subdtype (#247)

.gitignore

@@ -11,6 +11,7 @@ build/
 dist/
 MANIFEST
 *pytest_cache*
+*mypy_cache*
 .eggs
 
 # WebDAV file system cache files

CHANGES

@@ -6,6 +6,7 @@ pint-pandas Changelog
 --------------------
 
 - Added `__array_function__` support for numpy fuctions like clip.
+- `PintArray` now explictly shows its `subtype`, the data type of its magnitude array, eg `pint[m][Float64]`
 
 
 0.6.2 (2024-07-29)

docs/getting/tutorial.rst

@@ -1,5 +1,3 @@
-.. _tutorial:
-
 **************************
 Tutorial
 **************************

docs/index.rst

@@ -1,7 +1,7 @@
 :orphan:
 
 Pint-pandas: Unit support for pandas
-======================
+=====================================
 
 **Useful links**:
 `Code Repository <https://github.com/hgrecco/pint-pandas>`__ |
@@ -66,7 +66,6 @@ Pint-pandas: Unit support for pandas
 
     Getting started <getting/index>
     User Guide <user/index>
-    Advanced topics <advanced/index>
     ecosystem
 
 .. toctree::

docs/user/common.rst

@@ -61,6 +61,7 @@ The default operation of Pandas `pd.concat` function is to perform row-wise conc
     :suppress:
     :okexcept:
 
+        list_of_series = [pd.Series([1.0, 2.0], dtype="pint[m]") for i in range(0, 10)]
         df = pd.concat(list_of_series, axis=1)
 
 

docs/user/initializing.rst

@@ -4,35 +4,50 @@
 Initializing data
 **************************
 
-There are several ways to initialize PintArrays in a DataFrame. Here's the most common methods. We'll use `PA_` and `Q_` as shorthand for PintArray and Quantity.
+There are several ways to initialize a `PintArray`s` in a `DataFrame`. Here's the most common methods. We'll use `PA_` and `Q_` as shorthand for `PintArray` and `Quantity`.
 
 
 
 .. ipython:: python
+    :okwarning:
 
     import pandas as pd
     import pint
     import pint_pandas
-    import io
 
     PA_ = pint_pandas.PintArray
     ureg = pint_pandas.PintType.ureg
     Q_ = ureg.Quantity
 
     df = pd.DataFrame(
         {
-            "A": pd.Series([1.0, 2.0], dtype="pint[m]"),
-            "B": pd.Series([1.0, 2.0]).astype("pint[m]"),
-            "C": PA_([2.0, 3.0], dtype="pint[m]"),
-            "D": PA_([2.0, 3.0], dtype="m"),
-            "E": PA_([2.0, 3.0], dtype=ureg.m),
-            "F": PA_.from_1darray_quantity(Q_([2, 3], ureg.m)),
-            "G": PA_(Q_([2.0, 3.0], ureg.m)),
+            "Ser1": pd.Series([1, 2], dtype="pint[m]"),
+            "Ser2": pd.Series([1, 2]).astype("pint[m]"),
+            "Ser3": pd.Series([1, 2], dtype="pint[m][Int64]"),
+            "Ser4": pd.Series([1, 2]).astype("pint[m][Int64]"),
+            "PArr1": PA_([1, 2], dtype="pint[m]"),
+            "PArr2": PA_([1, 2], dtype="pint[m][Int64]"),
+            "PArr3": PA_([1, 2], dtype="m"),
+            "PArr4": PA_([1, 2], dtype=ureg.m),
+            "PArr5": PA_(Q_([1, 2], ureg.m)),
+            "PArr6": PA_([1, 2],"m"),
         }
     )
     df
 
 
+In the first two Series examples above, the data was converted to Float64.
+
+.. ipython:: python
+
+    df.dtypes
+
+
+To avoid this conversion, specify the subdtype (dtype of the magnitudes) in the dtype `"pint[m][Int64]"` when constructing using a `Series`. The default data dtype that pint-pandas converts to can be changed by modifying `pint_pandas.DEFAULT_SUBDTYPE`.
+
+`PintArray` infers the subdtype from the data passed into it when there is no subdtype specified in the dtype. It also accepts a pint `Unit`` or unit string as the dtype.
+
+
 .. note::
 
-   "pint[unit]" must be used for the Series or DataFrame constuctor.
+   `"pint[unit]"` or `"pint[unit][subdtype]"` must be used for the Series or DataFrame constuctor.