rebuild

Author: JohnMount

build/lib/data_algebra/PolarsSQL.py

@@ -29,7 +29,9 @@ def _polars_null_divide_expr(dbmodel, expression):
 
 
 class PolarsSQLModel(data_algebra.sql_model.SQLModel):
-    """A model of how SQL should be generated for PolarsSQL.
+    """
+    A model of how SQL should be generated for PolarsSQL.
+    Model is just a stand-in for now, as we don't have a good description of Polars SQL dialect yet.
     """
 
     def __init__(self):

build/lib/data_algebra/SQLite.py

@@ -22,6 +22,7 @@
 
 # map from op-name to special SQL formatting code
 
+
 # Standard SQL code for checking isbad doesn't work in SQLlite, so
 # at least capture to is_bad, which appears to not be implemented
 # unless we call prepare connection
@@ -83,6 +84,7 @@ def _sqlite_remainder_expr(dbmodel, expression):
         + ")"
     )
 
+
 def _sqlite_logical_or_expr(dbmodel, expression):
     """
     Return SQL or.

build/lib/data_algebra/SparkSQL.py

@@ -118,6 +118,7 @@ def _as_int(dbmodel, expression):
         + " AS INT)"
     )
 
+
 # map from op-name to special SQL formatting code
 SparkSQL_formatters = {
     "___": lambda dbmodel, expression: str(expression.to_python()),

build/lib/data_algebra/__init__.py

@@ -18,9 +18,19 @@
 
 import data_algebra.data_model
 import data_algebra.data_ops
+
 # import for easy access for package users
-from data_algebra.data_ops import TableDescription, SQLNode, describe_table, descr, data, ex
+from data_algebra.data_ops import (
+    TableDescription,
+    SQLNode,
+    describe_table,
+    descr,
+    data,
+    ex,
+)
 from data_algebra.expr_rep import lit, col, d_, one
 import data_algebra.pandas_model
+import data_algebra.data_schema
+
 
 data_algebra.pandas_model.register_pandas_model()

build/lib/data_algebra/arrow.py

@@ -41,9 +41,7 @@ class DataOpArrow(Arrow):
 
     """
 
-    def __init__(
-        self, pipeline, *, free_table_key=None
-    ):
+    def __init__(self, pipeline, *, free_table_key=None):
         assert isinstance(pipeline, data_algebra.data_ops.ViewRepresentation)
         self.pipeline = pipeline
         t_used = pipeline.get_tables()
@@ -71,7 +69,7 @@ def get_feature_names(self):
     def act_on(self, b, *, correct_ordered_first_call: bool = False):
         """
         Apply self onto b.
-        
+
         :param b: item to act on, or item that has been sent to self.
         :param correct_ordered_first_call: if True indicates this call is from __rshift__ or __rrshift__ and not the fallback paths.
         """
@@ -86,7 +84,9 @@ def act_on(self, b, *, correct_ordered_first_call: bool = False):
             excess = set(b.outgoing_columns) - set(self.incoming_columns)
             if len(excess) > 0:
                 raise ValueError("extra incoming columns: " + str(excess))
-            new_pipeline = self.pipeline.replace_leaves({self.free_table_key: b.pipeline})
+            new_pipeline = self.pipeline.replace_leaves(
+                {self.free_table_key: b.pipeline}
+            )
             new_pipeline.get_tables()  # check tables are compatible
             res = DataOpArrow(
                 pipeline=new_pipeline,
@@ -148,9 +148,7 @@ def required_columns(self):
         return self.incoming_columns.copy()
 
     # noinspection PyMethodMayBeStatic
-    def format_end_description(
-        self, *, required_cols, align_right=70, sep_width=2
-    ):
+    def format_end_description(self, *, required_cols, align_right=70, sep_width=2):
         in_rep = [str(c) for c in required_cols]
         in_rep = data_algebra.flow_text.flow_text(
             in_rep, align_right=align_right, sep_width=sep_width