Add PL PESEL recognizer (#1209)

presidio-analyzer/presidio_analyzer/predefined_recognizers/__init__.py

@@ -33,6 +33,7 @@
 from .au_tfn_recognizer import AuTfnRecognizer
 from .au_medicare_recognizer import AuMedicareRecognizer
 from .in_pan_recognizer import InPanRecognizer
+from .pl_pesel_recognizer import PlPeselRecognizer
 
 
 NLP_RECOGNIZERS = {
@@ -74,4 +75,5 @@
     "ItIdentityCardRecognizer",
     "ItPassportRecognizer",
     "InPanRecognizer",
+    "PlPeselRecognizer",
 ]

presidio-analyzer/presidio_analyzer/predefined_recognizers/pl_pesel_recognizer.py

@@ -0,0 +1,51 @@
+from typing import List, Optional
+
+from presidio_analyzer import Pattern, PatternRecognizer
+
+
+class PlPeselRecognizer(PatternRecognizer):
+    """
+    Recognize PESEL number using regex and checksum.
+
+    For more information about PESEL: https://en.wikipedia.org/wiki/PESEL
+
+    :param patterns: List of patterns to be used by this recognizer
+    :param context: List of context words to increase confidence in detection
+    :param supported_language: Language this recognizer supports
+    :param supported_entity: The entity this recognizer can detect
+    """
+
+    PATTERNS = [
+        Pattern(
+            "PESEL",
+            r"[0-9]{2}([02468][1-9]|[13579][012])(0[1-9]|1[0-9]|2[0-9]|3[01])[0-9]{5}",
+            0.4,
+        ),
+    ]
+
+    CONTEXT = ["PESEL"]
+
+    def __init__(
+        self,
+        patterns: Optional[List[Pattern]] = None,
+        context: Optional[List[str]] = None,
+        supported_language: str = "pl",
+        supported_entity: str = "PL_PESEL",
+    ):
+        patterns = patterns if patterns else self.PATTERNS
+        context = context if context else self.CONTEXT
+        super().__init__(
+            supported_entity=supported_entity,
+            patterns=patterns,
+            context=context,
+            supported_language=supported_language,
+        )
+
+    def validate_result(self, pattern_text: str) -> bool:  # noqa D102
+        digits = [int(digit) for digit in pattern_text]
+        weights = [1, 3, 7, 9, 1, 3, 7, 9, 1, 3]
+
+        checksum = sum(digit * weight for digit, weight in zip(digits[:10], weights))
+        checksum %= 10
+
+        return checksum == digits[10]

presidio-analyzer/presidio_analyzer/recognizer_registry.py

@@ -44,6 +44,7 @@
     ItPassportRecognizer,
     ItIdentityCardRecognizer,
     InPanRecognizer,
+    PlPeselRecognizer,
 )
 
 logger = logging.getLogger("presidio-analyzer")
@@ -109,6 +110,7 @@ def load_predefined_recognizers(
                 ItIdentityCardRecognizer,
                 ItPassportRecognizer,
             ],
+            "pl": [PlPeselRecognizer],
             "ALL": [
                 CreditCardRecognizer,
                 CryptoRecognizer,

presidio-analyzer/tests/test_pl_pesel_recognizer.py

@@ -0,0 +1,37 @@
+import pytest
+
+from tests import assert_result
+from presidio_analyzer.predefined_recognizers import PlPeselRecognizer
+
+
+@pytest.fixture(scope="module")
+def recognizer():
+    return PlPeselRecognizer()
+
+
+@pytest.fixture(scope="module")
+def entities():
+    return ["PL_PESEL"]
+
+
+@pytest.mark.parametrize(
+    "text, expected_len, expected_positions",
+    [
+        # fmt: off
+        # valid PESEL scores
+        ("11111111114", 1, ((0, 11),),),
+        ("My pesel is 11111111114.", 1, ((12, 23), )),
+        # invalid PESEL scores
+        ("1111321111", 0, ()),
+        ("11110021111", 0, ()),
+        ("11-11-11-11114", 0, ()),
+        # fmt: on
+    ],
+)
+def test_when_all_pl_pesels_then_succeed(
+    text, expected_len, expected_positions, recognizer, entities, max_score
+):
+    results = recognizer.analyze(text, entities)
+    assert len(results) == expected_len
+    for res, (st_pos, fn_pos) in zip(results, expected_positions):
+        assert_result(res, entities[0], st_pos, fn_pos, max_score)