feat: Implemented mechanism for skipping duplicates in dependency vulnerabilities evaluation

2 files changed, 18 insertions, 1 deletions

diff --git a/sec_certs/dataset/common_criteria.py b/sec_certs/dataset/common_criteria.py
index c5966c97..09dcbe23 100644
--- a/sec_certs/dataset/common_criteria.py
+++ b/sec_certs/dataset/common_criteria.py
@@ -754,6 +754,18 @@ class CCDataset(Dataset[CommonCriteriaCert], ComplexSerializableType):
         keywords.add("1.02.013")
         return {x for x in keywords if len(x) > config.minimal_token_length}
 
+    def get_dataset_cert_ids_occurrences(self) -> Dict[str, int]:
+        all_cert_ids: Dict[str, int] = {}
+
+        for cert_obj in self:
+            if cert_obj.heuristics.cert_id is None:
+                continue
+
+            cert_id = cert_obj.heuristics.cert_id
+            all_cert_ids[cert_id] = all_cert_ids.get(cert_id, 0) + 1
+
+        return all_cert_ids
+
 
 class CCDatasetMaintenanceUpdates(CCDataset, ComplexSerializableType):
     """
diff --git a/sec_certs/sample/common_criteria.py b/sec_certs/sample/common_criteria.py
index 6f797f6d..8e80a34b 100644
--- a/sec_certs/sample/common_criteria.py
+++ b/sec_certs/sample/common_criteria.py
@@ -708,13 +708,18 @@ class CommonCriteriaCert(
             return None
 
         vulnerabilities = set()
+        dataset_cert_id_occurrences = dset.get_dataset_cert_ids_occurrences()
 
         for cert_id in references:
+            cert_id_occurrence = dataset_cert_id_occurrences.get(cert_id, None)
+
+            # Not in dataset or more than 2 occurrences in dataset -> skip
+            if cert_id_occurrence is None or cert_id_occurrence >= 2:
+                continue
 
             for cert_obj in dset:
                 if cert_obj.heuristics.cert_id == cert_id:
                     if cert_obj.heuristics.related_cves:
-
                         vulnerabilities.update(cert_obj.heuristics.related_cves)
 
         if not vulnerabilities: