Fix of SF bug #475877 (Mutable subtype instances are hashable).

Rather than tweaking the inheritance of type object slots (which turns
out to be too messy to try), this fix adds a __hash__ to the list and
dict types (the only mutable types I'm aware of) that explicitly
raises an error.  This has the advantage that list.__hash__([]) also
raises an error (previously, this would invoke object.__hash__([]),
returning the argument's address); ditto for dict.__hash__.

The disadvantage for this fix is that 3rd party mutable types aren't
automatically fixed.  This should be added to the rules for creating
subclassable extension types: if you don't want your object to be
hashable, add a tp_hash function that raises an exception.

Also, it's possible that I've forgotten about other mutable types for
which this should be done.

1 files changed, 24 insertions, 0 deletions

diff --git a/Lib/test/test_descr.py b/Lib/test/test_descr.py
index f3b81ac..d76013e 100644
--- a/Lib/test/test_descr.py
+++ b/Lib/test/test_descr.py
@@ -2571,6 +2571,29 @@ def delhook():
     del c
     vereq(log, [1])
 
+def hashinherit():
+    if verbose: print "Testing hash of mutable subclasses..."
+
+    class mydict(dict):
+        pass
+    d = mydict()
+    try:
+        hash(d)
+    except TypeError:
+        pass
+    else:
+        raise TestFailed, "hash() of dict subclass should fail"
+
+    class mylist(list):
+        pass
+    d = mylist()
+    try:
+        hash(d)
+    except TypeError:
+        pass
+    else:
+        raise TestFailed, "hash() of list subclass should fail"
+
 def test_main():
     class_docstrings()
     lists()
@@ -2623,6 +2646,7 @@ def test_main():
     str_of_str_subclass()
     kwdargs()
     delhook()
+    hashinherit()
     if verbose: print "All OK"
 
 if __name__ == "__main__":