Merge pull request #59 from lyle-morrisBV/main

Added dictionary and binary search lookups

src/epss_api/epss.py

@@ -4,6 +4,7 @@
 from functools import cached_property
 from gzip import GzipFile
 from urllib.request import urlopen
+from bisect import bisect_left
 
 
 class Score(object):
@@ -17,7 +18,21 @@ def __init__(self, cve: str, epss: str, percentile: str):
 
 class EPSS(object):
     def __init__(self) -> None:
-        pass
+
+        url = 'https://epss.cyentia.com/epss_scores-current.csv.gz'
+
+        with urlopen(url) as res:
+            dec = GzipFile(fileobj=res)
+            epss_scores_str: str = dec.read().decode("utf-8")
+            epss_scores_list = epss_scores_str.split('\n')
+
+        self._download = epss_scores_list
+
+        scores = [row for row in csv.DictReader(self._download[1:])]
+
+        self._byCVE = {row['cve'] : Score(row['cve'], row['epss'], row['percentile']) for row in scores}
+
+        self._sortedScores = sorted(self._byCVE.values(),key=lambda x:x.percentile) 
 
     def scores(self) -> list[Score]:
         """Get all CVE's EPSS scores (downloaded data is cached in memory)
@@ -33,9 +48,7 @@ def scores(self) -> list[Score]:
         Returns:
             list[Score]: EPSS score's csv list
         """
-        scores = [row for row in csv.DictReader(self._download[1:])]
-        return [Score(row['cve'], row['epss'], row['percentile'])
-                for row in scores]
+        return list(self._sortedScores)
 
     def score(self, cve_id: str) -> Score:
         """Get EPSS score and percentile
@@ -49,11 +62,8 @@ def score(self, cve_id: str) -> Score:
         Returns:
             Score | None: EPSS score percentile
         """
-        rows = self._filter_by_cve_id(cve_id)
-        if len(rows) == 1:
-            return rows[0]
-        else:
-            return None
+
+        return self._byCVE.get(cve_id,None)
 
     def epss(self, cve_id: str) -> float:
         """Get EPSS score
@@ -64,11 +74,12 @@ def epss(self, cve_id: str) -> float:
         Returns:
             float | None: EPSS score (0.0-1.0)
         """
-        rows = self._filter_by_cve_id(cve_id)
-        if len(rows) == 1:
-            return rows[0].epss
-        else:
+
+        score = self._byCVE.get(cve_id,None)
+        if score is None:
             return None
+        else:
+            return score.epss
 
     def percentile(self, cve_id: str) -> float:
         """Get EPSS percentile
@@ -79,11 +90,11 @@ def percentile(self, cve_id: str) -> float:
         Returns:
             float | None: EPSS percentile (0.0-1.0)
         """
-        rows = self._filter_by_cve_id(cve_id)
-        if len(rows) == 1:
-            return rows[0].percentile
-        else:
+        score = self._byCVE.get(cve_id,None)
+        if score is None:
             return None
+        else:
+            return score.percentile
 
     def epss_gt(self, max: float) -> list[Score]:
         """Get CVEs with EPSS score greater or equal than the parameter
@@ -94,8 +105,9 @@ def epss_gt(self, max: float) -> list[Score]:
         Returns:
             list[Score] | None: EPSS score object list
         """
-        rows = [r for r in filter(lambda x: x.epss >= max, self.scores())]
-        return rows
+        i = bisect_left(self._sortedScores,min,key=lambda x:x.epss)
+
+        return list(self._sortedScores[i:])
 
     def percentile_gt(self, max: float) -> list[Score]:
         """Get CVEs with percentile greater or equal than the parameter
@@ -106,9 +118,9 @@ def percentile_gt(self, max: float) -> list[Score]:
         Returns:
             list[Score] | None: EPSS score object list
         """
-        rows = [r for r in
-                filter(lambda x: x.percentile >= max, self.scores())]
-        return rows
+        i = bisect_left(self._sortedScores,min,key=lambda x:x.percentile)
+
+        return list(self._sortedScores[i:])
 
     def epss_lt(self, min: float) -> list[Score]:
         """Get CVEs with EPSS score lower or equal than the parameter
@@ -119,8 +131,9 @@ def epss_lt(self, min: float) -> list[Score]:
         Returns:
             list[Score] | None: EPSS score object list
         """
-        rows = [r for r in filter(lambda x: x.epss <= min, self.scores())]
-        return rows
+        i = bisect_left(self._sortedScores[::-1],min,key=lambda x:1-x.epss)
+
+        return list(self._sortedScores[:len(self.sortedScores)-i])
 
     def percentile_lt(self, min: float) -> list[Score]:
         """Get CVEs with percentile lower or equal than the parameter
@@ -131,9 +144,9 @@ def percentile_lt(self, min: float) -> list[Score]:
         Returns:
             list[Score] | None: EPSS score object list
         """
-        rows = [r for r in
-                filter(lambda x: x.percentile <= min, self.scores())]
-        return rows
+        i = bisect_left(self._sortedScores[::-1],min,key=lambda x:1-x.percentile)
+
+        return list(self._sortedScores[:len(self.sortedScores)-i])
 
     def csv(self) -> list[str]:
         """Get csv data containing all epss scores.
@@ -150,18 +163,3 @@ def csv(self) -> list[str]:
         """
         return self._download
 
-    def _filter_by_cve_id(self, cve_id: str):
-        cve_filter = filter(lambda x: x.cve == cve_id, self.scores())
-        rows = [row for row in cve_filter]
-        return rows
-
-    @cached_property
-    def _download(self):
-        url = 'https://epss.cyentia.com/epss_scores-current.csv.gz'
-
-        with urlopen(url) as res:
-            dec = GzipFile(fileobj=res)
-            epss_scores_str: str = dec.read().decode("utf-8")
-            epss_scores_list = epss_scores_str.split('\n')
-
-        return epss_scores_list