Testing: Add makeTimeDataFrame and makeMixedDataFrame to pueblo.testing.pandas

.github/workflows/ngr.yml

@@ -8,12 +8,14 @@ on:
     - '.github/workflows/ngr.yml'
     - 'pueblo/ngr/**'
     - 'tests/ngr/**'
+    - 'pyproject.toml'
   push:
     branches: [ main ]
     paths:
     - '.github/workflows/ngr.yml'
     - 'pueblo/ngr/**'
     - 'tests/ngr/**'
+    - 'pyproject.toml'
 
   # Allow job to be triggered manually.
   workflow_dispatch:

CHANGES.md

@@ -1,6 +1,8 @@
 # Changes for pueblo
 
 ## Unreleased
+- Testing: Add `pueblo.testing.pandas.{makeTimeDataFrame,makeMixedDataFrame}`.
+  They have been removed from `pandas._testing` on behalf of pandas 2.2.0.
 
 ## 2024-01-19 v0.0.6
 - Packaging: Fix `MANIFEST.in`

pueblo/testing/dataframe.py

@@ -5,9 +5,10 @@
 from inspect import signature
 
 import pandas as pd
-from pandas._testing import makeMixedDataFrame, makeTimeDataFrame
 from pandas.io.formats.info import DataFrameInfo
 
+from pueblo.testing.pandas import makeMixedDataFrame, makeTimeDataFrame
+
 logger = logging.getLogger(__name__)
 
 

pueblo/testing/pandas.py

@@ -0,0 +1,39 @@
+import typing as t
+
+import numpy as np
+import pandas as pd
+
+
+def makeTimeDataFrame(nper=None, freq="B") -> pd.DataFrame:
+    """
+    pandas 2.2.0 removed `pandas._testing.makeTimeDataFrame`.
+
+    TST/CLN: Remove makeTime methods
+    https://github.com/pandas-dev/pandas/pull/56264
+
+    :param nper:
+    :param freq:
+    :return:
+    """
+    return pd.DataFrame(
+        np.random.default_rng(2).standard_normal((nper, 4)),
+        columns=pd.Index(list("ABCD"), dtype=object),
+        index=pd.date_range("2000-01-01", periods=nper, freq=freq),
+    )
+
+
+def getMixedTypeDict() -> t.Tuple[pd.Index, t.Dict[str, t.Any]]:
+    index = pd.Index(["a", "b", "c", "d", "e"])
+
+    data = {
+        "A": [0.0, 1.0, 2.0, 3.0, 4.0],
+        "B": [0.0, 1.0, 0.0, 1.0, 0.0],
+        "C": ["foo1", "foo2", "foo3", "foo4", "foo5"],
+        "D": pd.bdate_range("1/1/2009", periods=5),
+    }
+
+    return index, data
+
+
+def makeMixedDataFrame() -> pd.DataFrame:
+    return pd.DataFrame(getMixedTypeDict()[1])

pyproject.toml

@@ -82,6 +82,7 @@ dataframe = [
   "dask",
   "numpy",
   "pandas",
+  "pyarrow",
 ]
 develop = [
   "black[jupyter]<24",
@@ -258,7 +259,7 @@ format = [
   # Configure Ruff not to auto-fix (remove!):
   # Ignore unused imports (F401), unused variables (F841), `print` statements (T201), and commented-out code (ERA001).
   { cmd = "ruff --fix --ignore=ERA --ignore=F401 --ignore=F841 --ignore=T20 --ignore=ERA001 ." },
-  { cmd = "pyproject-fmt pyproject.toml" },
+  { cmd = "pyproject-fmt --keep-full-version pyproject.toml" },
 ]
 
 lint = [