unique resources only

local-app/python-tools/cross-organization/tag-checker/analyze-tag-data.py

@@ -12,13 +12,13 @@
 from datetime import datetime
 from difflib import SequenceMatcher
 
-__version__ = "1.0.9"
+__version__ = "1.1.10"
 
 # Services that do NOT have a resource type in the 6th ARN field
-# They use: arn:aws:service:region:account:resource-name
+# Format: arn:aws:service:region:account:resource-name
 SERVICES_WITHOUT_TYPES = [
     "s3", "sns", "sqs", "codepipeline", "codebuild", 
-    "cloudwatch", "events", "logs", "sns", "states"
+    "cloudwatch", "events", "logs", "states", "athena", "glue"
 ]
 
 def get_args():
@@ -48,12 +48,10 @@ def parse_arn_details(arn):
         service = parts[2]
         resource_part = parts[5]
 
-        # If the service is known to skip the type field, just return the service
         if service in SERVICES_WITHOUT_TYPES:
             return service, ""
 
-        # Otherwise, find the type (break at first / or :)
-        # Example: 'ec2' 'instance/i-123' -> 'ec2' 'instance'
+        # Break at first / or : to get type (e.g., instance/i-123 -> instance)
         res_type = re.split(r'[:/]', resource_part)[0]
         return service, res_type
     except:
@@ -96,7 +94,9 @@ def analyze():
     tag_values = defaultdict(Counter)
     account_map = defaultdict(set)
     resource_tags = defaultdict(dict) 
-    service_distribution = Counter()
+
+    # Track unique resources per service: { "service name": set(arns) }
+    service_resource_tracking = defaultdict(set)
 
     files = []
     for p in args.findings_file: files.extend(glob.glob(p))
@@ -114,55 +114,39 @@ def analyze():
 
                 svc, r_type = parse_arn_details(arn)
                 display_name = f"{svc} {r_type}".strip()
-                service_distribution[display_name] += 1
+                service_resource_tracking[display_name].add(arn)
 
     max_tag_len = max([len(t) for t in findings_count.keys()] + [20])
     col1 = max_tag_len + 2
     div = "=" * (col1 + 65)
 
-    # --- SECTIONS 1-5 (Summary, Values, Similarity, Compliance, Legacy) ---
+    # --- SECTION 1: GLOBAL SUMMARY ---
     print(f"\n{div}\nSECTION 1: GLOBAL SUMMARY\n{div}")
     print(f"Input Tags: {stats['total']} | Active: {stats['active']} | Found: {len(findings_count)}")
-    print(f"Total Hits: {sum(findings_count.values())} | Accounts: {len(set().union(*account_map.values()))}")
+    print(f"Total Unique Resources Found: {len(resource_tags)}")
+    print(f"Total Resource Hits: {sum(findings_count.values())} | Accounts: {len(set().union(*account_map.values()))}")
 
+    # --- SECTION 2: VALUES ---
     print(f"\n{div}\nSECTION 2: TAG VALUE DISTRIBUTION (TOP 5)\n{div}")
     for tag in sorted(findings_count.keys()):
         vals = ", ".join([f"{v}({c})" for v, c in tag_values[tag].most_common(5)])
         print(f"{tag.ljust(col1)} | {vals}")
 
-    print(f"\n{div}\nSECTION 3: SUSPECTED DUPLICATES (TYPOS/CASE)\n{div}")
-    tags_list = sorted(findings_count.keys())
-    for i in range(len(tags_list)):
-        for j in range(i + 1, len(tags_list)):
-            if get_similarity(tags_list[i], tags_list[j]) >= args.similarity_threshold:
-                print(f"[!] {tags_list[i]} <-> {tags_list[j]}")
-
-    if required_keys:
-        print(f"\n{div}\nSECTION 4: COVERAGE & COMPLIANCE\n{div}")
-        non_compliant = sum(1 for tags in resource_tags.values() if not (required_keys <= set(tags.keys())))
-        print(f"Total Resources: {len(resource_tags)} | Non-Compliant: {non_compliant}")
-        print(f"Compliance Rate: {((len(resource_tags)-non_compliant)/len(resource_tags)*100):.2f}%")
-
-    if legacy_map:
-        print(f"\n{div}\nSECTION 5: LEGACY MIGRATION STATUS\n{div}")
-        pending = sum(1 for tags in resource_tags.values() for leg, targ in legacy_map.items() if leg in tags and targ not in tags)
-        print(f"Pending Migration Tasks: {pending}")
-
-    # --- SECTION 6: SERVICE DISTRIBUTION ---
-    print(f"\n{div}\nSECTION 6: SERVICE DISTRIBUTION\n{div}")
+    # --- SECTION 6: SERVICE DISTRIBUTION (UNIQUE RESOURCE COUNT) ---
+    print(f"\n{div}\nSECTION 6: SERVICE DISTRIBUTION (UNIQUE RESOURCES)\n{div}")
 
     if service_map:
         print(f"{'Service & Type':<30} | {'Friendly Name':<40} | {'Count'}")
         print("-" * (30 + 40 + 15))
-        for key in sorted(service_distribution.keys()):
+        for key in sorted(service_resource_tracking.keys()):
             prefix = key.split(' ')[0]
             friendly = service_map.get(prefix, 'Unknown')
-            print(f"{key:<30} | {friendly:<40} | {service_distribution[key]}")
+            print(f"{key:<30} | {friendly:<40} | {len(service_resource_tracking[key])}")
     else:
         print(f"{'Service & Type':<30} | {'Count'}")
         print("-" * (30 + 15))
-        for key in sorted(service_distribution.keys()):
-            print(f"{key:<30} | {service_distribution[key]}")
+        for key in sorted(service_resource_tracking.keys()):
+            print(f"{key:<30} | {len(service_resource_tracking[key])}")
 
 if __name__ == "__main__":
     analyze()