test.py

# filename: test_presidio_en.py

from presidio_analyzer import AnalyzerEngine
from presidio_anonymizer import AnonymizerEngine


# --------------------------------------------------
# 1) Presidio 엔진 초기화
# --------------------------------------------------
analyzer = AnalyzerEngine()
anonymizer = AnonymizerEngine()


# --------------------------------------------------
# 2) 테스트 입력 텍스트 (영어)
# --------------------------------------------------
text = """
User John Doe's email is john.doe@example.com and his phone number is 415-555-2671.
The temporary password is qsd123! and credit card number is 4111-1111-1111-1111.
"""


# --------------------------------------------------
# 3) Presidio 분석 수행
# --------------------------------------------------
results = analyzer.analyze(text=text, language="en")

print("\n===== [1] 분석 결과 =====")
for r in results:
    print(f"{r.entity_type} ({r.start}-{r.end}) score={r.score:.2f}")


# --------------------------------------------------
# 4) Presidio 마스킹 수행
# --------------------------------------------------
anonymized = anonymizer.anonymize(
    text=text,
    analyzer_results=results
)

print("\n===== [2] 마스킹 결과 =====")
print(anonymized.text)