improve hash function

CHANGELOG.md

@@ -1,5 +1,9 @@
 ## Changelog
 
+### [1.0.5] - 2022-06-29
+#### Fixed
+- treat hash for -1 and -2 as different
+
 ### [1.0.4] - 2022-06-23
 #### Changed
 - add fallback implementations of `jarowinkler-cpp` back to wheel,

jarowinkler/__init__.py

@@ -1,5 +1,5 @@
 __author__ = "Max Bachmann"
 __license__ = "MIT"
-__version__ = "1.0.4"
+__version__ = "1.0.5"
 
 from ._initialize import *

jarowinkler/common.pxd

@@ -24,6 +24,11 @@ cdef extern from "common.hpp":
     RF_String convert_string(object py_str)
     void validate_string(object py_str, const char* err) except +
 
+cdef inline uint64_t rf_hash(val) except *:
+    if val == -1:
+        return <uint64_t>-1
+    return <uint64_t>hash(val)
+
 cdef inline RF_String hash_array(arr) except *:
     # TODO on Cpython this does not require any copies
     cdef RF_String s_proc
@@ -64,7 +69,7 @@ cdef inline RF_String hash_array(arr) except *:
         else: # float/double are hashed
             s_proc.kind = RF_StringType.RF_UINT64
             for i in range(s_proc.length):
-                (<uint64_t*>s_proc.data)[i] = <uint64_t>hash(arr[i])
+                (<uint64_t*>s_proc.data)[i] = rf_hash(arr[i])
     except Exception as e:
         free(s_proc.data)
         s_proc.data = NULL
@@ -91,7 +96,7 @@ cdef inline RF_String hash_sequence(seq) except *:
             if isinstance(elem, str) and len(elem) == 1:
                 (<uint64_t*>s_proc.data)[i] = <uint64_t><Py_UCS4>elem
             else:
-                (<uint64_t*>s_proc.data)[i] = <uint64_t>hash(elem)
+                (<uint64_t*>s_proc.data)[i] = rf_hash(elem)
     except Exception as e:
         free(s_proc.data)
         s_proc.data = NULL

setup.py

@@ -6,7 +6,7 @@
 
 setup(
     name="jarowinkler",
-    version="1.0.4",
+    version="1.0.5",
     url="https://github.com/maxbachmann/JaroWinkler",
     author="Max Bachmann",
     author_email="[email protected]",

tests/test_JaroWinkler.py

@@ -12,6 +12,9 @@ def _jaro_winkler_similarity(self, s1, s2, result):
         self.assertAlmostEqual(jarowinkler_similarity(s1, s2), result, places=4)
         self.assertAlmostEqual(jarowinkler_similarity(s2, s1), result, places=4)
 
+    def test_hash_special_case(self):
+        self._jaro_winkler_similarity([0, -1], [0, -2], 0.6666)
+
     def test_edge_case_lengths(self):
         self._jaro_winkler_similarity('', '', 0)
         self._jaro_winkler_similarity('0', '0', 1)