zero-quantity outcomes are pruned from Die/Deck construction

src/icepool/creation_args.py

@@ -104,8 +104,12 @@ def merge_weights_lcm(subdatas: Sequence[Mapping[T, int]],
     data: MutableMapping[Any, int] = defaultdict(int)
     for subdata, subdata_denominator, w in zip(subdatas, subdata_denominators,
                                                weights):
-        factor = denominator_lcm * w // subdata_denominator if subdata_denominator else 0
+        if subdata_denominator == 0 or w == 0:
+            continue
+        factor = denominator_lcm * w // subdata_denominator
         for outcome, weight in subdata.items():
+            if weight == 0:
+                continue
             data[outcome] += weight * factor
     return data
 
@@ -122,7 +126,11 @@ def merge_duplicates(subdatas: Sequence[Mapping[T, int]],
 
     data: MutableMapping[Any, int] = defaultdict(int)
     for subdata, subdup in zip(subdatas, duplicates):
+        if subdup == 0:
+            continue
         for outcome, dup in subdata.items():
+            if dup == 0:
+                continue
             data[outcome] += dup * subdup
 
     return data

src/icepool/function.py

@@ -198,11 +198,16 @@ def from_rv(rv, outcomes: Sequence[float], denominator: int,
 def from_rv(rv, outcomes: Sequence[int] | Sequence[float], denominator: int,
             **kwargs) -> 'icepool.Die[int] | icepool.Die[float]':
     """Constructs a `Die` from a rv object (as `scipy.stats`).
+
+    This is done using the CDF.
+
     Args:
         rv: A rv object (as `scipy.stats`).
         outcomes: An iterable of `int`s or `float`s that will be the outcomes
             of the resulting `Die`.
             If the distribution is discrete, outcomes must be `int`s.
+            Some outcomes may be omitted if their probability is too small
+            compared to the denominator.
         denominator: The denominator of the resulting `Die` will be set to this.
         **kwargs: These will be forwarded to `rv.cdf()`.
     """

src/icepool/population/deck.py

@@ -34,6 +34,9 @@ def __new__(cls,
                 times: Sequence[int] | int = 1) -> 'Deck[T_co]':
         """Constructor for a `Deck`.
 
+        All quantities must be non-negative. Outcomes with zero quantity will be
+        omitted.
+
         Args:
             outcomes: The cards of the `Deck`. This can be one of the following:
                 * A `Sequence` of outcomes. Duplicates will contribute

src/icepool/population/die.py

@@ -87,7 +87,8 @@ def __new__(
         * Use a dict: `Die({1:1, 2:1, 3:1, 4:1, 5:1, 6:1})`
         * Give the faces as a sequence: `Die([1, 2, 3, 4, 5, 6])`
 
-        All quantities must be non-negative, though they can be zero.
+        All quantities must be non-negative. Outcomes with zero quantity will be
+        omitted.
 
         Several methods and functions foward **kwargs to this constructor.
         However, these only affect the construction of the returned or yielded

tests/comparator_test.py

@@ -106,7 +106,7 @@ def test_cmp_len():
     assert len(icepool.d6.cmp(0)) == 1
     assert len(icepool.d6.cmp(7)) == 1
     assert len(icepool.Die([1]).cmp(1)) == 1
-    assert len(icepool.Die({-1: 0, 0: 0, 1: 0}).cmp(0)) == 3
+    assert len(icepool.Die({-1: 0, 0: 0, 1: 0}).cmp(0)) == 0
 
 
 def test_quantity_le():

tests/evaluator_test.py

@@ -51,7 +51,7 @@ def test_sum_descending_keep_highest():
 
 def test_zero_weight_outcomes():
     result = icepool.Die(range(5), times=[0, 1, 0, 1, 0]).highest(3, 2)
-    assert len(result) == 9
+    assert len(result) == 3
 
 
 def sum_dice_func(state, outcome, count):

tests/from_cumulative_test.py

@@ -27,5 +27,5 @@ def test_from_rv_norm():
                                1000000,
                                loc=die.mean(),
                                scale=die.standard_deviation())
-    assert die.probabilities('<=') == pytest.approx(
-        norm_die.probabilities('<='), abs=1e-3)
+    assert [die.probability('<=', x) for x in range(600)] == pytest.approx(
+        [norm_die.probability('<=', x) for x in range(600)], abs=1e-3)