Reallow non-callable key_fn in .cluster_objects()

Resolves #691

Thanks to @jfuruness for flagging.

CHANGELOG.md

@@ -11,6 +11,7 @@ All notable changes to this project will be documented in this file. The format
 ## Fixed
 
 - Fixed issue where `py.typed` file was not included in PyPi distribution. ([#698](https://github.com/jsvine/pdfplumber/issues/698) + [#703](https://github.com/jsvine/pdfplumber/pull/703)) [h/t @jhonatan-lopes]
+- Reinstated the ability to call `utils.cluster_objects(...)` with any hashable value (`str`, `int`, `tuple`, etc.) as the `key_fn` parameter, reverting breaking change in [58b1ab1](https://github.com/jsvine/pdfplumber/commit/58b1ab1). ([#691](https://github.com/jsvine/pdfplumber/issues/691)) [h/t @jfuruness]
 
 ### Development Changes
 

pdfplumber/utils.py

@@ -1,7 +1,7 @@
 import itertools
 import re
 import string
-from collections.abc import Sequence
+from collections.abc import Hashable, Sequence
 from operator import itemgetter
 from typing import (
     TYPE_CHECKING,
@@ -68,9 +68,12 @@ def make_cluster_dict(values: Iterable[T_num], tolerance: T_num) -> Dict[T_num,
 
 
 def cluster_objects(
-    xs: List[R], key_fn: Callable[[R], T_num], tolerance: T_num
+    xs: List[R], key_fn: Union[Hashable, Callable[[R], T_num]], tolerance: T_num
 ) -> List[List[R]]:
 
+    if not callable(key_fn):
+        key_fn = itemgetter(key_fn)
+
     values = map(key_fn, xs)
     cluster_dict = make_cluster_dict(values, tolerance)
 

tests/test_utils.py

@@ -43,6 +43,10 @@ def test_cluster_objects(self):
         a = ["a", "ab", "abc", "b"]
         assert utils.cluster_objects(a, len, 0) == [["a", "b"], ["ab"], ["abc"]]
 
+        b = [{"x": 1, 7: "a"}, {"x": 1, 7: "b"}, {"x": 2, 7: "b"}, {"x": 2, 7: "b"}]
+        assert utils.cluster_objects(b, "x", 0) == [[b[0], b[1]], [b[2], b[3]]]
+        assert utils.cluster_objects(b, 7, 0) == [[b[0]], [b[1], b[2], b[3]]]
+
     def test_resolve(self):
         annot = self.pdf.annots[0]
         annot_ad0 = utils.resolve(annot["data"]["A"]["D"][0])