Merge remote-tracking branch 'upstream/main' into ref/index_equiv

Author: mroeschke

Dockerfile

@@ -11,4 +11,5 @@ RUN apt-get install -y libhdf5-dev libgles2-mesa-dev
 RUN python -m pip install --upgrade pip
 COPY requirements-dev.txt /tmp
 RUN python -m pip install -r /tmp/requirements-dev.txt
+RUN git config --global --add safe.directory /home/pandas
 CMD ["/bin/bash"]

asv_bench/benchmarks/categoricals.py

@@ -88,7 +88,7 @@ def setup(self):
         )
 
         for col in ("int", "float", "timestamp"):
-            self.df[col + "_as_str"] = self.df[col].astype(str)
+            self.df[f"{col}_as_str"] = self.df[col].astype(str)
 
         for col in self.df.columns:
             self.df[col] = self.df[col].astype("category")

asv_bench/benchmarks/join_merge.py

@@ -328,6 +328,23 @@ def time_i8merge(self, how):
         merge(self.left, self.right, how=how)
 
 
+class UniqueMerge:
+    params = [4_000_000, 1_000_000]
+    param_names = ["unique_elements"]
+
+    def setup(self, unique_elements):
+        N = 1_000_000
+        self.left = DataFrame({"a": np.random.randint(1, unique_elements, (N,))})
+        self.right = DataFrame({"a": np.random.randint(1, unique_elements, (N,))})
+        uniques = self.right.a.drop_duplicates()
+        self.right["a"] = concat(
+            [uniques, Series(np.arange(0, -(N - len(uniques)), -1))], ignore_index=True
+        )
+
+    def time_unique_merge(self, unique_elements):
+        merge(self.left, self.right, how="inner")
+
+
 class MergeDatetime:
     params = [
         [

asv_bench/benchmarks/timeseries.py

@@ -183,7 +183,7 @@ def setup(self):
         self.dt_ts = Series(5, rng3, dtype="datetime64[ns]")
 
     def time_resample(self):
-        self.dt_ts.resample("1S").last()
+        self.dt_ts.resample("1s").last()
 
 
 class AsOf:

doc/source/_static/css/getting_started.css

@@ -248,6 +248,7 @@ ul.task-bullet > li > p:first-child {
 }
 
 .tutorial-card .card-header {
+  --bs-card-cap-color: var(--pst-color-text-base);
   cursor: pointer;
   background-color: var(--pst-color-surface);
   border: 1px solid var(--pst-color-border)
@@ -269,7 +270,7 @@ ul.task-bullet > li > p:first-child {
 
 
 .tutorial-card .gs-badge-link a {
-  color: var(--pst-color-text-base);
+  color: var(--pst-color-primary-text);
   text-decoration: none;
 }
 

doc/source/development/contributing_docstring.rst

@@ -940,7 +940,7 @@ Finally, docstrings can also be appended to with the ``doc`` decorator.
 
 In this example, we'll create a parent docstring normally (this is like
 ``pandas.core.generic.NDFrame``). Then we'll have two children (like
-``pandas.core.series.Series`` and ``pandas.core.frame.DataFrame``). We'll
+``pandas.core.series.Series`` and ``pandas.DataFrame``). We'll
 substitute the class names in this docstring.
 
 .. code-block:: python

doc/source/development/maintaining.rst

@@ -151,15 +151,15 @@ and then run::
     git bisect start
     git bisect good v1.4.0
     git bisect bad v1.5.0
-    git bisect run bash -c "python setup.py build_ext -j 4; python t.py"
+    git bisect run bash -c "python -m pip install -ve . --no-build-isolation --config-settings editable-verbose=true; python t.py"
 
 This finds the first commit that changed the behavior. The C extensions have to be
 rebuilt at every step, so the search can take a while.
 
 Exit bisect and rebuild the current version::
 
     git bisect reset
-    python setup.py build_ext -j 4
+    python -m pip install -ve . --no-build-isolation --config-settings editable-verbose=true
 
 Report your findings under the corresponding issue and ping the commit author to get
 their input.

doc/source/user_guide/enhancingperf.rst

@@ -453,7 +453,7 @@ by evaluate arithmetic and boolean expression all at once for large :class:`~pan
    :func:`~pandas.eval` is many orders of magnitude slower for
    smaller expressions or objects than plain Python. A good rule of thumb is
    to only use :func:`~pandas.eval` when you have a
-   :class:`.DataFrame` with more than 10,000 rows.
+   :class:`~pandas.core.frame.DataFrame` with more than 10,000 rows.
 
 Supported syntax
 ~~~~~~~~~~~~~~~~

doc/source/user_guide/io.rst

@@ -6400,7 +6400,7 @@ ignored.
    In [2]: df = pd.DataFrame({'A': np.random.randn(sz), 'B': [1] * sz})
 
    In [3]: df.info()
-   <class 'pandas.core.frame.DataFrame'>
+   <class 'pandas.DataFrame'>
    RangeIndex: 1000000 entries, 0 to 999999
    Data columns (total 2 columns):
    A    1000000 non-null float64

doc/source/user_guide/missing_data.rst

@@ -88,7 +88,7 @@ To detect these missing value, use the :func:`isna` or :func:`notna` methods.
 
 .. warning::
 
-   Experimental: the behaviour of :class:`NA`` can still change without warning.
+   Experimental: the behaviour of :class:`NA` can still change without warning.
 
 Starting from pandas 1.0, an experimental :class:`NA` value (singleton) is
 available to represent scalar missing values. The goal of :class:`NA` is provide a
@@ -105,7 +105,7 @@ dtype, it will use :class:`NA`:
     s[2]
     s[2] is pd.NA
 
-Currently, pandas does not yet use those data types using :class:`NA` by default
+Currently, pandas does not use those data types using :class:`NA` by default in
 a :class:`DataFrame` or :class:`Series`, so you need to specify
 the dtype explicitly. An easy way to convert to those dtypes is explained in the
 :ref:`conversion section <missing_data.NA.conversion>`.
@@ -253,8 +253,8 @@ Conversion
 ^^^^^^^^^^
 
 If you have a :class:`DataFrame` or :class:`Series` using ``np.nan``,
-:meth:`Series.convert_dtypes` and :meth:`DataFrame.convert_dtypes`
-in :class:`DataFrame` that can convert data to use the data types that use :class:`NA`
+:meth:`DataFrame.convert_dtypes` and :meth:`Series.convert_dtypes`, respectively,
+will convert your data to use the nullable data types supporting :class:`NA`,
 such as :class:`Int64Dtype` or :class:`ArrowDtype`. This is especially helpful after reading
 in data sets from IO methods where data types were inferred.