Modified subset to be a generic named subset_by_sequence.

This gives us some more flexibility for registering downstream classes.

src/biocutils/__init__.py

@@ -25,7 +25,6 @@
 from .normalize_subscript import normalize_subscript
 from .print_truncated import print_truncated, print_truncated_dict, print_truncated_list
 from .print_wrapped_table import create_floating_names, print_type, print_wrapped_table, truncate_strings
-from .subset import subset
 from .union import union
 
 from .combine import combine
@@ -35,5 +34,8 @@
 
 from .extract_row_names import extract_row_names
 from .extract_column_names import extract_column_names
+
+from .subset_sequence import subset_sequence
+
 from .show_as_cell import show_as_cell
 from .convert_to_dense import convert_to_dense

src/biocutils/print_wrapped_table.py

@@ -1,7 +1,7 @@
 from typing import List, Optional, Sequence
 import numpy
 
-from .subset import subset
+from .subset_sequence import subset_sequence
 
 
 def _get_max_width(col: List[str]):
@@ -118,7 +118,7 @@ def create_floating_names(
         List of strings containing floating names.
     """
     if names is not None:
-        return subset(names, indices)
+        return subset_sequence(names, indices)
     else:
         return ["[" + str(i) + "]" for i in indices]
 

src/biocutils/subset_sequence.py

@@ -0,0 +1,28 @@
+from typing import Any, Sequence, Union
+from functools import singledispatch
+import numpy
+
+
+@singledispatch
+def subset_sequence(x: Any, indices: Sequence) -> Any:
+    """
+    Subset ``x`` by ``indices`` to obtain a new object with the desired
+    subset of elements. This attempts to use ``x``'s ``__getitem__`` method.
+
+    Args:
+        x:
+            Any object that supports ``__getitem__`` with an integer sequence.
+
+        indices:
+            Sequence of non-negative integers specifying the integers of interest.
+
+    Returns:
+        The result of slicing ``x`` by ``indices``. The exact type
+        depends on what ``x``'s ``__getitem__`` method returns.
+    """
+    return x[indices]
+
+
+@subset_sequence.register
+def _subset_sequence_list(x: list, indices: Sequence) -> list:
+    return [x[i] for i in indices]

tests/test_subset_sequence.py

@@ -0,0 +1,21 @@
+from biocutils import subset_sequence
+import numpy as np
+
+
+def test_subset_list():
+    x = [1, 2, 3, 4, 5]
+    assert subset_sequence(x, [0, 2, 4]) == [1, 3, 5]
+
+    x = [1, 2, 3, 4, 5]
+    assert subset_sequence(x, range(5)) == x
+
+    x = [1, 2, 3, 4, 5]
+    assert subset_sequence(x, range(4, -1, -1)) == [5, 4, 3, 2, 1]
+
+
+def test_subset_numpy():
+    y = np.random.rand(10)
+    assert (subset_sequence(y, range(5)) == y[0:5]).all()
+
+    y = np.random.rand(10, 20)
+    assert (subset_sequence(y, range(5)) == y[0:5, :]).all()