From dd0887be186d208846cdc7c3df13dde020dfa957 Mon Sep 17 00:00:00 2001
From: Alexandre FLAMENT <alexandre.flament@hesge.ch>
Date: Fri, 2 Sep 2022 07:33:20 +0000
Subject: [PATCH] xpath engine: change raise_for_httperror to
 no_result_for_http_status

no_result_for_http_status contains a list of HTTP status.
These HTTP status are seen an empty result list.
In other cases an exception is thrown as usual.

Previously raise_for_httperror were ignoring all HTTP error,
which make defective engines invisible in the stats.
---
 searx/engines/xpath.py | 23 ++++++++++++++++++-----
 searx/settings.yml     |  2 +-
 2 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/searx/engines/xpath.py b/searx/engines/xpath.py
index 97656705a..f9528e92d 100644
--- a/searx/engines/xpath.py
+++ b/searx/engines/xpath.py
@@ -22,6 +22,7 @@ from urllib.parse import urlencode
 
 from lxml import html
 from searx.utils import extract_text, extract_url, eval_xpath, eval_xpath_list
+from searx.network import raise_for_httperror
 
 search_url = None
 """
@@ -60,9 +61,14 @@ lang_all = 'en'
 '''Replacement ``{lang}`` in :py:obj:`search_url` if language ``all`` is
 selected.
 '''
-raise_for_httperror = True
-'''True by default: raise an exception if the HTTP code of response is ``>=
-300``'''
+
+no_result_for_http_status = []
+'''Return empty result for these HTTP status codes instead of throwing an error.
+
+.. code:: yaml
+
+    no_result_for_http_status: []
+'''
 
 soft_max_redirects = 0
 '''Maximum redirects, soft limit. Record an error but don't stop the engine'''
@@ -179,12 +185,19 @@ def request(query, params):
 
     params['url'] = search_url.format(**fargs)
     params['soft_max_redirects'] = soft_max_redirects
-    params['raise_for_httperror'] = raise_for_httperror
+
+    params['raise_for_httperror'] = False
+
     return params
 
 
-def response(resp):
+def response(resp):  # pylint: disable=too-many-branches
     '''Scrap *results* from the response (see :ref:`engine results`).'''
+    if no_result_for_http_status and resp.status_code in no_result_for_http_status:
+        return []
+
+    raise_for_httperror(resp)
+
     results = []
     dom = html.fromstring(resp.text)
     is_onion = 'onions' in categories
diff --git a/searx/settings.yml b/searx/settings.yml
index f83c104fd..5587ed51a 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -1808,7 +1808,7 @@ engines:
     url_xpath: //div[@class="upper-synonyms"]/a/@href
     content_xpath: //div[@class="synonyms-list-group"]
     title_xpath: //div[@class="upper-synonyms"]/a
-    raise_for_httperror: false
+    no_result_for_http_status: [404]
     about:
       website: https://www.woxikon.de/
       wikidata_id:  # No Wikidata ID