Ponysearch/searx/engines/duden.py

79 lines
2.2 KiB
Python
Raw Normal View History

2018-08-18 19:24:02 +02:00
"""
Duden
@website https://www.duden.de
@provide-api no
@using-api no
@results HTML (using search portal)
@stable no (HTML can change)
@parse url, title, content
"""
from lxml import html, etree
import re
from urllib.parse import quote, urljoin
from searx.utils import extract_text, eval_xpath
2018-08-18 19:24:02 +02:00
from searx import logger
categories = ['general']
paging = True
language_support = False
# search-url
base_url = 'https://www.duden.de/'
2019-07-25 08:17:45 +02:00
search_url = base_url + 'suchen/dudenonline/{query}?search_api_fulltext=&page={offset}'
2018-08-18 19:24:02 +02:00
def request(query, params):
'''pre-request callback
params<dict>:
method : POST/GET
headers : {}
data : {} # if method == POST
url : ''
category: 'search category'
pageno : 1 # number of the requested page
'''
offset = (params['pageno'] - 1)
2019-07-25 08:17:45 +02:00
if offset == 0:
search_url_fmt = base_url + 'suchen/dudenonline/{query}'
params['url'] = search_url_fmt.format(query=quote(query))
else:
params['url'] = search_url.format(offset=offset, query=quote(query))
2018-08-18 19:24:02 +02:00
return params
def response(resp):
'''post-response callback
resp: requests response object
'''
results = []
dom = html.fromstring(resp.text)
try:
number_of_results_string =\
re.sub('[^0-9]', '',
eval_xpath(dom, '//a[@class="active" and contains(@href,"/suchen/dudenonline")]/span/text()')[0])
2018-08-18 19:24:02 +02:00
results.append({'number_of_results': int(number_of_results_string)})
except:
logger.debug("Couldn't read number of results.")
for result in eval_xpath(dom, '//section[not(contains(@class, "essay"))]'):
2018-08-18 19:24:02 +02:00
try:
url = eval_xpath(result, './/h2/a')[0].get('href')
2019-07-25 08:17:45 +02:00
url = urljoin(base_url, url)
title = eval_xpath(result, 'string(.//h2/a)').strip()
content = extract_text(eval_xpath(result, './/p'))
2018-08-18 19:24:02 +02:00
# append result
results.append({'url': url,
'title': title,
'content': content})
except:
logger.debug('result parse error in:\n%s', etree.tostring(result, pretty_print=True))
continue
return results