SIENTIAPDE-1579: Switched from replace logic to validation + mapping

This commit is contained in:
Kou Kinoshita
2026-02-18 18:57:41 -03:00
parent e7e851dbb5
commit 14f3fbadbe
5 changed files with 105 additions and 34 deletions

View File

@@ -17,14 +17,35 @@ NORMALIZATION = 'Normalization'
FEATURE_CREATION = 'Feature Creation'
LAG_CREATION = 'Lag Creation'
# Allowed frontend date formats and their strftime equivalents (single source of truth)
FRONTEND_DATE_FORMAT_TO_STRFTIME = {
'dd/MM/yyyy HH:mm:ss': '%d/%m/%Y %H:%M:%S',
'MM/dd/yyyy HH:mm:ss': '%m/%d/%Y %H:%M:%S',
'yyyy/MM/dd HH:mm:ss': '%Y/%m/%d %H:%M:%S',
'dd-MM-yyyy HH:mm:ss': '%d-%m-%Y %H:%M:%S',
'MM-dd-yyyy HH:mm:ss': '%m-%d-%Y %H:%M:%S',
'yyyy-MM-dd HH:mm:ss': '%Y-%m-%d %H:%M:%S',
}
ALLOWED_FRONTEND_DATE_FORMATS = frozenset(FRONTEND_DATE_FORMAT_TO_STRFTIME.keys())
def validate_frontend_date_format(fmt: str | None) -> None:
"""Raise ValueError if fmt is set and not one of the allowed frontend date formats."""
if not fmt or not fmt.strip():
return
if fmt not in ALLOWED_FRONTEND_DATE_FORMATS:
allowed = ', '.join(sorted(ALLOWED_FRONTEND_DATE_FORMATS))
raise ValueError(
f'Invalid date_format "{fmt}". Allowed formats: {allowed}'
)
def _frontend_date_format_to_strftime(fmt: str | None) -> str | None:
"""Convert front-end date format (e.g. dd/MM/yyyy HH:mm:ss) to Python strftime."""
"""Convert front-end date format to Python strftime. Validates format; returns None for empty."""
if not fmt:
return None
out = fmt.replace('yyyy', '%Y').replace('MM', '%m').replace('dd', '%d')
out = out.replace('HH', '%H').replace('mm', '%M').replace('ss', '%S')
return out
validate_frontend_date_format(fmt)
return FRONTEND_DATE_FORMAT_TO_STRFTIME[fmt]
class LinearRegressionModel(BaseEstimator, TransformerMixin):

View File

@@ -1,6 +1,8 @@
from dataclasses import dataclass
from typing import Any
from model_manager.sientia.models import validate_frontend_date_format
# Model name constants
MODEL_LINEAR_REGRESSION = 'Linear Regression'
MODEL_POLYNOMIAL_REGRESSION = 'Polynomial Regression'
@@ -217,6 +219,7 @@ class TrainModelParams:
self._validate_intervals_and_dates()
self._validate_limits()
self._validate_required_strings()
self._validate_date_format()
def _validate_numeric_ranges(self) -> None:
"""Validate numeric parameters are within acceptable ranges."""
@@ -327,3 +330,8 @@ class TrainModelParams:
if not self.experiment_name.strip():
raise ValueError('experiment_name cannot be empty or whitespace')
def _validate_date_format(self) -> None:
"""Validate date_format is one of the allowed frontend formats when set."""
if self.date_format:
validate_frontend_date_format(self.date_format)

View File

@@ -7,34 +7,56 @@ import pandas as pd
from pytest import raises
from model_manager.sientia.models import (
ALLOWED_FRONTEND_DATE_FORMATS,
FRONTEND_DATE_FORMAT_TO_STRFTIME,
DataPreprocessor,
LinearRegressionModel,
_frontend_date_format_to_strftime,
validate_frontend_date_format,
)
class TestFrontendDateFormatToStrftime:
"""Tests for _frontend_date_format_to_strftime (models module)."""
def test_none_or_empty_returns_none_or_empty(self):
"""None or empty string returns None or falsy."""
def test_none_or_empty_returns_none(self):
"""None or empty string returns None."""
assert _frontend_date_format_to_strftime(None) is None
assert _frontend_date_format_to_strftime('') is None
def test_dd_mm_yyyy_hh_mm_ss(self):
"""Converts dd/MM/yyyy HH:mm:ss to strftime."""
result = _frontend_date_format_to_strftime('dd/MM/yyyy HH:mm:ss')
assert result == '%d/%m/%Y %H:%M:%S'
def test_all_six_allowed_formats_convert_correctly(self):
"""All allowed frontend formats map to expected strftime."""
for frontend_fmt, strftime_fmt in FRONTEND_DATE_FORMAT_TO_STRFTIME.items():
assert _frontend_date_format_to_strftime(frontend_fmt) == strftime_fmt
def test_yyyy_mm_dd(self):
"""Converts yyyy-MM-dd to strftime."""
result = _frontend_date_format_to_strftime('yyyy-MM-dd')
assert result == '%Y-%m-%d'
def test_invalid_format_raises(self):
"""Invalid format raises ValueError with allowed list in message."""
with raises(ValueError, match='Invalid date_format'):
_frontend_date_format_to_strftime('yyyy-MM-dd')
with raises(ValueError, match='Allowed formats'):
_frontend_date_format_to_strftime('invalid')
def test_iso_datetime(self):
"""Converts yyyy-MM-dd HH:mm:ss to strftime."""
result = _frontend_date_format_to_strftime('yyyy-MM-dd HH:mm:ss')
assert result == '%Y-%m-%d %H:%M:%S'
class TestValidateFrontendDateFormat:
"""Tests for validate_frontend_date_format."""
def test_none_or_empty_does_not_raise(self):
"""None or empty string does not raise."""
validate_frontend_date_format(None)
validate_frontend_date_format('')
validate_frontend_date_format(' ')
def test_allowed_formats_do_not_raise(self):
"""All allowed formats pass validation."""
for fmt in ALLOWED_FRONTEND_DATE_FORMATS:
validate_frontend_date_format(fmt)
def test_invalid_format_raises(self):
"""Invalid format raises ValueError."""
with raises(ValueError, match='Invalid date_format'):
validate_frontend_date_format('yyyy-MM-dd')
with raises(ValueError, match='Invalid date_format'):
validate_frontend_date_format('custom-bad-format')
class _IterableWithContains:

View File

@@ -555,6 +555,27 @@ def test_validate_business_rules_valid_start_and_end_date(valid_train_params_dic
params.validate_business_rules() # Should not raise
def test_validate_business_rules_invalid_date_format(valid_train_params_dict):
"""Test validate_business_rules raises when date_format is not allowed."""
from model_manager.utils.models.train_model_params import TrainModelParams
params = TrainModelParams.from_dict(valid_train_params_dict)
params.date_format = 'yyyy-MM-dd'
with pytest.raises(ValueError, match='Invalid date_format'):
params.validate_business_rules()
def test_validate_business_rules_valid_date_format(valid_train_params_dict):
"""Test validate_business_rules accepts allowed date_format."""
from model_manager.utils.models.train_model_params import TrainModelParams
params = TrainModelParams.from_dict(valid_train_params_dict)
params.date_format = 'yyyy-MM-dd HH:mm:ss'
params.validate_business_rules() # Should not raise
def test_validate_business_rules_polynomial_regression_valid(valid_train_params_dict):
"""Test validate_business_rules accepts valid Polynomial Regression config."""
from model_manager.utils.models.train_model_params import TrainModelParams

View File

@@ -1125,27 +1125,23 @@ class TestExtractModelEquationPolynomial:
class TestFrontendFormatToStrftime:
"""Tests for _frontend_format_to_strftime."""
"""Tests for _frontend_format_to_strftime (alias from models)."""
def test_empty_or_none_returns_none(self):
"""Empty string or None returns None (same behavior as models module)."""
assert _frontend_format_to_strftime('') is None
assert _frontend_format_to_strftime(None) is None
def test_dd_mm_yyyy_hh_mm_ss(self):
"""Converts dd/MM/yyyy HH:mm:ss to strftime."""
result = _frontend_format_to_strftime('dd/MM/yyyy HH:mm:ss')
assert result == '%d/%m/%Y %H:%M:%S'
def test_allowed_formats_convert_correctly(self):
"""Allowed frontend formats convert to strftime via mapper."""
assert _frontend_format_to_strftime('dd/MM/yyyy HH:mm:ss') == '%d/%m/%Y %H:%M:%S'
assert _frontend_format_to_strftime('yyyy-MM-dd HH:mm:ss') == '%Y-%m-%d %H:%M:%S'
assert _frontend_format_to_strftime('MM/dd/yyyy HH:mm:ss') == '%m/%d/%Y %H:%M:%S'
def test_yyyy_mm_dd(self):
"""Converts yyyy-MM-dd to strftime."""
result = _frontend_format_to_strftime('yyyy-MM-dd')
assert result == '%Y-%m-%d'
def test_iso_like_datetime(self):
"""Converts yyyy-MM-dd HH:mm:ss to strftime."""
result = _frontend_format_to_strftime('yyyy-MM-dd HH:mm:ss')
assert result == '%Y-%m-%d %H:%M:%S'
def test_invalid_format_raises(self):
"""Invalid date format raises ValueError."""
with pytest.raises(ValueError, match='Invalid date_format'):
_frontend_format_to_strftime('yyyy-MM-dd')
class TestEnsureDateColumnParsed:
@@ -1215,8 +1211,11 @@ class TestEnsureDateColumnParsed:
def test_invalid_values_coerced_to_nat(self, date_params):
"""Invalid date strings are coerced to NaT when format is set."""
date_params.date_format = 'yyyy-MM-dd'
data = pd.DataFrame({'ts': ['2023-01-01', 'not-a-date', '2023-12-31'], 'x': [1, 2, 3]})
date_params.date_format = 'yyyy-MM-dd HH:mm:ss'
data = pd.DataFrame({
'ts': ['2023-01-01 00:00:00', 'not-a-date', '2023-12-31 00:00:00'],
'x': [1, 2, 3],
})
result = _ensure_date_column_parsed(data, date_params)
assert pd.isna(result['ts'].iloc[1])
assert result['ts'].iloc[0].year == 2023