3 files changed, 77 insertions, 74 deletions
diff --git a/searx/engines/duden.py b/searx/engines/duden.py
new file mode 100644
index 000000000..881ff9d9c
--- /dev/null
+++ 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
diff --git a/searx/engines/qwant.py b/searx/engines/qwant.py
index 239193b96..4b0f1c87c 100644
--- a/searx/engines/qwant.py
+++ b/searx/engines/qwant.py
@@ -28,7 +28,7 @@ category_to_keyword = {'general': 'web',
                        'social media': 'social'}
 
 # search-url
-url = 'https://api.qwant.com/api/search/{keyword}?count=10&offset={offset}&f=&{query}'
+url = 'https://api.qwant.com/api/search/{keyword}?count=10&offset={offset}&f=&{query}&t={keyword}&uiv=4'
 
 
 # do search-request
diff --git a/searx/engines/www500px.py b/searx/engines/www500px.py
deleted file mode 100644
index 7a2015ae9..000000000
--- a/searx/engines/www500px.py
+++ /dev/null
@@ -1,73 +0,0 @@
-"""
- 500px (Images)
-
- @website     https://500px.com
- @provide-api yes (https://developers.500px.com/)
-
- @using-api   no
- @results     HTML
- @stable      no (HTML can change)
- @parse       url, title, thumbnail, img_src, content
-
- @todo        rewrite to api
-"""
-
-from json import loads
-from searx.url_utils import urlencode, urljoin
-
-# engine dependent config
-categories = ['images']
-paging = True
-
-# search-url
-base_url = 'https://500px.com'
-search_url = 'https://api.500px.com/v1/photos/search?type=photos'\
-    '&{query}'\
-    '&image_size%5B%5D=4'\
-    '&image_size%5B%5D=20'\
-    '&image_size%5B%5D=21'\
-    '&image_size%5B%5D=1080'\
-    '&image_size%5B%5D=1600'\
-    '&image_size%5B%5D=2048'\
-    '&include_states=true'\
-    '&formats=jpeg%2Clytro'\
-    '&include_tags=true'\
-    '&exclude_nude=true'\
-    '&page={pageno}'\
-    '&rpp=50'\
-    '&sdk_key=b68e60cff4c929bedea36ca978830c5caca790c3'
-
-
-# do search-request
-def request(query, params):
-    params['url'] = search_url.format(pageno=params['pageno'],
-                                      query=urlencode({'term': query}))
-
-    return params
-
-
-# get response from search-request
-def response(resp):
-    results = []
-
-    response_json = loads(resp.text)
-
-    # parse results
-    for result in response_json['photos']:
-        url = urljoin(base_url, result['url'])
-        title = result['name']
-        # last index is the biggest resolution
-        img_src = result['image_url'][-1]
-        thumbnail_src = result['image_url'][0]
-        content = result['description'] or ''
-
-        # append result
-        results.append({'url': url,
-                        'title': title,
-                        'img_src': img_src,
-                        'content': content,
-                        'thumbnail_src': thumbnail_src,
-                        'template': 'images.html'})
-
-    # return results
-    return results