From 660c1801705e56a81cf02574f48cc194fec95b1f Mon Sep 17 00:00:00 2001
From: Alexandre Flament <alex@al-f.net>
Date: Mon, 6 Sep 2021 08:49:13 +0200
Subject: [mod] plugin: call on_result after each engine from the
 ResultContainer

Currently, searx.search.Search calls on_result once the engine results have been merged
(ResultContainer.order_results).

on_result plugins can rewrite the results: once the URL(s) are modified, even they can be merged,
it won't be the case since ResultContainer.order_results has already be called.

This commit call on_result inside for each result of each engines.
In addition the on_result function can return False to remove the result.

Note: the on_result function now run on the engine thread instead of the Flask thread.
---
 searx/search/__init__.py | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

(limited to 'searx/search')

diff --git a/searx/search/__init__.py b/searx/search/__init__.py
index d8d3e1e1c..6c750a3f9 100644
--- a/searx/search/__init__.py
+++ b/searx/search/__init__.py
@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
 # lint: pylint
-# pylint: disable=missing-module-docstring
+# pylint: disable=missing-module-docstring, too-few-public-methods
 
 import typing
 import threading
@@ -179,7 +179,18 @@ class SearchWithPlugins(Search):
     def __init__(self, search_query, ordered_plugin_list, request):
         super().__init__(search_query)
         self.ordered_plugin_list = ordered_plugin_list
-        self.request = request
+        self.result_container.on_result = self._on_result
+        # pylint: disable=line-too-long
+        # get the "real" request to use it outside the Flask context.
+        # see
+        # * https://github.com/pallets/flask/blob/d01d26e5210e3ee4cbbdef12f05c886e08e92852/src/flask/globals.py#L55
+        # * https://github.com/pallets/werkzeug/blob/3c5d3c9bd0d9ce64590f0af8997a38f3823b368d/src/werkzeug/local.py#L548-L559
+        # * https://werkzeug.palletsprojects.com/en/2.0.x/local/#werkzeug.local.LocalProxy._get_current_object
+        # pylint: enable=line-too-long
+        self.request = request._get_current_object()
+
+    def _on_result(self, result):
+        return plugins.call(self.ordered_plugin_list, 'on_result', self.request, self, result)
 
     def search(self):
         if plugins.call(self.ordered_plugin_list, 'pre_search', self.request, self):
@@ -187,9 +198,6 @@ class SearchWithPlugins(Search):
 
         plugins.call(self.ordered_plugin_list, 'post_search', self.request, self)
 
-        results = self.result_container.get_ordered_results()
-
-        for result in results:
-            plugins.call(self.ordered_plugin_list, 'on_result', self.request, self, result)
+        self.result_container.close()
 
         return self.result_container
-- 
cgit v1.2.3


From 0b27c8698f7b5bbca5083cf234fabbc7d7308349 Mon Sep 17 00:00:00 2001
From: Alexandre Flament <alex@al-f.net>
Date: Thu, 9 Sep 2021 11:23:57 +0200
Subject: [doc] update docs/dev/plugins.rst

---
 searx/search/__init__.py | 10 +++++-----
 searx/search/models.py   |  1 +
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'searx/search')

diff --git a/searx/search/__init__.py b/searx/search/__init__.py
index 6c750a3f9..69d7ffb25 100644
--- a/searx/search/__init__.py
+++ b/searx/search/__init__.py
@@ -39,7 +39,7 @@ class Search:
 
     __slots__ = "search_query", "result_container", "start_time", "actual_timeout"
 
-    def __init__(self, search_query):
+    def __init__(self, search_query: SearchQuery):
         # init vars
         super().__init__()
         self.search_query = search_query
@@ -163,7 +163,7 @@ class Search:
         return True
 
     # do search-request
-    def search(self):
+    def search(self) -> ResultContainer:
         self.start_time = default_timer()
         if not self.search_external_bang():
             if not self.search_answerers():
@@ -172,11 +172,11 @@ class Search:
 
 
 class SearchWithPlugins(Search):
-    """Similar to the Search class but call the plugins."""
+    """Inherit from the Search class, add calls to the plugins."""
 
     __slots__ = 'ordered_plugin_list', 'request'
 
-    def __init__(self, search_query, ordered_plugin_list, request):
+    def __init__(self, search_query: SearchQuery, ordered_plugin_list, request: "flask.Request"):
         super().__init__(search_query)
         self.ordered_plugin_list = ordered_plugin_list
         self.result_container.on_result = self._on_result
@@ -192,7 +192,7 @@ class SearchWithPlugins(Search):
     def _on_result(self, result):
         return plugins.call(self.ordered_plugin_list, 'on_result', self.request, self, result)
 
-    def search(self):
+    def search(self) -> ResultContainer:
         if plugins.call(self.ordered_plugin_list, 'pre_search', self.request, self):
             super().search()
 
diff --git a/searx/search/models.py b/searx/search/models.py
index 7233fac42..e48cb3611 100644
--- a/searx/search/models.py
+++ b/searx/search/models.py
@@ -4,6 +4,7 @@ import typing
 
 
 class EngineRef:
+    """Reference by names to an engine and category"""
 
     __slots__ = 'name', 'category'
 
-- 
cgit v1.2.3