Merge pull request ComplianceAsCode#12001 from Honny1/show-most-used-…

…rules-of-component

Show most used rules of component

build-scripts/profile_tool.py

@@ -314,6 +314,15 @@ def parse_most_used_components(subparsers):
         choices=get_available_products_with_components_root(),
         default=get_available_products_with_components_root(),
     )
+    parser_most_used_components.add_argument(
+        "--used-rules",
+        default=False,
+        action="store_true",
+        help=(
+            "For every component, show the usage of each component's rule "
+            "in the profiles in given product."
+        ),
+    )
 
 
 def get_available_products_with_components_root():

docs/manual/developer/05_tools_and_utilities.md

@@ -79,6 +79,13 @@ Optionally, you can use this command to limit the statistics for a specific prod
     $ ./build-scripts/profile_tool.py most-used-components --products rhel9
 ```
 
+You can also get a list of the most used components with used rules for the RHEL9 product, you can use the `--used-rules` flag.
+As shown in this command:
+
+```bash
+    $ ./build-scripts/profile_tool.py most-used-components --products rhel9 --used-rules
+```
+
 The result will be a list of rules with the number of uses in the profiles.
 The list can be generated as plain text, JSON or CVS.
 Via the `--format FORMAT` parameter.

utils/profile_tool/common.py

@@ -13,3 +13,19 @@ def generate_output(dict_, format, csv_header):
 
     for rule_id, rule_count in dict_.items():
         print(f_string.format(rule_id, rule_count))
+
+
+def _format_value_b(value_b, delim):
+    str_ = ""
+    if len(value_b) != 0:
+        values = ", ".join([f"{key}: {value}" for key, value in value_b.items()])
+        str_ = f"{delim}{values}"
+    return str_
+
+
+def merge_dicts(dict_a, dict_b, delim):
+    out = {}
+    for key, value in dict_a.items():
+        value_b = dict_b.get(key, {})
+        out[key] = str(value) + _format_value_b(value_b, delim)
+    return out

utils/profile_tool/most_used_components.py

@@ -5,7 +5,7 @@
 import ssg.components
 
 from .most_used_rules import _sorted_dict_by_num_value
-from .common import generate_output
+from .common import generate_output, merge_dicts
 
 PYTHON_2 = sys.version_info[0] < 3
 
@@ -17,10 +17,21 @@
     )
 
 
-def _count_components(components, rules_list, components_out):
+def _count_rules_components(component_name, rules, used_rules_of_components_out):
+    used_rules = defaultdict(int)
+    if component_name in used_rules_of_components_out:
+        used_rules = used_rules_of_components_out[component_name]
+    for rule in rules:
+        used_rules[rule] += 1
+    used_rules_of_components_out[component_name] = used_rules
+
+
+def _count_components(components, rules_list, components_out, used_rules_of_components_out):
     for component_name, component in components.items():
-        if len(set(component.rules).intersection(set(rules_list))) > 0:
+        intersection = set(component.rules).intersection(set(rules_list))
+        if len(intersection) > 0:
             components_out[component_name] += 1
+        _count_rules_components(component_name, intersection, used_rules_of_components_out)
 
 
 def load_components(product):
@@ -33,7 +44,7 @@ def load_components(product):
     return ssg.components.load(components_dir)
 
 
-def _process_all_products_from_controls(components_out, products):
+def _process_all_products_from_controls(components_out, used_rules_of_components_out, products):
     if PYTHON_2:
         raise Exception("This feature is not supported for python2.")
 
@@ -43,14 +54,32 @@ def _process_all_products_from_controls(components_out, products):
             continue
         controls_manager = load_controls_manager("./controls/", product)
         for profile in _get_profiles_for_product(controls_manager, product):
-            _count_components(components, profile.rules, components_out)
+            _count_components(
+                components, profile.rules, components_out, used_rules_of_components_out
+            )
+
+
+def _sort_rules_of_components(used_rules_of_components):
+    out = {}
+    for key, value in used_rules_of_components.items():
+        out[key] = _sorted_dict_by_num_value(value)
+    return out
 
 
 def command_most_used_components(args):
     components = defaultdict(int)
+    used_rules_of_components = {}
 
-    _process_all_products_from_controls(components, args.products)
+    _process_all_products_from_controls(components, used_rules_of_components, args.products)
 
     sorted_components = _sorted_dict_by_num_value(components)
     csv_header = "component_name,count_of_profiles"
+    if args.used_rules:
+        csv_header = "component_name,count_of_profiles,used_rules:count_of_profiles"
+        delim = " "
+        if args.format == "csv":
+            delim = ","
+        sorted_used_rules_of_components = _sort_rules_of_components(used_rules_of_components)
+        sorted_components = merge_dicts(sorted_components, sorted_used_rules_of_components, delim)
+
     generate_output(sorted_components, args.format, csv_header)

utils/profile_tool/most_used_rules.py

@@ -32,7 +32,8 @@ def _get_profiles_for_product(ctrls_mgr, product):
 
     profiles = []
     for file in profiles_files:
-        profiles.append(get_profile(profiles_files, file, ctrls_mgr.policies))
+        if "default.profile" not in file:
+            profiles.append(get_profile(profiles_files, file, ctrls_mgr.policies))
     return profiles
 
 

utils/profile_tool/profile.py

@@ -74,10 +74,14 @@ def _get_sel(selected):
         return policy, control, level
 
     @staticmethod
-    def _get_levels(policy, level):
-        levels = [level]
+    def _get_levels(policy, level, levels=None):
+        if levels is None:
+            levels = set()
+        levels.add(level)
+
         if policy.levels_by_id.get(level).inherits_from is not None:
-            levels.extend(policy.levels_by_id.get(level).inherits_from)
+            for parent_level in policy.levels_by_id.get(level).inherits_from:
+                levels.update(Profile._get_levels(policy, parent_level, levels))
         return levels
 
     def add_from_policy(self, policies, selected):