Merge pull request #27 from pysat/remove_panel

Remove panel

CHANGELOG.md

@@ -2,6 +2,11 @@
 All notable changes to this project will be documented in this file.
 This project adheres to [Semantic Versioning](http://semver.org/).
 
+## [0.2.0] - 2022-XX-XX
+- Removed deprecated `pandas.Panel` from functions.
+- Renamed `computational_form` to `to_xarray_dataset` and refocused.
+- Removed old __future__ imports.
+
 ## [0.1.3] - 2021-06-18
 - Updates style to match pysat 3.0.0 release candidate
 - Improves discussion of rationale for version caps on readme page

pysatSeasons/__init__.py

@@ -19,7 +19,7 @@
 from pysatSeasons import occur_prob  # noqa: F401
 from pysatSeasons import avg  # noqa: F401
 from pysatSeasons import plot  # noqa: F401
-from pysatSeasons._core import computational_form  # noqa: F401
+from pysatSeasons._core import to_xarray_dataset  # noqa: F401
 
 # set version
 here = os.path.abspath(os.path.dirname(__file__))

pysatSeasons/_core.py

@@ -1,38 +1,84 @@
 import pandas as pds
+import xarray as xr
 
 
-def computational_form(data):
-    """
-    Repackages numbers, Series, or DataFrames
+def to_xarray_dataset(data):
+    """Repackage input data into `xarray.Dataset`.
 
     Regardless of input format, mathematical operations may be performed on the
-    output via the same pandas mechanisms.
-
-    This method may be particularly useful in analysis methods that aim to be
-    instrument independent. pysat.Instrument objects can package data in a
-    variety of ways within a DataFrame, depending upon the scientific data
-    source. Thus, a variety of data types will be encountered by instrument
-    independent methods and computational_form method may reduce the effort
-    required to support more generalized processing.
+    output via the same xarray mechanisms. This method may be particularly
+    useful in analysis methods that aim to be instrument independent.
+    pysat.Instrument objects can package data in a variety of ways within a
+    DataFrame, depending upon the scientific data source. Thus, a variety of
+    data types will be encountered by instrument independent methods and
+    `to_xarray_dataset` method may reduce the effort required to support more
+    generalized processing.
 
     Parameters
     ----------
-    data : pandas.Series
-        Series of numbers, Series, DataFrames
+    data : pds.Series, pds.DataFrame, xr.DataArray, xr.DataSet, or list-like
+           of the same or numbers.
+        List-like of numbers, Series, DataFrames, or Datasets to be combined
+        into a single Dataset.
 
     Returns
     -------
-    pandas.Series, DataFrame, or Panel
-        repacked data, aligned by indices, ready for calculation
+    output : xr.Dataset
+        Repacked data, aligned by indices. If data is a list of multidimensional
+        objects then output will have a corresponding new dimension
+        'pysat_binning' to reflect that organization. If data is a list of
+        numbers, then output will have a single associated variable, `data`.
+        Otherwise, variable names are retained from the input data.
 
     """
 
-    if isinstance(data.iloc[0], pds.DataFrame):
-        dslice = pds.Panel.from_dict(dict([(i, data.iloc[i])
-                                     for i in range(len(data))]))
-    elif isinstance(data.iloc[0], pds.Series):
-        dslice = pds.DataFrame(data.tolist())
-        dslice.index = data.index
+    if isinstance(data, pds.DataFrame):
+        output = data.to_xarray()
+    elif isinstance(data, pds.Series):
+        output = xr.Dataset()
+        output[data.name] = data.to_xarray()
+    elif isinstance(data, xr.Dataset):
+        output = data
+    elif isinstance(data, xr.DataArray):
+        output = xr.Dataset()
+        output[data.name] = data
+    elif isinstance(data[0], xr.Dataset):
+        # Combine list-like of Datasets
+        vars = data[0].data_vars.keys()
+        output = xr.Dataset()
+
+        # Combine all info for each variable into a single data
+        # array.
+        for var in vars:
+            output[var] = xr.concat([item[var] for item in data],
+                                    'pysat_binning')
+    elif isinstance(data[0], xr.DataArray):
+        # Combine list-like of DataArrays
+        vars = [data[0].name]
+        output = xr.Dataset()
+
+        for var in vars:
+            output[var] = xr.concat(data, 'pysat_binning')
+    elif isinstance(data[0], pds.DataFrame):
+        # Convert list-like of DataFrames
+        info = [xr.Dataset.from_dataframe(item) for item in data]
+
+        vars = info[0].data_vars.keys()
+        output = xr.Dataset()
+
+        for var in vars:
+            output[var] = xr.concat([item[var] for item in info],
+                                    'pysat_binning')
+    elif isinstance(data[0], pds.Series):
+        # Convert list-like of pds.Series
+        vars = [data[0].name]
+        output = xr.Dataset()
+
+        for var in vars:
+            output[var] = xr.concat([xr.DataArray.from_series(item)
+                                     for item in data], 'pysat_binning')
     else:
-        dslice = data
-    return dslice
+        output = xr.Dataset()
+        output['data'] = data
+
+    return output