Merge branch 'master' into bug_in_HC_set

README.md

@@ -101,6 +101,11 @@ The arguments to `--resource_list` and to `--base_resource_list` should be one o
 - `--output_endpoints`\
   Choose endpoints type in output (pods/deployments).\
   *default:* deployments
+  - `--explain`\
+  A pair of node names (comma separated) to explain the policies affecting their connection or lack of it. Relevant only for connectivity query.\
+  Connections including IP-Blocks will show only the configurations of the node in that connection (since, IP-Blocks does
+  not have configurations). IP-Blocks should be places in CIDR format as seen in the query results (run the connectivity query first, to see the nodes there).\
+  e.g. default/pod-A1,default/deployment-B1.
   - `--print_ipv6`\
   include IPv6 range in the query results even when the policies of the config do not contain any IPv6 addresses.
 

docs/SchemeFileFormat.md

@@ -78,15 +78,15 @@ For example: `my_set/prod_ns/deny_all_policy`. If there are multiple policies na
 #### <a name="outputconfig"></a>Output Configuration object
 The supported entries in the outputConfiguration object are as follows:
 
-| Field            | Description                                                                                                     | Value                                                        |
-|------------------|-----------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------|
-| outputFormat     | Output format specification.                                                                                    | string [ txt / yaml / csv / md / dot / jpg/ txt_no_fw_rules] |
-| outputPath       | A file path to redirect output into.                                                                            | string                                                       |
-| outputEndpoints  | Choose endpoints type in output.                                                                                | string [ pods / deployments ]                                |
-| subset           | A dict object with the defined subset elements to display in the output                                         | [subset](#subset) object                                     |
-| fullExplanation  | Choose if to print all counterexamples causing the query result in the output                                   | bool                                                         |
-| excludeIPv6Range | If the policies of the config do not contain any IPv6 addresses, do not include IPv6 range in the query results | bool  [default: True]                                        |
-
+| Field            | Description                                                                                                                                                                                                                                                                                                                                                                                                                            | Value                                                        |
+|------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------|
+| outputFormat     | Output format specification.                                                                                                                                                                                                                                                                                                                                                                                                           | string [ txt / yaml / csv / md / dot / jpg/ txt_no_fw_rules] |
+| outputPath       | A file path to redirect output into.                                                                                                                                                                                                                                                                                                                                                                                                   | string                                                       |
+| outputEndpoints  | Choose endpoints type in output.                                                                                                                                                                                                                                                                                                                                                                                                       | string [ pods / deployments ]                                |
+| subset           | A dict object with the defined subset elements to display in the output                                                                                                                                                                                                                                                                                                                                                                | [subset](#subset) object                                     |
+| fullExplanation  | Choose if to print all counterexamples causing the query result in the output                                                                                                                                                                                                                                                                                                                                                          | bool                                                         |
+| excludeIPv6Range | If the policies of the config do not contain any IPv6 addresses, do not include IPv6 range in the query results                                                                                                                                                                                                                                                                                                                        | bool  [default: True]                                        |
+| explain          | A pair of node names (comma separated) to explain the policies affecting their connection or lack of it. Relevant only for connectivityMap query. Connections including IP-Blocks will show only the configurations of the node in that connection (since, IP-Blocks does not have configurations). IP-Blocks should be places in CIDR format as seen in the query results (run the connectivity query first, to see the nodes there). | string [ ns/node1,ns/node2 ]                           |
 
 #### <a name="subset"></a>Subset object
 The supported entries in the subset object are as follows:

nca/FWRules/ConnectivityGraph.py

@@ -343,7 +343,7 @@ def get_connectivity_dot_format_str(self, connectivity_restriction=None):
 
         dot_graph = DotGraph(name)
         peers_groups = self._get_equals_groups()
-        # we are going to treat a a peers_group as one peer.
+        # we are going to treat a peers_group as one peer.
         # the first peer in the peers_group is representing the group
         # we will add the text of all the peers in the group to this peer
         for peers_group, group_connection in peers_groups:

nca/FWRules/FWRule.py

@@ -383,7 +383,8 @@ def get_pod_str(self):
         """
         :return: string for the field src_pods or dst_pods in representation for txt rule format
         """
-        return f'[{self._get_pods_names()}]'
+        sorted_pods_names = ', '.join(sorted(self._get_pods_names().split(', ')))
+        return f'[{sorted_pods_names}]'
 
     def _get_pods_names(self):
         res = ''

nca/FWRules/InteractiveConnectivityGraph.py

@@ -127,7 +127,7 @@ def read_input_file(self):
             """
             try:
                 with open(self.input_svg_file) as svg_file:
-                    self.soup = BeautifulSoup(svg_file.read(), 'xml')
+                    self.soup = BeautifulSoup(svg_file.read(), 'html')
             except Exception as e:
                 print(f'Failed to open file: {self.input_svg_file}\n{e} for reading', file=sys.stderr)
 

nca/FileScanners/GenericTreeScanner.py

@@ -21,6 +21,7 @@ class YamlFile:
 
 class ObjectWithLocation:
     line_number = 0
+    path = ''
     column_number = 0
 
 
@@ -36,13 +37,15 @@ def to_yaml_objects(yaml_node):
     if isinstance(yaml_node, yaml.SequenceNode):
         res = YamlList()
         res.line_number = yaml_node.start_mark.line
+        res.path = yaml_node.start_mark.name
         res.column_number = yaml_node.start_mark.column
         for obj in yaml_node.value:
             res.append(to_yaml_objects(obj))
         return res
     if isinstance(yaml_node, yaml.MappingNode):
         res = YamlDict()
         res.line_number = yaml_node.start_mark.line + 1
+        res.path = yaml_node.start_mark.name
         res.column_number = yaml_node.start_mark.column + 1
         for obj in yaml_node.value:
             res[obj[0].value] = to_yaml_objects(obj[1])

nca/NetworkConfig/NetworkConfig.py

@@ -9,6 +9,7 @@
 from nca.CoreDS.ConnectivityProperties import ConnectivityProperties
 from nca.Resources.NetworkPolicy import NetworkPolicy, OptimizedPolicyConnections
 from .NetworkLayer import NetworkLayersContainer, NetworkLayerName
+from nca.Utils.ExplTracker import ExplTracker
 
 
 @dataclass
@@ -275,6 +276,8 @@ def allowed_connections_optimized(self, layer_name=None):
         :return: allowed_conns: all allowed connections for relevant peers.
         :rtype: OptimizedPolicyConnections
         """
+        if ExplTracker().is_active():
+            ExplTracker().set_peers(self.peer_container.peer_set)
         if layer_name is not None:
             if layer_name not in self.policies_container.layers:
                 return self.policies_container.layers.empty_layer_allowed_connections_optimized(self.peer_container,

nca/NetworkConfig/NetworkConfigQuery.py

@@ -24,6 +24,7 @@
     PoliciesAndRulesExplanations, PodsListsExplanations, ConnectionsDiffExplanation, IntersectPodsExplanation, \
     PoliciesWithCommonPods, PeersAndConnections, ComputedExplanation
 from .NetworkLayer import NetworkLayerName
+from nca.Utils.ExplTracker import ExplTracker
 
 
 class QueryType(Enum):
@@ -805,17 +806,24 @@ def compute_connectivity_output_optimized(self):
                                                                                "dst_peers": opt_peers_to_compare})
         base_peers_num = len(opt_peers_to_compare)
         subset_peers = self.compute_subset(opt_peers_to_compare)
+        all_peers = subset_peers
         if len(subset_peers) != base_peers_num:
             # remove connections where both of src_peers and dst_peers are out of the subset
             subset_conns = ConnectivityProperties.make_conn_props_from_dict({"src_peers": subset_peers}) | \
                            ConnectivityProperties.make_conn_props_from_dict({"dst_peers": subset_peers})
             all_conns_opt &= subset_conns
+            src_peers, dst_peers = ExplTracker().extract_peers(all_conns_opt)
+            all_peers = src_peers | dst_peers
         all_conns_opt = self.config.filter_conns_by_peer_types(all_conns_opt, opt_peers_to_compare)
+        expl_conns = all_conns_opt
         if self.config.policies_container.layers.does_contain_layer(NetworkLayerName.Istio):
             output_res, opt_fw_rules_tcp, opt_fw_rules_non_tcp = \
                 self.get_props_output_split_by_tcp(all_conns_opt, opt_peers_to_compare)
+            expl_conns, _ = self.convert_props_to_split_by_tcp(all_conns_opt)
         else:
             output_res, opt_fw_rules = self.get_props_output_full(all_conns_opt, opt_peers_to_compare)
+        if ExplTracker().is_active():
+            ExplTracker().set_connections_and_peers(expl_conns, all_peers)
         return output_res, opt_fw_rules, opt_fw_rules_tcp, opt_fw_rules_non_tcp
 
     def exec(self):
@@ -900,7 +908,9 @@ def get_props_output_full(self, props, peers_to_compare):
         if self.output_config.outputFormat in ['dot', 'jpg']:
             dot_full = self.dot_format_from_props(props, peers_to_compare)
             return dot_full, None
-        # TODO - handle 'txt_no_fw_rules' output format
+        if self.output_config.outputFormat == 'txt_no_fw_rules':
+            conns_wo_fw_rules = self.txt_no_fw_rules_format_from_props(props, peers_to_compare)
+            return conns_wo_fw_rules, None
         # handle other formats
         formatted_rules, fw_rules = self.fw_rules_from_props(props, peers_to_compare)
         return formatted_rules, fw_rules
@@ -966,7 +976,12 @@ def get_props_output_split_by_tcp(self, props, peers_to_compare):
             # concatenate the two graphs into one dot file
             res_str = dot_tcp + dot_non_tcp
             return res_str, None, None
-        # TODO - handle 'txt_no_fw_rules' output format
+        if self.output_config.outputFormat in ['txt_no_fw_rules']:
+            txt_no_fw_rules_tcp = self.txt_no_fw_rules_format_from_props(props_tcp, peers_to_compare, connectivity_tcp_str)
+            txt_no_fw_rules_non_tcp = self.txt_no_fw_rules_format_from_props(props_non_tcp, peers_to_compare,
+                                                                             connectivity_non_tcp_str)
+            res_str = txt_no_fw_rules_tcp + txt_no_fw_rules_non_tcp
+            return res_str, None, None
         # handle formats other than dot and txt_no_fw_rules
         formatted_rules_tcp, fw_rules_tcp = self.fw_rules_from_props(props_tcp, peers_to_compare, connectivity_tcp_str)
         formatted_rules_non_tcp, fw_rules_non_tcp = self.fw_rules_from_props(props_non_tcp, peers_to_compare,
@@ -1033,6 +1048,20 @@ def dot_format_from_props(self, props, peers, connectivity_restriction=None):
             conn_graph.add_edges_from_cube_dict(props.get_connectivity_cube(cube), self.config.peer_container)
         return conn_graph.get_connectivity_dot_format_str(connectivity_restriction)
 
+    def txt_no_fw_rules_format_from_props(self, props, peers, connectivity_restriction=None):
+        """
+        :param ConnectivityProperties props: properties describing allowed connections
+        :param PeerSet peers: the peers to consider for dot output
+        :param Union[str,None] connectivity_restriction: specify if connectivity is restricted to
+               TCP / non-TCP , or not
+        :rtype str
+        :return the connectivity map in txt_no_fw_rules format, considering connectivity_restriction if required
+        """
+        conn_graph = ConnectivityGraph(peers, self.config.get_allowed_labels(), self.output_config)
+        for cube in props:
+            conn_graph.add_edges_from_cube_dict(props.get_connectivity_cube(cube), self.config.peer_container)
+        return conn_graph.get_connections_without_fw_rules_txt_format(connectivity_restriction)
+
     def fw_rules_from_connections_dict(self, connections, peers_to_compare, connectivity_restriction=None):
         """
         :param dict connections: the connections' dict (map from connection-set to peer pairs)

nca/NetworkConfig/NetworkConfigQueryRunner.py

@@ -11,6 +11,7 @@
 from nca.Resources.NetworkPolicy import NetworkPolicy
 from .NetworkConfig import NetworkConfig
 from . import NetworkConfigQuery
+from nca.Utils.ExplTracker import ExplTracker
 
 
 @dataclass
@@ -40,9 +41,9 @@ def compute_final_results(self, output_format):
         extracts the final query results from self variables
         from self.query_iterations_output computes the final str output of the query,
         other results returned as is from query_result.
-        :param str output_format: the output format to form the final output
-        if output format is json, dumps the output list into one-top-leveled string
-        if output format is yaml, dumps the output list into str of a list of yaml objects
+        :param str output_format: the output format to form the final output.
+        if output format is json, dumps the output list into one-top-leveled string.
+        if output format is yaml, dumps the output list into str of a list of yaml objects.
         otherwise, writes the output list items split by \n
         :return the results: numerical result, output - str , num of not executed
         :rtype: int, str, int
@@ -169,7 +170,12 @@ def _run_query_for_each_config(self):
         query_result = QueryResult()
         for config in self.configs_array:
             query_result.update(self._execute_one_config_query(self.query_name, self._get_config(config)))
-        return query_result.compute_final_results(self.output_configuration.outputFormat)
+        expl_out = ''
+        if ExplTracker().is_active() and self.output_configuration.explain and \
+                ExplTracker().is_output_format_supported(self.output_configuration.outputFormat):
+            expl_out = '\n\nExplainability results:\n'+ExplTracker().explain(self.output_configuration.explain.split(','))
+        numerical_result, output, num_not_executed = query_result.compute_final_results(self.output_configuration.outputFormat)
+        return numerical_result, output + expl_out, num_not_executed
 
     def _run_query_on_configs_vs_base_config(self, cmd_line_flag):
         query_result = QueryResult()