Trac #14254: Add a signed_id() function returning an "id" as Py_ssize_t

src/sage/categories/homset.py

@@ -76,7 +76,7 @@
 # introduced in trac ticket #715
 
 from weakref import KeyedRef
-from sage.structure.coerce_dict import TripleDict
+from sage.structure.coerce_dict import signed_id, TripleDict
 _cache = TripleDict(53)
 
 def Hom(X, Y, category=None):
@@ -263,7 +263,7 @@ def Hom(X, Y, category=None):
     H = category.hom_category().parent_class(X, Y, category = category)
 
     ##_cache[key] = weakref.ref(H)
-    _cache[key] = KeyedRef(H, _cache.eraser, (id(X),id(Y),id(category)))
+    _cache[key] = KeyedRef(H, _cache.eraser, (signed_id(X),signed_id(Y),signed_id(category)))
     return H
 
 def hom(X, Y, f):

src/sage/structure/coerce_dict.pyx

@@ -44,23 +44,47 @@ cdef extern from "Python.h":
     PyObject* PyWeakref_GetObject(object ref)
     PyObject* Py_None
 
+cpdef inline Py_ssize_t signed_id(x):
+    """
+    A function like Python's :func:`id` returning *signed* integers,
+    which are guaranteed to fit in a ``Py_ssize_t``.
+
+    Theoretically, there is no guarantee that two different Python
+    objects have different ``signed_id()`` values. However, under the
+    mild assumption that a C pointer fits in a ``Py_ssize_t``, this
+    is guaranteed.
+
+    TESTS::
+
+        sage: a = 1.23e45  # some object
+        sage: from sage.structure.coerce_dict import signed_id
+        sage: s = signed_id(a)
+        sage: id(a) == s or id(a) == s + 2**32 or id(a) == s + 2**64
+        True
+        sage: signed_id(a) <= sys.maxsize
+        True
+    """
+    return <Py_ssize_t><void *>(x)
+
 import gc
 
 ############################################
 # A note about how to store "id" keys in python structures:
 #
-# In python a "pointer length integer" (size_t) normally, is encoded
-# as a *signed* integer, of type Py_ssize_t. This has an advantage in that
-# if the value gets encoded as a *python integer* it can do so in a sign-preserving
-# way and still make use of all the bits that python offers to store (small) integers.
+# We use the type Py_ssize_t to store "id"s generated by the signed_id
+# function defined above. Assuming that Py_ssize_t is the same as a C
+# long (which is true on most Unix-like systems), this also has the
+# advantage that these Py_ssize_t values are stored as a Python "int"
+# (as opposed to "long"), which allow for fast conversion to/from C
+# types.
 #
 # There is one place where we have to be careful about signs:
 # Our hash values most naturally live in Py_ssize_t. We convert those into
 # an index into our bucket list by taking the hash modulo the number of buckets.
 # However, the modulo operator in C preserves the sign of the number we take the
 # modulus of, which is not what we want.
 # The solution is to always do
-# (<size_t) h)% modulus
+# (<size_t> h) % modulus
 # to ensure we're doing an unsigned modulus.
 
 ############################################
@@ -528,7 +552,7 @@ cdef class MonoDict:
             sage: 15 in L
             False
         """
-        cdef Py_ssize_t h = <Py_ssize_t><void *>k
+        cdef Py_ssize_t h = signed_id(k)
         cdef Py_ssize_t i
         cdef list all_buckets = self.buckets
         cdef list bucket = <object>PyList_GET_ITEM(all_buckets, (<size_t>h)% PyList_GET_SIZE(all_buckets))
@@ -571,7 +595,7 @@ cdef class MonoDict:
         return self.get(k)
 
     cdef get(self, object k):
-        cdef Py_ssize_t h =<Py_ssize_t><void *>k
+        cdef Py_ssize_t h = signed_id(k)
         cdef Py_ssize_t i
         cdef list all_buckets = self.buckets
         cdef list bucket = <object>PyList_GET_ITEM(all_buckets, (<size_t>h) % PyList_GET_SIZE(all_buckets))
@@ -608,7 +632,7 @@ cdef class MonoDict:
     cdef set(self,object k, value):
         if self.threshold and self._size > len(self.buckets) * self.threshold:
             self.resize()
-        cdef Py_ssize_t h = <Py_ssize_t><void *>k
+        cdef Py_ssize_t h = signed_id(k)
         cdef Py_ssize_t i
         cdef list bucket = <object>PyList_GET_ITEM(self.buckets,(<size_t> h) % PyList_GET_SIZE(self.buckets))
         cdef object r
@@ -679,7 +703,7 @@ cdef class MonoDict:
             sage: a in L
             False
         """
-        cdef Py_ssize_t h = <Py_ssize_t><void *>k
+        cdef Py_ssize_t h = signed_id(k)
         cdef object r
         cdef Py_ssize_t i
         cdef object tmp
@@ -1101,9 +1125,9 @@ cdef class TripleDict:
         return self.get(k1, k2, k3)
 
     cdef get(self, object k1, object k2, object k3):
-        cdef Py_ssize_t h1 = <Py_ssize_t><void *>k1
-        cdef Py_ssize_t h2 = <Py_ssize_t><void *>k2
-        cdef Py_ssize_t h3 = <Py_ssize_t><void *>k3
+        cdef Py_ssize_t h1 = signed_id(k1)
+        cdef Py_ssize_t h2 = signed_id(k2)
+        cdef Py_ssize_t h3 = signed_id(k3)
         cdef Py_ssize_t h = (h1 + 13*h2 ^ 503*h3)
 
         cdef object r1,r2,r3
@@ -1149,9 +1173,9 @@ cdef class TripleDict:
     cdef set(self, object k1, object k2, object k3, value):
         if self.threshold and self._size > len(self.buckets) * self.threshold:
             self.resize()
-        cdef Py_ssize_t h1 = <Py_ssize_t><void *>k1
-        cdef Py_ssize_t h2 = <Py_ssize_t><void *>k2
-        cdef Py_ssize_t h3 = <Py_ssize_t><void *>k3
+        cdef Py_ssize_t h1 = signed_id(k1)
+        cdef Py_ssize_t h2 = signed_id(k2)
+        cdef Py_ssize_t h3 = signed_id(k3)
         cdef Py_ssize_t h = (h1 + 13*h2 ^ 503*h3)
 
         cdef object r1,r2,r3
@@ -1234,9 +1258,9 @@ cdef class TripleDict:
             k1, k2, k3 = k
         except (TypeError,ValueError):
             raise KeyError, k
-        cdef Py_ssize_t h1 = <Py_ssize_t><void *>k1
-        cdef Py_ssize_t h2 = <Py_ssize_t><void *>k2
-        cdef Py_ssize_t h3 = <Py_ssize_t><void *>k3
+        cdef Py_ssize_t h1 = signed_id(k1)
+        cdef Py_ssize_t h2 = signed_id(k2)
+        cdef Py_ssize_t h3 = signed_id(k3)
         cdef Py_ssize_t h = (h1 + 13*h2 ^ 503*h3)
 
         cdef Py_ssize_t i