test: add regression tests for passport number normalization (#253)

tests/tests/test_normalization.py

@@ -5,6 +5,7 @@
 
 import sys
 import malaya
+import malaya.normalizer.rules
 import logging
 
 logging.basicConfig(level=logging.DEBUG)
@@ -23,3 +24,30 @@ def test_normalization():
     normalizer = malaya.normalize.normalizer(corrector)
     string = 'boleh dtg 8pagi esok tak atau minggu depan? 2 oktober 2019 2pm, tlong bayar rm 3.2k sekali tau'
     normalizer.normalize(string)
+
+
+def test_passport_normalization():
+    """
+    Test that passport numbers are normalized digit-by-digit,
+    not expanded as cardinal numbers. Regression test for #253.
+    """
+    normalizer = malaya.normalizer.rules.load()
+
+    # Passport numbers should be spelled out character-by-character
+    test_cases = [
+        'your passport is EL9568719',
+        'ini adalah passport saya A20964577',
+    ]
+
+    for s in test_cases:
+        result = normalizer.normalize(s)
+        normalized = result['normalize']
+
+        # Should NOT contain cardinal expansions like "million", "thousand", "juta", "ribu"
+        cardinal_words = ['million', 'thousand', 'hundred', 'juta', 'ribu', 'ratus']
+        for word in cardinal_words:
+            assert word not in normalized.lower(), (
+                f'Passport number in "{s}" was expanded as cardinal number '
+                f'(found "{word}" in "{normalized}"). '
+                f'Expected digit-by-digit expansion.'
+            )