duden.de engine

2018-08-18 19:24:02 +02:00 · 2018-08-18 19:24:02 +02:00 · b575f898c0
parent d377563723
commit b575f898c0
3 changed files with 122 additions and 0 deletions
--- a/searx/engines/duden.py
+++ b/searx/engines/duden.py
@ -0,0 +1,76 @@
+"""
+ 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 searx.engines.xpath import extract_text
+from searx.url_utils import quote
+from searx import logger
+
+categories = ['general']
+paging = True
+language_support = False
+
+# search-url
+base_url = 'https://www.duden.de/'
+search_url = base_url + 'suchen/dudenonline/{query}?page={offset}'
+
+
+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)
+    params['url'] = search_url.format(offset=offset, query=quote(query))
+    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]', '', dom.xpath(
+            '//a[@class="active" and contains(@href,"/suchen/dudenonline")]/span/text()')[0]
+        )
+
+        results.append({'number_of_results': int(number_of_results_string)})
+
+    except:
+        logger.debug("Couldn't read number of results.")
+        pass
+
+    for result in dom.xpath('//section[@class="wide" and not(contains(@style,"overflow:hidden"))]'):
+        try:
+            logger.debug("running for %s" % str(result))
+            link = result.xpath('.//h2/a')[0]
+            url = link.attrib.get('href')
+            title = result.xpath('string(.//h2/a)')
+            content = extract_text(result.xpath('.//p'))
+            # 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
--- a/searx/settings.yml
+++ b/searx/settings.yml
@ -714,6 +714,11 @@ engines:
    shortcut : 1337x
    disabled : True

+  - name : Duden
+    engine : duden
+    shortcut : du
+    disabled : True
+
 #  - name : yacy
 #    engine : yacy
 #    shortcut : ya
--- a/tests/unit/engines/test_duden.py
+++ b/tests/unit/engines/test_duden.py
@ -0,0 +1,41 @@
+from collections import defaultdict
+import mock
+from searx.engines import duden
+from searx.testing import SearxTestCase
+from datetime import datetime
+
+
+class TestDudenEngine(SearxTestCase):
+
+    def test_request(self):
+        query = 'Haus'
+        dic = defaultdict(dict)
+        dic['pageno'] = 1
+        params = duden.request(query, dic)
+        self.assertTrue('url' in params)
+        self.assertTrue(query in params['url'])
+        self.assertTrue('duden.de' in params['url'])
+
+    def test_response(self):
+        resp = mock.Mock(text='<html></html>')
+        self.assertEqual(duden.response(resp), [])
+
+        html = """
+        <section class="wide">
+        <h2><a href="https://this.is.the.url/" class="hidden-link"><strong>This is the title</strong> also here</a></h2>
+        <p>This is the <strong>content</strong></p>
+        <a href="https://this.is.the.url/">Zum vollst&auml;ndigen Artikel</a>
+        </section>
+        """
+
+        resp = mock.Mock(text=html)
+        results = duden.response(resp)
+
+        self.assertEqual(len(results), 1)
+        self.assertEqual(type(results), list)
+
+        # testing result (dictionary entry)
+        r = results[0]
+        self.assertEqual(r['url'], 'https://this.is.the.url/')
+        self.assertEqual(r['title'], 'This is the title also here')
+        self.assertEqual(r['content'], 'This is the content')