From d1911fce8b9aa4006659d55740c22d3e8a40dbb5 Mon Sep 17 00:00:00 2001
From: J08nY
Date: Wed, 7 Feb 2024 10:09:54 +0100
Subject: Fix cert_id canonicalization.

---
 src/sec_certs/rules.yaml   |  4 ++--
 src/sec_certs/sample/cc.py | 22 ++++++++++++++++++----
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/src/sec_certs/rules.yaml b/src/sec_certs/rules.yaml
index 4305b8cd..250c45b9 100644
--- a/src/sec_certs/rules.yaml
+++ b/src/sec_certs/rules.yaml
@@ -70,7 +70,7 @@ cc_cert_id:
     # KECS-CISS-1210-2023
   JP:
     - "(?:CRP|ACR)-C(?P<counter>[0-9]+)-(?P<digit>[0-9]+)"
-    - "JISEC-CC-CRP-C(?P<counter>[0-9]+)-(?P<digit>[0-9]+)-(?P<year>[0-9]{4})"
+    - "JISEC-CC-CRP-C(?P<counter>[0-9]+)(?:-(?P<digit>[0-9]{2}))?(?:-(?P<year>[0-9]{4}))?"
     - "Certification No. [cC](?P<counter>[0-9]+)"
     # Examples:
     # CRP-C0595-01
@@ -144,7 +144,7 @@ cc_filename_cert_id:
   ES:
     - "(?P<year>[0-9]{4})[-‐](?P<project>[0-9]+)[-‐]INF[-‐](?P<counter>[0-9]+)[ -‐_]{1,2}[vV](?P<version>[0-9])"
   KR:
-    - "(?P<word>ISIS|NISS|CISS)[-‐](?P<counter>[0-9]{2,4})(?:[-‐](?P<year>[0-9]{4}))?"
+    - "(?P<word>ISIS|NISS|CISS)[-‐](?P<counter>[0-9]{2,4})[-‐](?P<year>[0-9]{4})"
   JP:
     - "[cC](?P<counter>[0-9]+)"
   MY:
diff --git a/src/sec_certs/sample/cc.py b/src/sec_certs/sample/cc.py
index ef1ff984..904c794d 100644
--- a/src/sec_certs/sample/cc.py
+++ b/src/sec_certs/sample/cc.py
@@ -417,14 +417,28 @@ class CCCertificate(
             candidates: dict[str, float] = defaultdict(lambda: 0.0)
             # TODO: Add heuristic based on ordering of ids (and extracted year + increment)
             # TODO: Add heuristic based on length
+            # TODO: Add heuristic based on id "richness", we want to prefer IDs that have more components.
+            # If we cannot canonicalize, just skip that ID.
             for candidate, count in frontpage_id.items():
-                candidates[canonicalize(candidate, scheme)] += count * 1.5
+                try:
+                    candidates[canonicalize(candidate, scheme)] += count * 1.5
+                except Exception:
+                    continue
             for candidate, count in metadata_id.items():
-                candidates[canonicalize(candidate, scheme)] += count * 1.2
+                try:
+                    candidates[canonicalize(candidate, scheme)] += count * 1.2
+                except Exception:
+                    continue
             for candidate, count in keywords_id.items():
-                candidates[canonicalize(candidate, scheme)] += count * 1.0
+                try:
+                    candidates[canonicalize(candidate, scheme)] += count * 1.0
+                except Exception:
+                    continue
             for candidate, count in filename_id.items():
-                candidates[canonicalize(candidate, scheme)] += count * 1.0
+                try:
+                    candidates[canonicalize(candidate, scheme)] += count * 1.0
+                except Exception:
+                    continue
             return candidates
 
     @dataclass
-- 
cgit v1.3.1