Move Filter Users plugin into pre_analysis (and keep display part)

plugins/post_analysis/filter_users.py

@@ -0,0 +1,145 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright Grégory Soutadé 2020
+
+# This file is part of iwla
+
+# iwla is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# iwla is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with iwla.  If not, see <http://www.gnu.org/licenses/>.
+#
+
+import re
+
+from iwla import IWLA
+from iplugin import IPlugin
+from display import *
+
+"""
+Post analysis hook
+
+Filter users with given user conditions
+
+Plugin requirements :
+    None
+
+Conf values needed :
+    filtered_users : list of filters
+    filtered_ip : list of ip (string)
+    create_filtered_page*
+
+Filter is a list of filter description combined by AND operator
+Filter description is a list of 3 elements :
+
+  * Field to match in visits
+  * Operator '=', '==', '!=', '>', '>=', '<', '<=' for int value
+  * Operator '=', '==', '!=', 'in', 'match' for str value
+  * Target value
+
+For easiest config, you can indicate both 'remote_addr' or 'ip' in field element
+
+Output files :
+    None
+
+Statistics creation :
+visits :
+   remote_addr =>
+       filtered
+
+Statistics update :
+visits :
+   remote_addr =>
+       keep_requests
+
+Statistics deletion :
+    None
+"""
+
+class IWLAPostAnalysisFilterUsers(IPlugin):
+    def __init__(self, iwla):
+        super(IWLAPostAnalysisFilterUsers, self).__init__(iwla)
+        self.API_VERSION = 1
+
+    def _check_filter(self, _filter):
+        if len(_filter) != 3:
+            raise Exception('Bad filter ' + ' '.join(_filter))
+        (field, operator, value) = _filter
+        try:
+            if type(value) == str:
+                value = int(value, 10)
+                _filter[2] = value
+            if operator not in ('=', '==', '!=', '>', '>=', '<', '<='):
+                raise Exception('Bad filter ' + ' '.join(_filter))
+        except Exception as e:
+            if field == 'ip':
+                _filter[0] = 'remote_addr'
+            if operator not in ('=', '==', '!=', 'in', 'match'):
+                raise Exception('Bad filter ' + ' '.join(_filter))
+            if operator == 'match':
+                _filter[2] = re.compile(value)
+    
+    def load(self):
+        self.filters = self.iwla.getConfValue('filtered_users', [])
+        self.ip_filters = self.iwla.getConfValue('filtered_ip', [])
+        for _filter in self.filters:
+            for sub_filter in _filter:
+                self._check_filter(sub_filter)
+        return True
+    
+    def __do_filter(self, hit, _filter):
+        (field, operator, value) = _filter
+        if not field in hit.keys(): return False
+        hit_value = hit[field]
+        # In dict, we have something like : {dayX : value} were day0 is total
+        if type(value) == int and type(hit_value) == dict:
+            hit_value = list(hit_value.values())[0]
+        if operator == '=' or operator == '==':
+            return hit_value == value
+        elif operator == 'in':
+            return hit_value in value
+        elif operator == 'match':
+            return re.match(value, hit_value)
+        elif operator == '!=':
+            return hit_value != value
+        elif operator == '<':
+            return hit_value < value
+        elif operator == '<=':
+            return hit_value <= value
+        elif operator == '>':
+            return hit_value > value
+        elif operator == '>=':
+            return hit_value >= value
+    
+    def _do_filter(self, ip, hits):
+        if ip in self.ip_filters:
+            return True
+        
+        for _filter in self.filters:
+            filtered = True
+            for sub_filter in _filter:
+                if not self.__do_filter(hits[ip], sub_filter):
+                    filtered = False
+                    break
+            if filtered:
+                return True
+        return False
+    
+    def hook(self):
+        hits = self.iwla.getValidVisitors()
+
+        if len(self.filters) or len(self.ip_filters):
+            for (key,value) in hits.items():
+                if self._do_filter(key, hits):
+                    print('Keep for {}'.format(key))
+                    value['filtered'] = True
+                    value['keep_requests'] = True
+