dcs-sdk 1.6.5__tar.gz → 1.6.6__tar.gz

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dcs-sdk
-Version: 1.6.5
+Version: 1.6.6
 Summary: SDK for DataChecks
 Author: Waterdip Labs
 Author-email: hello@waterdip.ai
@@ -84,7 +84,7 @@ Requires-Dist: vertica-python (>=1.4.0) ; extra == "vertica" or extra == "all-db
 Description-Content-Type: text/markdown
 
 <h1 align="center">
-  DCS SDK v1.6.4
+  DCS SDK v1.6.6
 </h1>
 
 > SDK for DataChecks

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/README.md

@@ -1,5 +1,5 @@
 <h1 align="center">
-  DCS SDK v1.6.4
+  DCS SDK v1.6.6
 </h1>
 
 > SDK for DataChecks

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/dcs_core/integrations/databases/mssql.py

@@ -13,6 +13,7 @@
 #  limitations under the License.
 
 import datetime
+import math
 from decimal import Decimal
 from typing import Any, Dict, List, Optional, Tuple, Union
 from uuid import UUID
@@ -706,13 +707,15 @@ class MssqlDataSource(SQLDataSource):
         cursor = self.connection.cursor()
         try:
             cursor.execute(query)
-            columns = [column[0] for column in cursor.description]
-            result_row = cursor.fetchone()
+            if cursor.description:
+                columns = [column[0] for column in cursor.description]
+                result_row = cursor.fetchone()
+                row = dict(zip(columns, result_row)) if result_row else {}
+            else:
+                row = {}
         finally:
             cursor.close()
 
-        row = dict(zip(columns, result_row))
-
         def _normalize_metrics(value):
             """Safely normalize DB metric values for JSON serialization."""
             if value is None:
@@ -737,11 +740,158 @@ class MssqlDataSource(SQLDataSource):
             col_metrics = {}
 
             for key, value in row.items():
-                if key.startswith(f"{name}_"):
-                    metric_name = key[len(name) + 1 :]
+                clean_key = key.replace("[", "").replace("]", "")
+                if clean_key.startswith(f"{name}_"):
+                    metric_name = clean_key[len(name) + 1 :]
                     col_metrics[metric_name] = _normalize_metrics(value)
 
             column_wise.append({"column_name": name, "metrics": col_metrics})
+
+        for col_data in column_wise:
+            metrics = col_data["metrics"]
+            distinct_count = metrics.get("distinct")
+            col_name = col_data["column_name"]
+
+            dtype = next(c["data_type"].lower() for c in column_info if c["column_name"] == col_name)
+
+            quoted = self.quote_column(col_name)
+
+            is_dtype_numeric = (
+                True
+                if dtype
+                in (
+                    "int",
+                    "integer",
+                    "bigint",
+                    "smallint",
+                    "tinyint",
+                    "decimal",
+                    "numeric",
+                    "float",
+                    "real",
+                    "money",
+                    "smallmoney",
+                )
+                else False
+            )
+
+            if is_dtype_numeric:
+                col_min = metrics.get("min")
+                col_max = metrics.get("max")
+
+                if col_min is not None and col_max is not None and col_min != col_max:
+                    bucket_count = 20
+                    bucket_size = (float(col_max) - float(col_min)) / bucket_count
+
+                    bucket_queries = []
+                    for i in range(bucket_count):
+                        start = float(col_min) + i * bucket_size
+                        end = float(col_min) + (i + 1) * bucket_size
+
+                        bucket_queries.append(
+                            f"SUM(CASE WHEN {quoted} >= {start} AND {quoted} < {end} THEN 1 ELSE 0 END) AS bucket_{i}"
+                        )
+
+                    bucket_sql = f"SELECT {', '.join(bucket_queries)} FROM {qualified_table}"
+
+                    try:
+                        bucket_result = self.connection.execute(text(bucket_sql)).fetchone()
+                        distribution = []
+
+                        for i in range(bucket_count):
+                            start_raw = float(col_min) + i * bucket_size
+                            end_raw = float(col_min) + (i + 1) * bucket_size
+
+                            if dtype in ("int", "integer", "bigint", "smallint", "tinyint"):
+                                start = math.floor(start_raw)
+                                end = math.ceil(end_raw)
+                            else:
+                                start = round(start_raw, 2)
+                                end = round(end_raw, 2)
+
+                            count = bucket_result[i] if bucket_result and bucket_result[i] is not None else 0
+
+                            distribution.append(
+                                {
+                                    "col_val": f"{start} - {end}",
+                                    "count": count,
+                                }
+                            )
+
+                        metrics["distribution_graph"] = distribution
+
+                    except Exception as e:
+                        print(f"Failed to generate numeric distribution for {col_name}: {e}")
+
+                continue
+
+            if isinstance(distinct_count, (int, float)) and distinct_count <= 20:
+                if dtype in ("text", "ntext", "xml"):
+                    group_expr = f"CAST({quoted} AS NVARCHAR(MAX))"
+                else:
+                    group_expr = quoted
+
+                dist_query = (
+                    f"SELECT {group_expr}, COUNT(*) "
+                    f"FROM {qualified_table} GROUP BY {group_expr} ORDER BY COUNT(*) DESC"
+                )
+
+                try:
+                    dist_cursor = self.connection.cursor()
+                    dist_cursor.execute(dist_query)
+                    dist_result = dist_cursor.fetchall()
+                    dist_cursor.close()
+
+                    distribution = []
+
+                    for r in dist_result:
+                        val = _normalize_metrics(r[0])
+                        distribution.append(
+                            {
+                                "col_val": val,
+                                "count": r[1],
+                            }
+                        )
+
+                    metrics["distribution_graph"] = distribution
+
+                except Exception as e:
+                    print(f"Failed to generate distribution graph for column {col_name}: {e}")
+
+        for col_data in column_wise:
+            metrics = col_data["metrics"]
+            distinct_count = metrics.get("distinct")
+            col_name = col_data["column_name"]
+            dtype = next(c["data_type"].lower() for c in column_info if c["column_name"] == col_name)
+
+            quoted = self.quote_column(col_name)
+
+            is_dtype_numeric = (
+                True
+                if dtype
+                in (
+                    "int",
+                    "integer",
+                    "bigint",
+                    "smallint",
+                    "tinyint",
+                    "decimal",
+                    "numeric",
+                    "float",
+                    "real",
+                    "money",
+                    "smallmoney",
+                )
+                else False
+            )
+
+            formatted_metrics_data = {
+                "general_data": {key: value for key, value in metrics.items() if key != "distribution_graph"},
+                "is_dtype_numeric": is_dtype_numeric,
+                "distribution_data": metrics.get("distribution_graph", []),
+            }
+            col_data["metrics"] = formatted_metrics_data
+
         return column_wise
 
     def fetch_sample_values_from_database(

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/dcs_core/integrations/databases/postgres.py

@@ -13,6 +13,7 @@
 #  limitations under the License.
 
 import datetime
+import math
 from decimal import Decimal
 from typing import Any, Dict, List, Optional, Tuple
 from uuid import UUID
@@ -411,9 +412,73 @@ class PostgresDataSource(SQLDataSource):
             col_name = col_data["column_name"]
             dtype = next(c["data_type"].lower() for c in column_info if c["column_name"] == col_name)
 
-            if isinstance(distinct_count, (int, float)) and distinct_count < 20:
-                quoted = self.quote_column(col_name)
+            quoted = self.quote_column(col_name)
+
+            is_dtype_numeric = (
+                True
+                if dtype
+                in (
+                    "int",
+                    "integer",
+                    "bigint",
+                    "smallint",
+                    "decimal",
+                    "numeric",
+                    "float",
+                    "double",
+                )
+                else False
+            )
+
+            if is_dtype_numeric:
+                col_min = metrics.get("min")
+                col_max = metrics.get("max")
+
+                if col_min is not None and col_max is not None and col_min != col_max:
+                    bucket_count = 20
+                    bucket_size = (col_max - col_min) / bucket_count
+
+                    bucket_queries = []
+                    for i in range(bucket_count):
+                        start = col_min + i * bucket_size
+                        end = col_min + (i + 1) * bucket_size
+
+                        bucket_queries.append(
+                            f"SUM(CASE WHEN {quoted} >= {start} AND {quoted} < {end} THEN 1 ELSE 0 END) AS bucket_{i}"
+                        )
+
+                    bucket_sql = f"SELECT {', '.join(bucket_queries)} FROM {qualified_table}"
+
+                    try:
+                        bucket_result = self.connection.execute(text(bucket_sql)).fetchone()
+                        distribution = []
+
+                        for i in range(bucket_count):
+                            start_raw = col_min + i * bucket_size
+                            end_raw = col_min + (i + 1) * bucket_size
+                            if dtype in ("int", "integer", "bigint", "smallint"):
+                                start = math.floor(start_raw)
+                                end = math.ceil(end_raw)
+                            else:
+                                start = round(start_raw, 2)
+                                end = round(end_raw, 2)
+                            count = bucket_result[i]
+
+                            distribution.append(
+                                {
+                                    "col_val": f"{start} - {end}",
+                                    "count": count,
+                                }
+                            )
 
+                        metrics["distribution_graph"] = distribution
+
+                    except Exception as e:
+                        print(f"Failed to generate numeric distribution for {col_name}: {e}")
+
+                continue
+
+            if isinstance(distinct_count, (int, float)) and distinct_count <= 20:
                 if dtype in ("json", "jsonb"):
                     group_expr = f"{quoted}::text"
                 else:
@@ -444,8 +509,31 @@ class PostgresDataSource(SQLDataSource):
 
         for col_data in column_wise:
             metrics = col_data["metrics"]
+            distinct_count = metrics.get("distinct")
+            col_name = col_data["column_name"]
+            dtype = next(c["data_type"].lower() for c in column_info if c["column_name"] == col_name)
+
+            quoted = self.quote_column(col_name)
+
+            is_dtype_numeric = (
+                True
+                if dtype
+                in (
+                    "int",
+                    "integer",
+                    "bigint",
+                    "smallint",
+                    "decimal",
+                    "numeric",
+                    "float",
+                    "double",
+                )
+                else False
+            )
+
             formatted_metrics_data = {
                 "general_data": {key: value for key, value in metrics.items() if key != "distribution_graph"},
+                "is_dtype_numeric": is_dtype_numeric,
                 "distribution_data": metrics.get("distribution_graph", []),
             }
             col_data["metrics"] = formatted_metrics_data

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/dcs_sdk/__version__.py

@@ -12,4 +12,4 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 
-__version__ = "1.6.4"
+__version__ = "1.6.6"

{dcs_sdk-1.6.5 → dcs_sdk-1.6.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "dcs-sdk"
-version = "1.6.5"
+version = "1.6.6"
 description = "SDK for DataChecks"
 authors = ["Waterdip Labs <hello@waterdip.ai>"]
 readme = "README.md"