duckguard 2.0.0__py3-none-any.whl → 2.2.0__py3-none-any.whl

duckguard/__init__.py

@@ -3,7 +3,7 @@ DuckGuard - Data quality that just works.
 
 A Python-native data quality tool built on DuckDB for speed.
 Features YAML-based rules, semantic type detection, data contracts,
-and anomaly detection.
+anomaly detection, notifications, and dbt integration.
 
 Quick Start:
     # Python API
@@ -12,61 +12,80 @@ Quick Start:
     assert orders.row_count > 0
     assert orders.customer_id.null_percent == 0
 
+    # With row-level error capture
+    result = orders.quantity.between(1, 100)
+    if not result:
+        print(result.summary())  # See which rows failed
+
+    # Notifications
+    from duckguard.notifications import SlackNotifier
+    slack = SlackNotifier(webhook_url="...")
+    slack.send_failure_alert(result)
+
     # CLI
     $ duckguard check data.csv
     $ duckguard discover data.csv --output duckguard.yaml
     $ duckguard contract generate data.csv
 
-Documentation: https://github.com/duckguard/duckguard
+Documentation: https://github.com/XDataHubAI/duckguard
 """
 
 # Core classes
-from duckguard.core.dataset import Dataset
+# Anomaly detection
+from duckguard.anomaly import (
+    AnomalyDetector,
+    AnomalyResult,
+    detect_anomalies,
+)
+
+# Connectors
+from duckguard.connectors import connect
+
+# Data contracts
+from duckguard.contracts import (
+    DataContract,
+    diff_contracts,
+    generate_contract,
+    load_contract,
+    validate_contract,
+)
 from duckguard.core.column import Column
+from duckguard.core.dataset import Dataset
 from duckguard.core.engine import DuckGuardEngine
-from duckguard.core.result import ValidationResult, CheckResult
+from duckguard.core.result import CheckResult, FailedRow, ValidationResult
 from duckguard.core.scoring import QualityScore, QualityScorer, score
 
-# Connectors
-from duckguard.connectors import connect
+# Error classes
+from duckguard.errors import (
+    ColumnNotFoundError,
+    ContractViolationError,
+    DuckGuardError,
+    RuleParseError,
+    UnsupportedConnectorError,
+    ValidationError,
+)
 
 # Profiling
-from duckguard.profiler import profile, AutoProfiler
+from duckguard.profiler import AutoProfiler, profile
 
 # Rules (YAML-based)
 from duckguard.rules import (
-    load_rules,
-    load_rules_from_string,
+    RuleSet,
     execute_rules,
     generate_rules,
-    RuleSet,
+    load_rules,
+    load_rules_from_string,
 )
 
 # Semantic type detection
 from duckguard.semantic import (
-    SemanticType,
     SemanticAnalyzer,
+    SemanticType,
     detect_type,
     detect_types_for_dataset,
 )
 
-# Data contracts
-from duckguard.contracts import (
-    DataContract,
-    load_contract,
-    validate_contract,
-    generate_contract,
-    diff_contracts,
-)
-
-# Anomaly detection
-from duckguard.anomaly import (
-    AnomalyDetector,
-    AnomalyResult,
-    detect_anomalies,
-)
-
-__version__ = "2.0.0"
+__version__ = "2.2.0"
 
 __all__ = [
     # Core classes
@@ -75,6 +94,7 @@ __all__ = [
     "DuckGuardEngine",
     "ValidationResult",
     "CheckResult",
+    "FailedRow",
     # Scoring
     "QualityScore",
     "QualityScorer",
@@ -105,6 +125,13 @@ __all__ = [
     "AnomalyDetector",
     "AnomalyResult",
     "detect_anomalies",
+    # Errors
+    "DuckGuardError",
+    "ColumnNotFoundError",
+    "ContractViolationError",
+    "RuleParseError",
+    "UnsupportedConnectorError",
+    "ValidationError",
     # Version
     "__version__",
 ]

duckguard/anomaly/__init__.py

@@ -17,9 +17,9 @@ from duckguard.anomaly.detector import (
     detect_column_anomalies,
 )
 from duckguard.anomaly.methods import (
-    ZScoreMethod,
     IQRMethod,
     PercentChangeMethod,
+    ZScoreMethod,
 )
 
 __all__ = [

duckguard/anomaly/detector.py

@@ -10,15 +10,11 @@ from datetime import datetime
 from enum import Enum
 from typing import Any
 
-from duckguard.core.dataset import Dataset
 from duckguard.anomaly.methods import (
-    AnomalyMethod,
     AnomalyScore,
-    ZScoreMethod,
-    IQRMethod,
-    PercentChangeMethod,
     create_method,
 )
+from duckguard.core.dataset import Dataset
 
 
 class AnomalyType(Enum):

duckguard/anomaly/methods.py

@@ -5,10 +5,10 @@ Implements various statistical methods for detecting anomalies in data.
 
 from __future__ import annotations
 
+import math
 from abc import ABC, abstractmethod
 from dataclasses import dataclass, field
 from typing import Any
-import math
 
 
 @dataclass
@@ -177,8 +177,6 @@ class IQRMethod(AnomalyMethod):
         if not clean_values:
             return
 
-        n = len(clean_values)
-
         # Calculate Q1 and Q3
         self._q1 = self._percentile(clean_values, 25)
         self._q3 = self._percentile(clean_values, 75)