Feature control checks (#134)

* Formated and refactored

* Added control completeness

* Added control checks

cuallee/__init__.py

@@ -148,7 +148,7 @@ def validate_data_types(self, rules: List[Rule], dataframe: Any) -> bool:
     def summary(self, check: Any, dataframe: Any) -> Any:
         """Computes all predicates and expressions for check summary"""
 
-
+        
 class Check:
     def __init__(
         self,
@@ -627,27 +627,12 @@ def validate(self, dataframe: Any):
         ), "Invalid data types between rules and dataframe"
         return self.compute_engine.summary(self, dataframe)
 
-    # def samples(self, dataframe: Any, rule_index: int = None) -> Any:
-    #     if not rule_index:
-    #         return reduce(
-    #             DataFrame.unionAll,
-    #             [dataframe.filter(predicate) for predicate in self.predicates],
-    #         ).drop_duplicates()
-    #     elif isinstance(rule_index, int):
-    #         return reduce(
-    #             DataFrame.unionAll,
-    #             [
-    #                 dataframe.filter(predicate)
-    #                 for index, predicate in enumerate(self.predicates, 1)
-    #                 if rule_index == index
-    #             ],
-    #         )
-    #     elif isinstance(rule_index, list):
-    #         return reduce(
-    #             DataFrame.unionAll,
-    #             [
-    #                 dataframe.filter(predicate)
-    #                 for index, predicate in enumerate(self.predicates, 1)
-    #                 if index in rule_index
-    #             ],
-    #         ).drop_duplicates()
+
+
+class Control():
+    @staticmethod
+    def completeness(dataframe):
+        """Control of null values on data frames"""
+        check = Check(CheckLevel.WARNING, "Completeness")
+        [check.is_complete(c) for c in dataframe.columns];
+        return check.validate(dataframe)

cuallee/polars_validation.py

@@ -451,7 +451,7 @@ def _calculate_pass_rate(result, nrows):
                     if result.imag > nrows:
                         return nrows / result.imag
                     else:
-                        return result.imag / nrows
+                        return nrows / (nrows + result.imag)
                 else:
                     return 1.0
 

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "cuallee"
-version = "0.6.1"
+version = "0.7.0"
 authors = [
   { name="Herminio Vazquez", email="canimus@gmail.com"},
   { name="Virginie Grosboillot", email="vestalisvirginis@gmail.com" }

setup.cfg

@@ -1,5 +1,5 @@
 [metadata]
 name = cuallee
-version = 0.6.1
+version = 0.7.0
 [options]
 packages = find:

test/unit/class_control/test_methods.py

@@ -0,0 +1,9 @@
+from cuallee import Control
+import pandas as pd
+
+def test_has_completeness():
+    assert hasattr(Control, "completeness")
+
+def test_completeness_result():
+    df = pd.DataFrame({"A" : [1,2,3,4,5]})
+    assert Control.completeness(df).status.eq("PASS").all()

test/unit/polars_dataframe/test_is_daily.py

@@ -9,23 +9,23 @@ def test_positive(check: Check):
     df = pl.DataFrame({
         "id": pl.date_range(start=date(2022,11,20), end=date(2022,11,26), interval="1d", eager=True)
     })
-    result = check.validate(df).select(pl.col("status")) == "PASS"
-    assert all(result.to_series().to_list())
+
+    assert check.validate(df).select(pl.col("status").eq("PASS").all()).item() 
 
 
 def test_negative(check: Check):
     check.is_daily("id")
     df = pl.DataFrame(
         {"id": [datetime.today() + timedelta(days=i) for i in range(1, 10, 2)]}
     )
-    result = check.validate(df).select(pl.col("status")) == "FAIL"
-    assert all(result.to_series().to_list())
+    assert check.validate(df).select(pl.col("status").eq("FAIL").all()).item() 
 
 
 def test_coverage(check: Check):
     check.is_daily("id", pct=0.6)
     df = pl.DataFrame(
         {"id": [datetime(2022, 12, 12) + timedelta(days=i) for i in range(1, 10, 2)]}
     )
-    result = check.validate(df).select(pl.col("status")) == "PASS"
-    assert all(result.to_series().to_list())
+
+    assert check.validate(df).select(pl.col("status").eq("PASS").all()).item() 
+