Add ValueErrorTuple

This just prints a value and an error nicely, e.g. in the report.

src/reportengine/floatformatting.py

@@ -5,6 +5,8 @@
 it looks like.
 """
 import decimal
+import numbers
+from typing import NamedTuple
 
 def significant_digits(value, digits):
     """Return a `Decimal` object with all the digits less signingicant than
@@ -69,3 +71,12 @@ def format_error_value_columns(df, valcol, errcol, inplace=False, **kwargs):
 
     if not inplace:
         return df
+
+class ValueErrorTuple(NamedTuple):
+    """A class to represent a value and its error. The only functionality it
+    adds is the ``__str__`` method, to represent the quantity."""
+    value: numbers.Real
+    error: numbers.Real
+    def __str__(self):
+        valstr, errstr = format_value_error(self.value, self.error)
+        return f'{valstr}±{errstr}'

src/reportengine/tests/test_floatformatting.py

@@ -5,7 +5,9 @@
 import pandas as pd
 import numpy as np
 
-from reportengine.floatformatting import format_number, significant_digits, format_error_value_columns
+from reportengine.floatformatting import (format_number, significant_digits,
+                                          format_error_value_columns,
+                                          ValueErrorTuple)
 
 @given(floats(allow_nan=False))
 def test_format_rountrip(x):
@@ -24,3 +26,7 @@ def test_cols(valerr):
     formatted = format_error_value_columns(df, 'Val', 'Err')
     format_error_value_columns(df, 'Val', 'Err', inplace=True)
     assert (formatted == df).all().all()
+
+@given(floats(), floats())
+def test_valueerrortuple(value, error):
+    assert '±' in str(ValueErrorTuple(value,error))