Fix #50

skcriteria/cmp/ranks_cmp.py

@@ -88,11 +88,13 @@ class RanksComparator(Sequence, DiffEqualityMixin):
 
     def __init__(self, ranks):
         ranks = list(ranks)
-        self._validate_ranks(ranks)
         self._ranks = ranks
+        self._validate_ranks()
 
     # INTERNALS ===============================================================
-    def _validate_ranks(self, ranks):
+    def _validate_ranks(self):
+        ranks = self._ranks
+
         if len(ranks) <= 1:
             raise ValueError("Please provide more than one ranking")
 
@@ -228,6 +230,8 @@ def to_dataframe(self, *, untied=False):
 
         return df
 
+    # STATISTICALS ============================================================
+
     def corr(self, *, untied=False, **kwargs):
         """Compute pairwise correlation of rankings, excluding NA/null values.
 
@@ -371,6 +375,41 @@ def distance(self, *, untied=False, metric="hamming", **kwargs):
         )
         return dis_df
 
+    def extra_get(self, key, default=None):
+        """Retrieve a specific key from each rank, returning a \
+        dictionary of results.
+
+        This method iterates through all ranks and attempts to get the value
+        associated with the specified key. If the key is not found in a rank,
+        the default value is used.
+
+        Parameters
+        ----------
+        key : hashable
+            The key to look up in each rank.
+        default : any, optional
+            The value to return if the key is not found in a rank.
+            Defaults to None.
+
+        Returns
+        -------
+        dict
+            A dictionary where each key is the name of a rank, and each value
+            is the result of calling `get(key, default)` on that rank.
+
+        Notes
+        -----
+        The returned dictionary will have an entry for every rank, even if the
+        key was not found and the default value was used.
+
+        """
+        return {
+            rank_name: rank.extra_.get(key, default)
+            for rank_name, rank in self._ranks
+        }
+
+    eget = extra_get  # shortcut
+
     # ACCESSORS (YES, WE USE CACHED PROPERTIES IS THE EASIEST WAY) ============
 
     @methodtools.lru_cache(maxsize=None)

skcriteria/utils/bunch.py

@@ -108,3 +108,11 @@ def __repr__(self):
     def __dir__(self):
         """x.__dir__() <==> dir(x)."""
         return super().__dir__() + list(self._data)
+
+    def __setstate__(self, state):
+        """Needed for multiprocessing environment."""
+        self.__dict__.update(state)
+
+    def get(self, key, default=None):
+        """Get item from bunch."""
+        return self._data.get(key, default)

tests/cmp/test_ranks_cmp.py

@@ -271,6 +271,24 @@ def test_RanksComparator_hash():
     assert id(rcmp) == hash(rcmp)
 
 
+def test_RanksComparator_extra_get():
+    rank0 = agg.RankResult(
+        "test", ["a", "b"], [1, 1], {"alpha": 1, "bravo": 2}
+    )
+    rank1 = agg.RankResult(
+        "test", ["a", "b"], [1, 1], {"alpha": 1, "delta": 3}
+    )
+    rcmp = ranks_cmp.mkrank_cmp(rank0, rank1)
+
+    assert rcmp.extra_get("alpha") == {"test_1": 1, "test_2": 1}
+    assert rcmp.extra_get("bravo") == {"test_1": 2, "test_2": None}
+    assert rcmp.extra_get("delta", "foo") == {"test_1": "foo", "test_2": 3}
+    assert rcmp.extra_get("charly", "foo") == {
+        "test_1": "foo",
+        "test_2": "foo",
+    }
+
+
 def test_RanksComparator_plot():
     rank0 = agg.RankResult("test", ["a", "b"], [1, 1], {})
     rank1 = agg.RankResult("test", ["a", "b"], [1, 1], {})

tests/utils/test_bunch.py

@@ -19,6 +19,7 @@
 # =============================================================================
 
 import copy
+import pickle
 
 import pytest
 
@@ -97,3 +98,19 @@ def test_Bunch_assign_fails():
     foo_bunch = bunch.Bunch("foo", {})
     with pytest.raises(AttributeError, match="Bunch 'foo' is read-only"):
         foo_bunch.some_key = 1
+
+
+def test_Bunch_setstate():
+    md = bunch.Bunch("foo", {"alfa": 1})
+    md_c = pickle.loads(pickle.dumps(md))
+
+    assert md is not md_c
+    assert md._name == md_c._name  # string are inmutable never deep copy
+    assert md._data == md_c._data and md._data is not md_c._data
+
+
+def test_Bunch_get():
+    md = bunch.Bunch("foo", {"alfa": 1})
+    assert md.get("alfa") == 1
+    assert md.get("bravo") is None
+    assert md.get("bravo", 2) == 2