Revert copy on write optimization

In case of multiple renames, all of the resulting relations will point
to the same data, so writing on one and marking its only known copy
as writable would be incorrect.

For this to work, all the references should be kown to all the other
instances.

It is probably not worth to implement it.

Implemented test for the update

relational/relation.py

@@ -54,7 +54,6 @@ class Relation (object):
 
     def __init__(self, filename=""):
         self._readonly = False
-        self._copy = None
         self.content = set()
 
         if len(filename) == 0:  # Empty relation
@@ -71,8 +70,6 @@ class Relation (object):
 
         self._readonly = True
         copy._readonly = True
-        copy._copy = self
-        self._copy = copy
 
     def _make_writable(self, copy_content=True):
         '''If this relation is marked as readonly, this
@@ -83,8 +80,6 @@ class Relation (object):
 
         if self._readonly:
             self._readonly = False
-            self._copy._readonly = False
-            self._copy = None
 
             if copy_content:
                 self.content = set(self.content)

test/update.py

@@ -0,0 +1,16 @@
+p1=people
+p2=p1.rename({'id':'i'})
+p2=p2.rename({'i':'id'})
+assert p1==p2
+assert p1._readonly
+assert p2._readonly
+# It is VERY important to not change the original relations
+# or other tests might fail randomly, since the relations are
+# only loaded once
+
+p2.update('age==20', {'age':50})
+assert p2._readonly == False
+assert p1!=p2
+p3 = p2.selection('age!=50')
+p4 = p1.selection('age!=20')
+assert p3==p4