From 520ce49b95c7f79481a68f23ec74bffbc93a80c1 Mon Sep 17 00:00:00 2001
From: J08nY <johny@neuromancer.sk>
Date: Mon, 5 Feb 2024 14:05:28 +0100
Subject: [PATCH] Improve Indian rules.

---
 src/sec_certs/rules.yaml                  | 10 +++++++---
 src/sec_certs/sample/cc_certificate_id.py | 14 ++++++++++++--
 tests/cc/test_cc_misc.py                  |  2 +-
 3 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/src/sec_certs/rules.yaml b/src/sec_certs/rules.yaml
index 5e000144..41357015 100644
--- a/src/sec_certs/rules.yaml
+++ b/src/sec_certs/rules.yaml
@@ -94,9 +94,13 @@ cc_cert_id:
     # CSEC2019015
     # CSEC 2019012
   IN:
-    # India (IC3S/DEL01/VALIANT/EAL1/0317/0007/CR  STQC/CC/14-15/12/ETR/0017 IC3S/MUM01/CISCO/cPP/0119/0016/CR)
-    # will miss STQC/CC/14-15/12/ETR/0017
-    - "(?:IC3S|STQC/CC)/[^ ]+? ?/CR"
+    - "IC3S/(?P<lab>[A-Z]+[0-9]+)/(?P<vendor>[a-zA-Z_]+)/(?P<level>[a-zA-Z0-9]+)/(?P<number1>[0-9]+)/(?P<number2>[0-9]+) ?(?:/CR)?"
+    # XXX: The cert IDs are often present only in the certificate and not in the report.
+    #      The report often only has the report id, of the format "STQC/CC/1617/18/CR"
+    # Examples:
+    # IC3S/BG01/HALTDOS/EAL2/0317/0008/CR
+    # IC3S/KOL01/ADVA/EAL2/0520/0021/CR
+    # IC3S/MUM01/Symantec/NDcPP/0722/0032/CR
   SG:
     - "CSA_CC_(?P<year>[0-9]{2})(?P<counter>[0-9]{3})"
     # Examples:
diff --git a/src/sec_certs/sample/cc_certificate_id.py b/src/sec_certs/sample/cc_certificate_id.py
index 5facf1a5..7997b846 100644
--- a/src/sec_certs/sample/cc_certificate_id.py
+++ b/src/sec_certs/sample/cc_certificate_id.py
@@ -117,7 +117,17 @@ def _canonical_es(self) -> str:
         return new_cert_id
 
     def _canonical_in(self):
-        return self.clean.replace(" ", "")
+        new_cert_id = self.clean
+        for rule in rules["cc_cert_id"]["IN"]:
+            if match := re.match(rule, new_cert_id):
+                groups = match.groupdict()
+                lab = groups["lab"]
+                vendor = groups["vendor"]
+                level = groups["level"]
+                number1, number2 = groups["number1"], groups["number2"]
+                new_cert_id = f"IC3S/{lab}/{vendor}/{level}/{number1}/{number2}"
+                break
+        return new_cert_id
 
     def _canonical_se(self):
         new_cert_id = self.clean
@@ -232,7 +242,7 @@ def canonical(self) -> str:
             "AU": self._canonical_au,
             "KR": self._canonical_kr,
             # SG is canonical by default
-            # IT is canonucal by default
+            # IT is canonical by default
         }
 
         if self.scheme in schemes:
diff --git a/tests/cc/test_cc_misc.py b/tests/cc/test_cc_misc.py
index 50bd81c0..ca32fe91 100644
--- a/tests/cc/test_cc_misc.py
+++ b/tests/cc/test_cc_misc.py
@@ -31,7 +31,7 @@ def test_canonicalize_es():
 
 
 def test_canonicalize_in():
-    assert canonicalize("IC3S/KOL01/ADVA/EAL2/0520/0021 /CR", "IN") == "IC3S/KOL01/ADVA/EAL2/0520/0021/CR"
+    assert canonicalize("IC3S/KOL01/ADVA/EAL2/0520/0021 /CR", "IN") == "IC3S/KOL01/ADVA/EAL2/0520/0021"
 
 
 def test_canonicalize_se():