template': 'images.html',
'url': a_tag.attrib.get('href'),
'img_src': img_tag.attrib.get('src'),
'title': img_tag.attrib.get('alt'),
})
# return results
return results