Skip to content

Commit

Permalink
Add SSN and bank account to the default pii list
Browse files Browse the repository at this point in the history
  • Loading branch information
Anthony Naddeo committed Feb 18, 2024
1 parent b1f92c3 commit b57db35
Show file tree
Hide file tree
Showing 3 changed files with 18 additions and 2 deletions.
2 changes: 1 addition & 1 deletion Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ clean: ## Clean the project and generated files
rm -rf ./dist/

test: ## Run the tests
poetry run pytest tests -o log_level=INFO -o log_cli=true
poetry run pytest tests -vvv -o log_level=INFO -o log_cli=true

test-cache-constraint:
docker build -f ./Dockerfile.cache_test . -t langkit_cache_test
Expand Down
2 changes: 1 addition & 1 deletion langkit/metrics/pii.py
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ def to_dict(self) -> List[Dict[str, Any]]:
]


__default_entities = ["PHONE_NUMBER", "EMAIL_ADDRESS", "CREDIT_CARD", "IP_ADDRESS"]
__default_entities = ["PHONE_NUMBER", "EMAIL_ADDRESS", "CREDIT_CARD", "IP_ADDRESS", "US_SSN", "US_BANK_NUMBER"]


def __create_pii_metric_name(input_name: str, entity: str) -> str:
Expand Down
16 changes: 16 additions & 0 deletions tests/langkit/metrics/test_pii.py
Original file line number Diff line number Diff line change
Expand Up @@ -67,12 +67,16 @@ def test_prompt_response_pii_metric_whylogs():
"prompt.pii.ip_address",
"prompt.pii.phone_number",
"prompt.pii.redacted",
"prompt.pii.us_bank_number",
"prompt.pii.us_ssn",
"response",
"response.pii.credit_card",
"response.pii.email_address",
"response.pii.ip_address",
"response.pii.phone_number",
"response.pii.redacted",
"response.pii.us_bank_number",
"response.pii.us_ssn",
]

logged = _log(item=df, conf=all_config)
Expand All @@ -84,10 +88,14 @@ def test_prompt_response_pii_metric_whylogs():
assert logged["distribution/max"]["prompt.pii.email_address"] == 1
assert logged["distribution/max"]["prompt.pii.credit_card"] == 0.0
assert logged["distribution/max"]["prompt.pii.ip_address"] == 0.0
assert logged["distribution/max"]["prompt.pii.us_ssn"] == 0.0
assert logged["distribution/max"]["prompt.pii.us_bank_number"] == 0.0
assert logged["distribution/max"]["response.pii.phone_number"] == 0.0
assert logged["distribution/max"]["response.pii.email_address"] == 0.0
assert logged["distribution/max"]["response.pii.credit_card"] == 0.0
assert logged["distribution/max"]["response.pii.ip_address"] == 0.0
assert logged["distribution/max"]["response.pii.us_ssn"] == 0.0
assert logged["distribution/max"]["response.pii.us_bank_number"] == 0.0


def test_prompt_response_pii_metric():
Expand All @@ -107,11 +115,15 @@ def test_prompt_response_pii_metric():
"prompt.pii.email_address",
"prompt.pii.credit_card",
"prompt.pii.ip_address",
"prompt.pii.us_ssn",
"prompt.pii.us_bank_number",
"prompt.pii.redacted",
"response.pii.phone_number",
"response.pii.email_address",
"response.pii.credit_card",
"response.pii.ip_address",
"response.pii.us_ssn",
"response.pii.us_bank_number",
"response.pii.redacted",
"id",
]
Expand All @@ -128,10 +140,14 @@ def test_prompt_response_pii_metric():
assert logged["prompt.pii.email_address"][0] == 1
assert logged["prompt.pii.credit_card"][0] == 0
assert logged["prompt.pii.ip_address"][0] == 0
assert logged["prompt.pii.us_ssn"][0] == 0
assert logged["prompt.pii.us_bank_number"][0] == 0
assert logged["response.pii.phone_number"][0] == 0
assert logged["response.pii.email_address"][0] == 0
assert logged["response.pii.credit_card"][0] == 0
assert logged["response.pii.ip_address"][0] == 0
assert logged["response.pii.us_ssn"][0] == 0
assert logged["response.pii.us_bank_number"][0] == 0
assert logged["prompt.pii.redacted"][0] == "Hey! Here is my phone number: <PHONE_NUMBER>, and my email is <EMAIL_ADDRESS>."
assert logged["response.pii.redacted"][0] is None
assert logged["id"][0] == "0"

0 comments on commit b57db35

Please sign in to comment.