Merge pull request #14 from Hugoberry/dax-columns

Surface DAX columns

Author: Hugoberry

README.md

@@ -59,6 +59,12 @@ To access DAX measures in a dataframe with `TableName`, `Name`, `Expression`, `D
 dax_measures = model.dax_measures
 print(dax_measures)
 ```
+### Calculated Columns
+To access calculated column DAX expressions in a dataframe with `TableName`,`ColumnName` and `Expression` columns:
+```python
+dax_columns = model.dax_columns
+print(dax_columns)
+```
 ### Schema
 To get details about the data model schema and column types in a dataframe with `TableName`, `ColumnName`, and `PandasDataType` columns:
 ```python

data/2020SU11 Blog Demo - November.pbix

@@ -36,4 +36,18 @@
 ic(model.size)
 ic(model.schema)
 ic(model.relationships)
-ic(model.get_table("Fruit_RLE"))
+ic(model.dax_columns)
+ic(model.get_table("Fruit_RLE"))
+
+model = PBIXRay(r"C:\git\hub\pbixray\data\2020SU11 Blog Demo - November.pbix")
+ic(model.tables)
+ic(model.metadata)
+ic(model.power_query)
+ic(model.statistics)
+ic(model.dax_tables)
+ic(model.dax_measures)
+ic(model.dax_columns)
+ic(model.size)
+ic(model.schema)
+ic(model.relationships)
+ic(model.get_table("Reseller"))

data/output/metadata.sqlitedb

@@ -132,7 +132,8 @@ def __init__(self, _io, _parent=None, _root=None):
 
         def _read(self):
             self.cp = IdfmetaParser.CPElement(self._io, self, self._root)
-            self.sdos = IdfmetaParser.SDOsElement(self._io, self, self._root)
+            if self._io.size() > self._io.pos():
+                self.sdos = IdfmetaParser.SDOsElement(self._io, self, self._root)
 
 
     class CSDOs1Element(KaitaiStruct):

demo.py

@@ -39,6 +39,10 @@ def dax_tables(self):
     def dax_measures(self):
         return self._metadata_handler.metadata.dax_measures_df
 
+    @property
+    def dax_columns(self):
+        return self._metadata_handler.metadata.dax_columns_df
+    
     @property
     def metadata(self):
         return self._metadata_handler.metadata.metadata_df

pbixray/column_data/idfmeta.py

@@ -8,8 +8,9 @@ def __init__(self, sqlite_handler):
         self.m_df = self.__populate_m()
         self.dax_tables_df = self.__populate_dax_tables()
         self.dax_measures_df = self.__populate_dax_measures()
-        self.metadata_df = self.populate_metadata()
-        self.relationships_df = self.populate_relationships()
+        self.dax_columns_df = self.__populate_dax_columns()
+        self.metadata_df = self.__populate_metadata()
+        self.relationships_df = self.__populate_relationships()
         self.handler.close_connection()
 
     def __populate_schema(self):
@@ -41,7 +42,7 @@ def __populate_schema(self):
         --IDF
         JOIN ColumnPartitionStorage cps ON cps.ColumnStorageID = cs.ID
         JOIN StorageFile sfi ON sfi.ID = cps.StorageFileID
-        WHERE c.Type = 1
+        WHERE c.Type IN (1,2)
         ORDER BY t.Name, cs.StoragePosition
         """
         return self.handler.execute_query(sql)
@@ -80,16 +81,28 @@ def __populate_dax_measures(self):
         JOIN [Table] t ON m.TableID = t.ID;
         """
         return self.handler.execute_query(sql)
+    
+    def __populate_dax_columns(self):
+        sql = """ 
+        SELECT 
+            t.Name AS TableName,
+            c.ExplicitName AS ColumnName,
+            c.Expression
+        FROM Column c 
+        JOIN [Table] t ON c.TableID = t.ID
+        WHERE c.Type = 2;
+        """
+        return self.handler.execute_query(sql)
 
-    def populate_metadata(self):
+    def __populate_metadata(self):
         sql = """
         SELECT Name,Value 
         FROM Annotation 
         WHERE ObjectType = 1
         """
         return self.handler.execute_query(sql)
 
-    def populate_relationships(self):
+    def __populate_relationships(self):
         sql = """
         SELECT 
             ft.Name AS FromTableName,

pbixray/core.py

@@ -2,7 +2,7 @@
 
 setup(
     name='pbixray',
-    version='0.1.19',
+    version='0.1.20',
     packages=find_packages(),
     install_requires=[
         'kaitaistruct',