Ticket #13184: trac_13184_homset_unique_parent

# HG changeset patch
# User Xavier Caruso <xavier.caruso@normalesup.org>
# Date 1341738209 -7200
# Node ID 09ba9b7f74b31e781e40c0f2bbe765e6c9c7bdb2
# Parent  c4bdbf18dd78c13ff3032e07a3f355b8632ac054
Trac #13184: Unique parent assumption for homset

diff --git a/sage/categories/homset.py b/sage/categories/homset.py
--- a/sage/categories/homset.py
+++ b/sage/categories/homset.py
@@ -154,9 +154,17 @@
 
     TESTS:
 
-    Some doc tests in :mod:`sage.rings` (need to) break the unique parent assumption.
-    But if domain or codomain are not unique parents, then the hom set won't fit.
-    That's to say, the hom set found in the cache will have a (co)domain that is
+    Homset are unique parents::
+
+        sage: k = GF(5)
+        sage: H1 = Hom(k,k)
+        sage: H2 = Hom(k,k)
+        sage: H1 is H2
+        True
+
+    However, some doc tests in :mod:`sage.rings` (need to) break the unique parent 
+    assumption. If domain or codomain are not unique parents, then the hom set won't 
+    fit. That's to say, the hom set found in the cache will have a (co)domain that is
     equal to, but not identic with, the given (co)domain.
 
     By trac ticket #9138, we abandon the uniqueness of hom sets, if the domain or
@@ -221,8 +229,12 @@
             return H
 
     try:
-        # Apparently X._Hom_ is supposed to be cached
-        return X._Hom_(Y, category)
+        # Actually X._Hom_ is supposed to be cached
+        # but it is not in some cases (e.g. X is a finite field)
+        # To be investigated
+        H = X._Hom_(Y,category)
+        _cache[key] = weakref.ref(H)
+        return H
     except (AttributeError, TypeError):
         pass