retest and rebuild

Author: JohnMount

build/lib/data_algebra/BigQuery.py

@@ -249,20 +249,10 @@ def query_to_csv(self, q, *, res_name) -> None:
             q = q.to_sql(self.db_model)
         else:
             q = str(q)
-
-        def open_regular():
-            """open regular"""
-            return lambda: open(res_name, "w")
-
-        def open_gzip():
-            """open gzipped"""
-            return lambda: gzip.open(res_name, "w")
-
         if res_name.endswith(".gz"):
-            op = open_gzip
+            op = lambda: gzip.open(res_name, "w", encoding="utf-8")
         else:
-            op = open_regular()
-
+            op = lambda: open(res_name, "w", encoding="utf-8")
         with op() as res:
             res_iter = self.conn.query(q).result().to_dataframe_iterable()
             is_first = True

build/lib/data_algebra/__init__.py

@@ -16,8 +16,5 @@
 the `rquery`<https://github.com/WinVector/rquery> and `rqdatatable`<https://github.com/WinVector/rqdatatable> packages.
 """
 
-import data_algebra.pandas_model
-
-
-# set up what pandas supplier we are using
-default_data_model = data_algebra.pandas_model.PandasModel()
+# import for easy access for package users
+from data_algebra.data_ops import TableDescription, SQLNode, describe_table, descr, data, ex

build/lib/data_algebra/arrow.py

@@ -1,6 +1,5 @@
 import abc
 
-import data_algebra
 import data_algebra.data_ops
 import data_algebra.flow_text
 

build/lib/data_algebra/cdata.py

@@ -7,8 +7,7 @@
 from typing import Iterable, List, Optional
 
 import numpy
-
-import data_algebra
+import data_algebra.pandas_model
 import data_algebra.util
 
 
@@ -36,7 +35,7 @@ def __init__(
         :param local_data_model: data.frame data model
         """
         if local_data_model is None:
-            local_data_model = data_algebra.default_data_model
+            local_data_model = data_algebra.pandas_model.default_data_model
         control_table = control_table.reset_index(inplace=False, drop=True)
         if control_table.shape[0] < 1:
             raise ValueError("control table should have at least 1 row")
@@ -194,7 +193,7 @@ def blocks_to_rowrecs(data, *, blocks_in: RecordSpecification, local_data_model=
     if len(ck) != len(set(ck)):
         raise ValueError("blocks_in can not have duplicate content keys")
     if local_data_model is None:
-        local_data_model = data_algebra.default_data_model
+        local_data_model = data_algebra.pandas_model.default_data_model
     data = data.reset_index(drop=True)
     missing_cols = set(blocks_in.control_table_keys).union(blocks_in.record_keys) - set(
         data.columns
@@ -281,7 +280,7 @@ def rowrecs_to_blocks(
     """
     assert isinstance(blocks_out, RecordSpecification)
     if local_data_model is None:
-        local_data_model = data_algebra.default_data_model
+        local_data_model = data_algebra.pandas_model.default_data_model
     data = data.reset_index(drop=True)
     missing_cols = set(blocks_out.record_keys) - set(data.columns)
     if len(missing_cols) > 0:
@@ -431,7 +430,7 @@ def example_input(self, *, local_data_model=None):
         :return: example result data frame.
         """
         if local_data_model is None:
-            local_data_model = data_algebra.default_data_model
+            local_data_model = data_algebra.pandas_model.default_data_model
         if self.blocks_in is not None:
             example = self.blocks_in.control_table.copy()
             nrow = example.shape[0]
@@ -461,7 +460,7 @@ def transform(
         if len(unknown) > 0:
             raise ValueError("missing required columns: " + str(unknown))
         if local_data_model is None:
-            local_data_model = data_algebra.default_data_model
+            local_data_model = data_algebra.pandas_model.default_data_model
         X = X.reset_index(drop=True)
         if self.blocks_in is not None:
             X = blocks_to_rowrecs(
@@ -651,7 +650,7 @@ def pivot_blocks_to_rowrecs(
     """
 
     if local_data_model is None:
-        local_data_model = data_algebra.default_data_model
+        local_data_model = data_algebra.pandas_model.default_data_model
     control_table = local_data_model.data_frame(
         {
             attribute_key_column: record_value_columns,
@@ -687,7 +686,7 @@ def pivot_rowrecs_to_blocks(
     """
 
     if local_data_model is None:
-        local_data_model = data_algebra.default_data_model
+        local_data_model = data_algebra.pandas_model.default_data_model
     control_table = local_data_model.data_frame(
         {
             attribute_key_column: record_value_columns,
@@ -730,7 +729,7 @@ def pivot_specification(
     assert len(known_cols) == len(set(known_cols))
     record_map = RecordMap(
         blocks_in=RecordSpecification(
-            control_table=data_algebra.pandas_model.pd.DataFrame(
+            control_table=data_algebra.pandas_model.default_data_model.pd.DataFrame(
                 {
                     col_name_key: value_cols,
                     col_value_key: value_cols,
@@ -770,7 +769,7 @@ def unpivot_specification(
     assert len(known_cols) == len(set(known_cols))
     record_map = RecordMap(
         blocks_out=RecordSpecification(
-            control_table=data_algebra.pandas_model.pd.DataFrame(
+            control_table=data_algebra.pandas_model.default_data_model.pd.DataFrame(
                 {
                     col_name_key: value_cols,
                     col_value_key: value_cols,

build/lib/data_algebra/data_ops.py

@@ -11,11 +11,10 @@
 
 import numpy
 
-import data_algebra
+import data_algebra.pandas_model
 import data_algebra.expr_parse
 import data_algebra.flow_text
 import data_algebra.data_model
-import data_algebra.pandas_model
 import data_algebra.expr_rep
 from data_algebra.data_ops_types import MethodUse, OperatorPlatform
 import data_algebra.data_ops_utils
@@ -480,7 +479,7 @@ def eval(
         """
         assert isinstance(data_map, dict)
         if data_model is None:
-            data_model = data_algebra.default_data_model
+            data_model = data_algebra.pandas_model.default_data_model
         assert isinstance(data_model, data_algebra.data_model.DataModel)
         self.columns_used()  # for table consistency check/raise
         tables = self.get_tables()
@@ -512,7 +511,7 @@ def transform(
         :return: transformed data frame
         """
         if data_model is None:
-            data_model = data_algebra.default_data_model
+            data_model = data_algebra.pandas_model.default_data_model
         assert isinstance(data_model, data_algebra.data_model.DataModel)
         self.columns_used()  # for table consistency check/raise
         tables = self.get_tables()

build/lib/data_algebra/db_model.py

@@ -11,8 +11,7 @@
 
 import pandas.io.sql
 
-import data_algebra
-
+import data_algebra.pandas_model
 import data_algebra.near_sql
 import data_algebra.expr_rep
 import data_algebra.util
@@ -797,7 +796,7 @@ def __init__(
         union_all_term_end: str = ")",
     ):
         if local_data_model is None:
-            local_data_model = data_algebra.default_data_model
+            local_data_model = data_algebra.pandas_model.default_data_model
         self.local_data_model = local_data_model
         if sql_formatters is None:
             sql_formatters = {}

build/lib/data_algebra/eval_cache.py

@@ -2,7 +2,7 @@
 
 from typing import Any, Dict, List, NamedTuple, Optional, Tuple
 import hashlib
-import data_algebra
+import data_algebra.pandas_model
 import data_algebra.db_model
 
 
@@ -13,9 +13,9 @@ def hash_data_frame(d) -> str:
     :param d: data frame
     :return: hash code as a string
     """
-    data_algebra.default_data_model.is_appropriate_data_instance(d)
+    data_algebra.pandas_model.default_data_model.is_appropriate_data_instance(d)
     hash_str = hashlib.sha256(
-        data_algebra.default_data_model.pd.util.hash_pandas_object(d).values
+        data_algebra.pandas_model.default_data_model.pd.util.hash_pandas_object(d).values
     ).hexdigest()
     return f"{d.shape}_{list(d.columns)}_{hash_str}"
 
@@ -44,7 +44,7 @@ def make_cache_key(
     data_map_keys.sort()
     for k in data_map_keys:
         assert isinstance(k, str)
-        assert data_algebra.default_data_model.is_appropriate_data_instance(data_map[k])
+        assert data_algebra.pandas_model.default_data_model.is_appropriate_data_instance(data_map[k])
     return EvalKey(
         db_model_name=str(db_model),
         sql=sql,
@@ -74,7 +74,7 @@ def get(
         """get result from cache, raise KeyError if not present"""
         k = make_cache_key(db_model=db_model, sql=sql, data_map=data_map)
         res = self.result_cache[k]
-        assert data_algebra.default_data_model.is_appropriate_data_instance(res)
+        assert data_algebra.pandas_model.default_data_model.is_appropriate_data_instance(res)
         return res.copy()
 
     def store(
@@ -86,7 +86,7 @@ def store(
         res,
     ) -> None:
         """Store result to cache, mark dirty if change."""
-        assert data_algebra.default_data_model.is_appropriate_data_instance(res)
+        assert data_algebra.pandas_model.default_data_model.is_appropriate_data_instance(res)
         op_key = make_cache_key(db_model=db_model, sql=sql, data_map=data_map)
         try:
             previous = self.result_cache[op_key]

build/lib/data_algebra/expr_rep.py

@@ -7,8 +7,8 @@
 
 import numpy
 
-import data_algebra
 import data_algebra.util
+import data_algebra.pandas_model
 
 
 # for some ideas in capturing expressions in Python see:
@@ -1333,14 +1333,14 @@ def _can_find_method_by_name(op):
     # first check chosen mappings
     try:
         # noinspection PyUnusedLocal
-        check_val = data_algebra.default_data_model.user_fun_map[op]  # for KeyError
+        check_val = data_algebra.pandas_model.default_data_model.user_fun_map[op]  # for KeyError
         return True
     except KeyError:
         pass
     # check chosen mappings
     try:
         # noinspection PyUnusedLocal
-        check_val = data_algebra.default_data_model.impl_map[op]  # for KeyError
+        check_val = data_algebra.pandas_model.default_data_model.impl_map[op]  # for KeyError
         return True
     except KeyError:
         pass
@@ -1454,13 +1454,13 @@ def evaluate(self, data_frame):
         # check user fns
         # first check chosen mappings
         try:
-            method_to_call = data_algebra.default_data_model.user_fun_map[self.op]
+            method_to_call = data_algebra.pandas_model.default_data_model.user_fun_map[self.op]
             return method_to_call(*args)
         except KeyError:
             pass
         # check chosen mappings
         try:
-            method_to_call = data_algebra.default_data_model.impl_map[self.op]
+            method_to_call = data_algebra.pandas_model.default_data_model.impl_map[self.op]
             return method_to_call(*args)
         except KeyError:
             pass

build/lib/data_algebra/op_catalog.py

@@ -1,6 +1,7 @@
-import data_algebra
 
-pd = data_algebra.default_data_model.pd
+import data_algebra.pandas_model
+
+pd = data_algebra.pandas_model.default_data_model.pd
 
 
 methods_table = pd.DataFrame(