pandas AttributeError : 'Series' object has no attribute 'iteritems' when using accepted.tolerance() on a MultiIndexed Series

[colinmatthieu]

The following example code using pandas 2.2.2 throws an AttributeError :

import pandas as pd
from datatest import validate, accepted

a = pd.DataFrame([['A', 1, 1], ['B', 2, 2]], columns=['c1', 'c2', 'c3']).set_index(['c1', 'c2'])

with accepted.tolerance(1):
    validate(a['c3'], 0)

The following error is raised :

File "[...]\venv\Lib\site-packages\datatest\acceptances.py", line 191, in __exit__
    raise exc_value
  File "<input>", line 2, in <module>
  File "[...]\venv\Lib\site-packages\datatest\validation.py", line 285, in __call__
    result = requirement_object(data)  # <- Apply requirement.
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "[...]\venv\Lib\site-packages\datatest\requirements.py", line 203, in __call__
    result = self.check_data(data)
             ^^^^^^^^^^^^^^^^^^^^^
  File "[...]venv\Lib\site-packages\datatest\requirements.py", line 267, in check_data
    data = normalize(data, lazy_evaluation=True)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "[...]\venv\Lib\site-packages\datatest\_normalize.py", line 159, in normalize
    obj = _normalize_lazy(obj)
          ^^^^^^^^^^^^^^^^^^^^
  File "[...]\venv\Lib\site-packages\datatest\_normalize.py", line 93, in _normalize_lazy
    return IterItems(obj.iteritems())  # <- EXIT!
                     ^^^^^^^^^^^^^
  File "[...]\venv\Lib\site-packages\pandas\core\generic.py", line 6299, in __getattr__
    return object.__getattribute__(self, name)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'Series' object has no attribute 'iteritems'

According to pandas docs, Series.iteritems is deprecated since pandas 2.0, and obj.items should now be used.

It is worth noting that changing obj.iteritems() to obj.items() in line 93 of _normalize.py removes this error and gives the expected result :

datatest.ValidationError: does not satisfy `0` (1 difference): {
    ('B', 2): Deviation(+2, 0),
}
```