Remove backports.tempfile for Python compatibility and bump version (…

…#134)

* remove backports.tempfile; add deps; bump version

* pin csvkit

* set a range for csvkit

* update test for pandas syntax

CITATION.cff

@@ -26,5 +26,5 @@ authors:
   given-names: "Anne"
   orcid: "https://orcid.org/0000-0003-1555-8261"
 title: "cytominer-database"
-version: 0.3.4
+version: 0.3.5
 url: "https://github.com/cytomining/cytominer-database"

cytominer_database/ingest.py

@@ -44,7 +44,7 @@
 import zlib
 
 import pandas as pd
-import backports.tempfile
+import tempfile
 import sqlalchemy.exc
 from sqlalchemy import create_engine
 

cytominer_database/ingest_variable_engine.py

@@ -48,7 +48,7 @@
 import click
 import zlib
 import pandas as pd
-import backports.tempfile
+import tempfile
 import sqlalchemy.exc
 from sqlalchemy import create_engine
 import pyarrow

cytominer_database/load.py

@@ -3,7 +3,7 @@
 import warnings
 import zlib
 import pandas as pd
-import backports.tempfile
+import tempfile
 import sqlalchemy.exc
 from sqlalchemy import create_engine
 import pyarrow

cytominer_database/tableSchema.py

@@ -4,7 +4,7 @@
 import warnings
 import zlib
 import pandas as pd
-import backports.tempfile
+import tempfile
 import sqlalchemy.exc
 from sqlalchemy import create_engine
 import pyarrow

cytominer_database/write.py

@@ -3,7 +3,7 @@
 import warnings
 import zlib
 import pandas as pd
-import backports.tempfile
+import tempfile
 import sqlalchemy.exc
 from sqlalchemy import create_engine
 import pyarrow

setup.py

@@ -1,7 +1,7 @@
 import setuptools
 
 project = "cytominer_database"
-version = "0.3.3"
+version = "0.3.5"
 
 # Note about authors:
 # setuptools makes it difficult to have multiple authors
@@ -30,11 +30,11 @@
     packages=setuptools.find_packages(exclude=["tests", "doc"]),
     include_package_data=True,
     install_requires=[
-        "backports.tempfile>=1.0rc1",
         "click>=6.7",
         "configparser>=3.5.0",
-        "csvkit>=1.0.2",
+        "csvkit>=1.0.2,<2",
         "pandas>=0.20.3",
+        "pyarrow",
     ],
     license="BSD",
     url="https://github.com/cytomining/cytominer-database",

tests/commands/test_command_ingest.py

@@ -2,7 +2,7 @@
 import pandas as pd
 import pytest
 import click.testing
-import backports.tempfile
+import tempfile
 from sqlalchemy import create_engine
 
 import cytominer_database.command
@@ -34,7 +34,7 @@ def test_run(dataset, runner):
 
     opts += [dataset["data_dir"]]
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         sqlite_file = os.path.join(temp_dir, "test.db")
 
         opts += ["sqlite:///{}".format(sqlite_file)]
@@ -50,7 +50,7 @@ def test_run(dataset, runner):
             engine = create_engine(target)
             con = engine.connect()
 
-            df = pd.read_sql(sql=table_name, con=con, index_col=0)
+            df = pd.read_sql(sql=f"SELECT * FROM {table_name}", con=con)
 
             assert df.shape[0] == blob["nrows"]
             assert df.shape[1] == blob["ncols"] + 1
@@ -68,7 +68,7 @@ def test_run_variable_engine_default(dataset, runner):
     # SOURCE
     opts = [dataset["data_dir"]]
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         # TARGET
         sqlite_file = os.path.join(temp_dir, "test.db")
         opts += ["sqlite:///{}".format(sqlite_file)]
@@ -96,7 +96,7 @@ def test_run_variable_engine_default(dataset, runner):
             engine = create_engine(target)
             con = engine.connect()
 
-            df = pd.read_sql(sql=table_name, con=con, index_col=0)
+            df = pd.read_sql(sql=f"SELECT * FROM {table_name}", con=con)
 
             assert df.shape[0] == blob["nrows"]
             assert df.shape[1] == blob["ncols"] + 1
@@ -112,7 +112,7 @@ def test_run_variable_engine_sqlite(dataset, runner):
     CONFIG_CHOICE = "config_SQLite.ini"
     # SOURCE
     opts = [dataset["data_dir"]]
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         # TARGET
         sqlite_file = os.path.join(temp_dir, "test.db")
         opts += ["sqlite:///{}".format(sqlite_file)]
@@ -139,7 +139,7 @@ def test_run_variable_engine_sqlite(dataset, runner):
             engine = create_engine(target)
             con = engine.connect()
 
-            df = pd.read_sql(sql=table_name, con=con, index_col=0)
+            df = pd.read_sql(sql=f"SELECT * FROM {table_name}", con=con)
 
             assert df.shape[0] == blob["nrows"]
             assert df.shape[1] == blob["ncols"] + 1
@@ -156,7 +156,7 @@ def test_run_variable_engine_parquet(dataset, runner):
     # SOURCE
     opts = [dataset["data_dir"]]
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         # TARGET
         # create output directory
         target = os.path.join(temp_dir, "test_parquet_output")
@@ -211,7 +211,7 @@ def test_run_defaults(cellpainting, runner):
 
     opts += [cellpainting["data_dir"]]
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         sqlite_file = os.path.join(temp_dir, "test.db")
 
         opts += ["sqlite:///{}".format(sqlite_file)]
@@ -227,7 +227,7 @@ def test_run_defaults(cellpainting, runner):
             engine = create_engine(target)
             con = engine.connect()
 
-            df = pd.read_sql(sql=table_name, con=con, index_col=0)
+            df = pd.read_sql(sql=f"SELECT * FROM {table_name}", con=con)
 
             assert df.shape[0] == blob["nrows"]
             assert df.shape[1] == blob["ncols"] + 1

tests/test_ingest.py

@@ -1,7 +1,7 @@
 import os.path
 
 import pandas as pd
-import backports.tempfile
+import tempfile
 from sqlalchemy import create_engine
 
 import cytominer_database.ingest
@@ -27,7 +27,7 @@ def test_seed(dataset):
     if munge:
         cytominer_database.munge.munge(config_file, data_dir)
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         sqlite_file = os.path.join(temp_dir, "test.db")
 
         cytominer_database.ingest.seed(

tests/test_ingest_variable_engine.py

@@ -1,7 +1,7 @@
 import os
 
 import pandas as pd
-import backports.tempfile
+import tempfile
 from sqlalchemy import create_engine
 
 # import cytominer_database.ingest
@@ -24,7 +24,7 @@ def test_seed(dataset, config_choice):
     if munge:
         cytominer_database.munge.munge(config_path, data_dir)
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         # set up
         if engine_type == "Parquet":
             # create output directory

tests/test_munge.py

@@ -1,9 +1,9 @@
 import os.path
 
-import backports.tempfile
+import tempfile
 import cytominer_database.munge
 import pandas as pd
-import pandas.util.testing
+from pandas.testing import assert_frame_equal
 
 
 def test_munge(dataset):
@@ -12,7 +12,7 @@ def test_munge(dataset):
 
     config_file = os.path.join(dataset["data_dir"], "config.ini")
 
-    with backports.tempfile.TemporaryDirectory() as temp_dir:
+    with tempfile.TemporaryDirectory() as temp_dir:
         valid_directories = cytominer_database.munge.munge(
             config_path=config_file, source=dataset["data_dir"], target=temp_dir
         )
@@ -32,4 +32,4 @@ def test_munge(dataset):
                     )
                 )
 
-                pandas.util.testing.assert_frame_equal(input_csv, output_csv)
+                assert_frame_equal(input_csv, output_csv)