Merge pull request #2564 from dalf/fix-seznam

[fix] fix seznam engine

2 files changed, 65 insertions, 17 deletions

diff --git a/searx/engines/seznam.py b/searx/engines/seznam.py
new file mode 100644
index 000000000..1df92a845
--- /dev/null
+++ b/searx/engines/seznam.py
@@ -0,0 +1,64 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""
+ Seznam
+"""
+
+from urllib.parse import urlencode, urlparse
+from lxml import html
+from searx.poolrequests import get
+from searx.exceptions import SearxEngineAccessDeniedException
+from searx.utils import extract_text, eval_xpath_list, eval_xpath_getindex
+
+# about
+about = {
+    "website": "https://www.seznam.cz/",
+    "wikidata_id": "Q3490485",
+    "official_api_documentation": "https://api.sklik.cz/",
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": "HTML",
+}
+
+base_url = 'https://search.seznam.cz/'
+
+
+def request(query, params):
+    response_index = get(base_url, headers=params['headers'], raise_for_httperror=True)
+    dom = html.fromstring(response_index.text)
+
+    url_params = {'q': query}
+    for e in eval_xpath_list(dom, '//input[@type="hidden"]'):
+        name = e.get('name')
+        value = e.get('value')
+        url_params[name] = value
+
+    params['url'] = base_url + '?' + urlencode(url_params)
+    params['cookies'] = response_index.cookies
+    return params
+
+
+def response(resp):
+    resp_url = urlparse(resp.url)
+    if resp_url.path.startswith('/verify'):
+        raise SearxEngineAccessDeniedException()
+
+    results = []
+
+    dom = html.fromstring(resp.content.decode())
+    for result_element in eval_xpath_list(dom, '//div[@id="searchpage-root"]//div[@data-dot="results"]/div'):
+        dot_data = eval_xpath_getindex(result_element, './div/div[@data-dot-data]/@data-dot-data', 0, default=None)
+        if dot_data is None:
+            title_element = eval_xpath_getindex(result_element, './/h3/a', 0)
+            results.append({
+                'url': title_element.get('href'),
+                'title': extract_text(title_element),
+                'content': extract_text(eval_xpath_getindex(title_element, '../../div[2]', 0)),
+            })
+        elif dot_data == '{"reporter_name":"hint/related/relates"}':
+            suggestions_element = eval_xpath_getindex(result_element,
+                                                      './div/div[@data-dot="main-box"]', 0, default=None)
+            if suggestions_element is not None:
+                for suggestion in eval_xpath_list(suggestions_element, './/ul/li'):
+                    results.append({'suggestion': extract_text(suggestion)})
+
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
index 87008eb20..4e926d73c 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -1170,24 +1170,8 @@ engines:
 
   - name : seznam
     shortcut: szn
-    engine: xpath
-    paging : True
-    search_url : https://search.seznam.cz/?q={query}&count=10&from={pageno}
-    results_xpath: //div[@class="Page-content"]//div[contains(@class, "Result ")]
-    url_xpath : ./h3/a/@href
-    title_xpath : ./h3
-    content_xpath : .//p[@class="Result-description"]
-    suggestion_xpath: //div[@class="Related-container"]//div[@class="RelatedItem"]/div/span/a
-    first_page_num : 0
-    page_size : 10
+    engine: seznam
     disabled : True
-    about:
-      website: https://www.seznam.cz/
-      wikidata_id: Q3490485
-      official_api_documentation: https://api.sklik.cz/
-      use_official_api: false
-      require_api_key: false
-      results: HTML
 
   - name : mojeek
     shortcut: mjk