added reindex to History, and added tests for that; fixed old bug in test History

saqc/flagger/history.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 from typing import Tuple, Type
 import pandas as pd
 import numpy as np
+from saqc.common import *
 
 
 class History:
@@ -149,7 +150,9 @@ class History:
         -------
         History
         """
-        # internal detail:
+        # Note:
+        # all following code must handle a passed empty series
+
         # ensure continuous increasing columns
         assert 0 <= pos <= len(self)
 
@@ -191,11 +194,8 @@ class History:
         """
         s = self._validate_value(value)
 
-        if s.empty:
-            raise ValueError('Cannot append empty pd.Series')
-
-        if not self.empty and not s.index.equals(self.index):
-            raise ValueError("Index must be equal to FH's index")
+        if len(self) > 0 and not s.index.equals(self.index):
+            raise ValueError("Index must be equal to FlagHistory index")
 
         self._insert(value, pos=len(self), force=force)
         return self
@@ -279,6 +279,28 @@ class History:
         """
         return self._constructor(hist=self, copy=deep)
 
+    def reindex(self, index: pd.Index, fill_value_last: float = UNFLAGGED) -> History:
+        """
+        Reindex the History. Be careful this alters the past.
+
+        Parameters
+        ----------
+        index : pd.Index
+            the index to reindex to.
+        fill_value_last : float, default 0
+            value to fill nan's (UNTOUCHED) in the last column. Defaults to 0 (UNFLAGGED).
+
+        Returns
+        -------
+        History
+        """
+        self.hist = self.hist.reindex(index=index, copy=False, fill_value=np.nan)
+        self.mask = self.mask.reindex(index=index, copy=False, fill_value=False)
+        # Note: all following code must handle empty frames
+        self.hist.iloc[:, -1:] = self.hist.iloc[:, -1:].fillna(fill_value_last)
+        self.mask.iloc[:, -1:] = True
+        return self
+
     def __copy__(self, deep: bool = True):
         return self.copy(deep=deep)
 

test/flagger/test_history.py

@@ -102,7 +102,11 @@ def check_invariants(hist):
     # or the entire row is True
     if not hist.empty:
         idxmax = hist.mask.idxmax(axis=1)
+        print(f'idxmax: {idxmax}')
         for row, col in idxmax.items():
+            # this is contra intuitive, it gets the positional (for iloc)
+            row = idxmax.index.get_loc(row)
+
             assert all(hist.mask.iloc[row, :col] == False)
             assert all(hist.mask.iloc[row, col:] == True)
 
@@ -183,6 +187,41 @@ def test_copy(data):
     assert shallow.mask is hist.mask
 
 
+@pytest.mark.parametrize('data', data + [None])
+def test_reindex_trivial_cases(data):
+    df = pd.DataFrame(data, dtype=float)
+    orig = History(hist=df)
+
+    # checks
+    for index in [df.index, pd.Index([])]:
+        hist = orig.copy()
+        ref = hist.reindex(index)
+        assert ref is hist  # check if working inplace
+        check_invariants(hist)
+
+
+@pytest.mark.parametrize('data', data + [None])
+def test_reindex_missing_indicees(data):
+    df = pd.DataFrame(data, dtype=float)
+    hist = History(hist=df)
+    index = df.index[1:-1]
+    # checks
+    ref = hist.reindex(index)
+    assert ref is hist  # check if working inplace
+    check_invariants(hist)
+
+
+@pytest.mark.parametrize('data', data + [None])
+def test_reindex_extra_indicees(data):
+    df = pd.DataFrame(data, dtype=float)
+    hist = History(hist=df)
+    index = df.index.append(pd.Index(range(len(df.index), len(df.index) + 5)))
+    # checks
+    ref = hist.reindex(index)
+    assert ref is hist  # check if working inplace
+    check_invariants(hist)
+
+
 @pytest.fixture(scope='module')
 def __hist():
     # this FH is filled by