[Feature Store] Add MinMaxLenValidator and RegexValidator

mlrun/features.py

@@ -1,3 +1,4 @@
+import re
 from typing import Dict, List, Optional
 
 from .data_types import ValueType
@@ -146,7 +147,108 @@ def check(self, value):
         return ok, args
 
 
+class MinMaxLenValidator(Validator):
+    """Validate min/max length value ranges"""
+
+    kind = "minmaxlen"
+    _dict_fields = Validator._dict_fields + ["min", "max"]
+
+    def __init__(self, check_type=None, severity=None, min=None, max=None):
+        """Validate min/max length value ranges
+
+        example::
+
+            from mlrun.features import MinMaxValidator
+
+            # Add length validator to the feature 'ticker', where valid
+            # minimal length is 1 and maximal length is 10
+            quotes_set["ticker"].validator = MinMaxLenValidator(
+                min=1,
+                max=10,
+                severity="info"
+            )
+
+        :param check_type:  ..
+        :param severity:    severity name e.g. info, warning, etc.
+        :param min:         minimal valid length size
+        :param max:         maximal valid length size
+        """
+        super().__init__(check_type, severity)
+        self.min = min
+        self.max = max
+
+    def check(self, value):
+        ok, args = super().check(value)
+        if ok:
+            if self.min is not None:
+                if len(value) < self.min:
+                    return (
+                        False,
+                        {
+                            "message": "Length value is smaller than min",
+                            "min": self.min,
+                            "length value": len(value),
+                        },
+                    )
+            if self.max is not None:
+                if len(value) > self.max:
+                    return (
+                        False,
+                        {
+                            "message": "Length value is greater than max",
+                            "max": self.max,
+                            "length value": len(value),
+                        },
+                    )
+        return ok, args
+
+
+class RegexValidator(Validator):
+    """Validate value based on regular expression"""
+
+    kind = "regex"
+    _dict_fields = Validator._dict_fields + ["regex"]
+
+    def __init__(self, check_type=None, severity=None, regex=None):
+        """Validate value based on regular expression
+
+        example::
+
+            from mlrun.features import RegexValidator
+
+            # Add regular expression validator to the feature 'name'
+            quotes_set["name"].validator = RegexValidator(
+                regex=r"(\b[A-Za-z]{1}[0-9]{7}\b)",
+                severity="info"
+            )
+
+        :param check_type:  ..
+        :param severity:    severity name e.g. info, warning, etc.
+        :param regex:       regular expression for validation
+        """
+        super().__init__(check_type, severity)
+        self.regex = regex
+        self.regex_compile = re.compile(self.regex)
+
+    def check(self, value):
+        ok, args = super().check(value)
+        if ok:
+            if self.regex is not None:
+                if not re.fullmatch(self.regex_compile, value):
+                    return (
+                        False,
+                        {
+                            "message": "Value is not valid with regular expression",
+                            "regexp": self.regex,
+                            "value": value,
+                        },
+                    )
+        return ok, args
+
+
 validator_kinds = {
     "": Validator,
     "minmax": MinMaxValidator,
+    "minmaxlen": MinMaxLenValidator,
+    "regex": RegexValidator,
 }

tests/system/feature_store/test_feature_store.py

@@ -39,7 +39,7 @@
 from mlrun.feature_store.feature_set import aggregates_step
 from mlrun.feature_store.feature_vector import FixedWindowType
 from mlrun.feature_store.steps import FeaturesetValidator
-from mlrun.features import MinMaxValidator
+from mlrun.features import MinMaxLenValidator, MinMaxValidator
 from tests.system.base import TestMLRunSystem
 
 from .data_sample import quotes, stocks, trades
@@ -141,6 +141,9 @@ def _ingest_quotes_featureset(self):
         self._logger.info(f"quotes spec: {quotes_set.spec.to_yaml()}")
         assert df["zz"].mean() == 9, "map didnt set the zz column properly"
         quotes_set["bid"].validator = MinMaxValidator(min=52, severity="info")
+        quotes_set["ticker"].validator = MinMaxLenValidator(
+            min=1, max=10, severity="info"
+        )
 
         quotes_set.plot(
             str(self.results_path / "pipe.png"), rankdir="LR", with_targets=True