M kovalsky/dax

docs/source/conf.py

@@ -13,7 +13,7 @@
 project = 'semantic-link-labs'
 copyright = '2026, Microsoft and community'
 author = 'Microsoft and community'
-release = '0.14.3'
+release = '0.15.0'
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration

pyproject.toml

@@ -7,7 +7,7 @@ name="semantic-link-labs"
 authors = [
     { name = "Microsoft Corporation" },
 ]
-version="0.14.3"
+version="0.15.0"
 description="Semantic Link Labs for Microsoft Fabric"
 readme="README.md"
 requires-python=">=3.10,<3.13"

src/sempy_labs/_a_lib_info.py

@@ -4,7 +4,7 @@
 from pathlib import Path
 
 lib_name = "semanticlinklabs"
-lib_version = "0.14.3"
+lib_version = "0.15.0"
 
 NUGET_BASE_URL = "https://www.nuget.org/api/v2/package"
 current_dir = Path(__file__).parent

src/sempy_labs/_generate_semantic_model.py

@@ -404,30 +404,30 @@ def deploy_semantic_model(
             )
 
             # Remove mini model annotations from the master model if they exist (cleanup)
-            #ann_to_remove = [
+            # ann_to_remove = [
             #    a.Name
             #    for a in tom.model.Annotations
             #    if a.Name.startswith(icons.prefix_mini)
-            #]
-            #for ann in ann_to_remove:
+            # ]
+            # for ann in ann_to_remove:
             #    tom.remove_annotation(object=tom.model, name=ann)
 
     # Set annotations to the master model
-    #if filters is not None or perspective is not None:
+    # if filters is not None or perspective is not None:
     #    with connect_semantic_model(
     #        dataset=source_dataset_id, workspace=source_workspace_id, readonly=False
     #    ) as tom:
 
     #        ann_name = f"{icons.prefix_mini}_{perspective}"
 
-            # --- Get existing annotation safely ---
+    # --- Get existing annotation safely ---
     #        try:
     #            ann_value = tom.get_annotation_value(object=tom.model, name=ann_name)
     #            ann_list = ast.literal_eval(ann_value) if ann_value else []
     #        except Exception:
     #            ann_list = []
 
-            # --- Build lookup (faster than loop) ---
+    # --- Build lookup (faster than loop) ---
     #        index = {a.get("datasetId"): a for a in ann_list}
 
     #        if target_dataset_id in index:
@@ -455,7 +455,7 @@ def deploy_semantic_model(
     #                }
     #            )
 
-            # --- Save once ---
+    # --- Save once ---
     #        tom.set_annotation(object=tom.model, name=ann_name, value=str(ann_list))
     if refresh_target_dataset:
         refresh_semantic_model(dataset=target_dataset_id, workspace=target_workspace_id)

src/sempy_labs/_model_bpa_rules.py

@@ -3,6 +3,12 @@
 import re
 from typing import Optional
 from sempy._utils._log import log
+from sempy_labs.dax._analysis import (
+    find_fully_qualified_measures,
+    find_non_numeric_aggregations,
+    find_unqualified_columns,
+    uses_function,
+)
 
 
 @log
@@ -434,20 +440,26 @@ def model_bpa_rules(
                 "Measure",
                 "Warning",
                 "Avoid using the IFERROR function",
-                lambda obj, tom: re.search(
-                    r"iferror\s*\(", obj.Expression, flags=re.IGNORECASE
-                ),
+                lambda obj, tom: uses_function(obj.Expression, "IFERROR"),
                 "Avoid using the IFERROR function as it may cause performance degradation. If you are concerned about a divide-by-zero error, use the DIVIDE function as it naturally resolves such errors as blank (or you can customize what should be shown in case of such an error).",
                 "https://www.elegantbi.com/post/top10bestpractices",
             ),
+            (
+                "DAX Expressions",
+                "Measure",
+                "Error",
+                "Avoid aggregating non-numeric columns",
+                lambda obj, tom: any(
+                    find_non_numeric_aggregations(obj.Expression, tom)
+                ),
+                "Numeric aggregation functions (SUM, SUMX, AVERAGE, AVERAGEX, MIN, MINX, MAX, MAXX, PRODUCT, PRODUCTX) should be applied to numeric columns (Int64, Decimal, Double). Aggregating a non-numeric column will either fail at query time or force an implicit conversion, both of which usually indicate a modeling mistake.",
+            ),
             (
                 "DAX Expressions",
                 "Measure",
                 "Warning",
                 "Use the TREATAS function instead of INTERSECT for virtual relationships",
-                lambda obj, tom: re.search(
-                    r"intersect\s*\(", obj.Expression, flags=re.IGNORECASE
-                ),
+                lambda obj, tom: uses_function(obj.Expression, "INTERSECT"),
                 "The TREATAS function is more efficient and provides better performance than the INTERSECT function when used in virutal relationships.",
                 "https://www.sqlbi.com/articles/propagate-filters-using-treatas-in-dax",
             ),
@@ -456,11 +468,7 @@ def model_bpa_rules(
                 "Measure",
                 "Warning",
                 "The EVALUATEANDLOG function should not be used in production models",
-                lambda obj, tom: re.search(
-                    r"evaluateandlog\s*\(",
-                    obj.Expression,
-                    flags=re.IGNORECASE,
-                ),
+                lambda obj, tom: uses_function(obj.Expression, "EVALUATEANDLOG"),
                 "The EVALUATEANDLOG function is meant to be used only in development/test environments and should not be used in production models.",
                 "https://pbidax.wordpress.com/2022/08/16/introduce-the-dax-evaluateandlog-function",
             ),
@@ -576,7 +584,7 @@ def model_bpa_rules(
                 "Error",
                 "Column references should be fully qualified",
                 lambda obj, tom: any(
-                    tom.unqualified_columns(object=obj, dependencies=dependencies)
+                    find_unqualified_columns(tom._get_expression(obj), tom)
                 ),
                 "Using fully qualified column references makes it easier to distinguish between column and measure references, and also helps avoid certain errors. When referencing a column in DAX, first specify the table name, then specify the column name in square brackets.",
                 "https://www.elegantbi.com/post/top10bestpractices",
@@ -592,7 +600,7 @@ def model_bpa_rules(
                 "Error",
                 "Measure references should be unqualified",
                 lambda obj, tom: any(
-                    tom.fully_qualified_measures(object=obj, dependencies=dependencies)
+                    find_fully_qualified_measures(tom._get_expression(obj), tom)
                 ),
                 "Using unqualified measure references makes it easier to distinguish between column and measure references, and also helps avoid certain errors. When referencing a measure using DAX, do not specify the table name. Use only the measure name in square brackets.",
                 "https://www.elegantbi.com/post/top10bestpractices",

src/sempy_labs/dax/__init__.py

@@ -0,0 +1,11 @@
+from ._parser import (
+    parse_dax,
+)
+from ._format import (
+    format_dax,
+)
+
+__all__ = [
+    "parse_dax",
+    "format_dax",
+]