forked from Ponysearch/Ponysearch
[fix] fix duden engine (#1594)
This commit is contained in:
parent
87baa74a86
commit
3b1122c5fa
2 changed files with 28 additions and 19 deletions
|
@ -11,7 +11,7 @@
|
||||||
from lxml import html, etree
|
from lxml import html, etree
|
||||||
import re
|
import re
|
||||||
from searx.engines.xpath import extract_text
|
from searx.engines.xpath import extract_text
|
||||||
from searx.url_utils import quote
|
from searx.url_utils import quote, urljoin
|
||||||
from searx import logger
|
from searx import logger
|
||||||
|
|
||||||
categories = ['general']
|
categories = ['general']
|
||||||
|
@ -20,7 +20,7 @@ language_support = False
|
||||||
|
|
||||||
# search-url
|
# search-url
|
||||||
base_url = 'https://www.duden.de/'
|
base_url = 'https://www.duden.de/'
|
||||||
search_url = base_url + 'suchen/dudenonline/{query}?page={offset}'
|
search_url = base_url + 'suchen/dudenonline/{query}?search_api_fulltext=&page={offset}'
|
||||||
|
|
||||||
|
|
||||||
def request(query, params):
|
def request(query, params):
|
||||||
|
@ -35,6 +35,10 @@ def request(query, params):
|
||||||
'''
|
'''
|
||||||
|
|
||||||
offset = (params['pageno'] - 1)
|
offset = (params['pageno'] - 1)
|
||||||
|
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))
|
params['url'] = search_url.format(offset=offset, query=quote(query))
|
||||||
return params
|
return params
|
||||||
|
|
||||||
|
@ -58,12 +62,11 @@ def response(resp):
|
||||||
logger.debug("Couldn't read number of results.")
|
logger.debug("Couldn't read number of results.")
|
||||||
pass
|
pass
|
||||||
|
|
||||||
for result in dom.xpath('//section[@class="wide" and not(contains(@style,"overflow:hidden"))]'):
|
for result in dom.xpath('//section[not(contains(@class, "essay"))]'):
|
||||||
try:
|
try:
|
||||||
logger.debug("running for %s" % str(result))
|
url = result.xpath('.//h2/a')[0].get('href')
|
||||||
link = result.xpath('.//h2/a')[0]
|
url = urljoin(base_url, url)
|
||||||
url = link.attrib.get('href')
|
title = result.xpath('string(.//h2/a)').strip()
|
||||||
title = result.xpath('string(.//h2/a)')
|
|
||||||
content = extract_text(result.xpath('.//p'))
|
content = extract_text(result.xpath('.//p'))
|
||||||
# append result
|
# append result
|
||||||
results.append({'url': url,
|
results.append({'url': url,
|
||||||
|
|
|
@ -10,24 +10,30 @@ class TestDudenEngine(SearxTestCase):
|
||||||
def test_request(self):
|
def test_request(self):
|
||||||
query = 'Haus'
|
query = 'Haus'
|
||||||
dic = defaultdict(dict)
|
dic = defaultdict(dict)
|
||||||
dic['pageno'] = 1
|
data = [
|
||||||
|
[1, 'https://www.duden.de/suchen/dudenonline/Haus'],
|
||||||
|
[2, 'https://www.duden.de/suchen/dudenonline/Haus?search_api_fulltext=&page=1']
|
||||||
|
]
|
||||||
|
for page_no, exp_res in data:
|
||||||
|
dic['pageno'] = page_no
|
||||||
params = duden.request(query, dic)
|
params = duden.request(query, dic)
|
||||||
self.assertTrue('url' in params)
|
self.assertTrue('url' in params)
|
||||||
self.assertTrue(query in params['url'])
|
self.assertTrue(query in params['url'])
|
||||||
self.assertTrue('duden.de' in params['url'])
|
self.assertTrue('duden.de' in params['url'])
|
||||||
|
self.assertEqual(params['url'], exp_res)
|
||||||
|
|
||||||
def test_response(self):
|
def test_response(self):
|
||||||
resp = mock.Mock(text='<html></html>')
|
resp = mock.Mock(text='<html></html>')
|
||||||
self.assertEqual(duden.response(resp), [])
|
self.assertEqual(duden.response(resp), [])
|
||||||
|
|
||||||
html = """
|
html = """
|
||||||
<section class="wide">
|
<section class="vignette">
|
||||||
<h2><a href="https://this.is.the.url/" class="hidden-link"><strong>This is the title</strong> also here</a></h2>
|
<h2"> <a href="/rechtschreibung/Haus">
|
||||||
<p>This is the <strong>content</strong></p>
|
<strong>This is the title also here</strong>
|
||||||
<a href="https://this.is.the.url/">Zum vollständigen Artikel</a>
|
</a> </h2>
|
||||||
|
<p>This is the content</p>
|
||||||
</section>
|
</section>
|
||||||
"""
|
"""
|
||||||
|
|
||||||
resp = mock.Mock(text=html)
|
resp = mock.Mock(text=html)
|
||||||
results = duden.response(resp)
|
results = duden.response(resp)
|
||||||
|
|
||||||
|
@ -36,6 +42,6 @@ class TestDudenEngine(SearxTestCase):
|
||||||
|
|
||||||
# testing result (dictionary entry)
|
# testing result (dictionary entry)
|
||||||
r = results[0]
|
r = results[0]
|
||||||
self.assertEqual(r['url'], 'https://this.is.the.url/')
|
self.assertEqual(r['url'], 'https://www.duden.de/rechtschreibung/Haus')
|
||||||
self.assertEqual(r['title'], 'This is the title also here')
|
self.assertEqual(r['title'], 'This is the title also here')
|
||||||
self.assertEqual(r['content'], 'This is the content')
|
self.assertEqual(r['content'], 'This is the content')
|
||||||
|
|
Loading…
Reference in a new issue