- flush() refactor merged from uow_nontree branch r3871-r3885

- topological.py cleaned up, presents three public facing functions which
return list/tuple based structures, without exposing any internals.  only
the third function returns the "hierarchical" structure.  when results
include "cycles" or "child" items, 2- or 3- tuples are used to represent
results.
- unitofwork uses InstanceState almost exclusively now.  new and deleted lists
are now dicts which ref the actual object to provide a strong ref for the
duration that they're in those lists.  IdentitySet is only used for the public
facing versions of "new" and "deleted".
- unitofwork topological sort no longer uses the "hierarchical" version of the sort
for the base sort, only for the "per-object" secondary sort where it still
helps to group non-dependent operations together and provides expected insert
order.  the default sort deals with UOWTasks in a straight list and is greatly
simplified.  Tests all pass but need to see if svilen's stuff still works,
one block of code in _sort_cyclical_dependencies() seems to not be needed anywhere
but i definitely put it there for a reason at some point; if not hopefully we
can derive more test coverage from that.
- the UOWEventHandler is only applied to object-storing attributes, not
scalar (i.e. column-based) ones.  cuts out a ton of overhead when setting
non-object based attributes.
- InstanceState also used throughout the flush process, i.e. dependency.py,
mapper.save_obj()/delete_obj(), sync.execute() all expect InstanceState objects
in most cases now.
- mapper/property cascade_iterator() takes InstanceState as its argument,
but still returns lists of object instances so that they are not dereferenced.
- a few tricks needed when dealing with InstanceState, i.e. when loading a list
of items that are possibly fresh from the DB, you *have* to get the actual objects
into a strong-referencing datastructure else they fall out of scope immediately.
dependency.py caches lists of dependent objects which it loads now (i.e. history
collections).
- AttributeHistory is gone, replaced by a function that returns a 3-tuple of
added, unchanged, deleted.  these collections still reference the object
instances directly for the strong-referencing reasons mentiontioned, but
it uses less IdentitySet logic to generate.

2 files changed, 14 insertions, 16 deletions

diff --git a/lib/sqlalchemy/sql/compiler.py b/lib/sqlalchemy/sql/compiler.py
index 94950b872..3af8f97ca 100644
--- a/lib/sqlalchemy/sql/compiler.py
+++ b/lib/sqlalchemy/sql/compiler.py
@@ -459,10 +459,7 @@ class DefaultCompiler(engine.Compiled):
 
         stack_entry = {'select':select}
         
-        if asfrom:
-            stack_entry['is_subquery'] = True
-            column_clause_args = {}
-        elif self.stack and 'select' in self.stack[-1]:
+        if asfrom or (self.stack and 'select' in self.stack[-1]):
             stack_entry['is_subquery'] = True
             column_clause_args = {}
         else:
@@ -546,6 +543,7 @@ class DefaultCompiler(engine.Compiled):
 
     def get_select_precolumns(self, select):
         """Called when building a ``SELECT`` statement, position is just before column list."""
+
         return select._distinct and "DISTINCT " or ""
 
     def order_by_clause(self, select):
@@ -624,8 +622,8 @@ class DefaultCompiler(engine.Compiled):
             self.binds[col.key] = bindparam
             return self.bindparam_string(self._truncate_bindparam(bindparam))
 
-        self.postfetch = util.Set()
-        self.prefetch = util.Set()
+        self.postfetch = []
+        self.prefetch = []
         
         # no parameters in the statement, no parameters in the
         # compiled params - return binds for all columns
@@ -651,7 +649,7 @@ class DefaultCompiler(engine.Compiled):
                 if sql._is_literal(value):
                     value = create_bind_param(c, value)
                 else:
-                    self.postfetch.add(c)
+                    self.postfetch.append(c)
                     value = self.process(value.self_group())
                 values.append((c, value))
             elif isinstance(c, schema.Column):
@@ -663,35 +661,35 @@ class DefaultCompiler(engine.Compiled):
                             (c.default is not None and 
                              not isinstance(c.default, schema.Sequence))):
                             values.append((c, create_bind_param(c, None)))
-                            self.prefetch.add(c)
+                            self.prefetch.append(c)
                     elif isinstance(c.default, schema.ColumnDefault):
                         if isinstance(c.default.arg, sql.ClauseElement):
                             values.append((c, self.process(c.default.arg.self_group())))
                             if not c.primary_key:
                                 # dont add primary key column to postfetch
-                                self.postfetch.add(c)
+                                self.postfetch.append(c)
                         else:
                             values.append((c, create_bind_param(c, None)))
-                            self.prefetch.add(c)
+                            self.prefetch.append(c)
                     elif isinstance(c.default, schema.PassiveDefault):
                         if not c.primary_key:
-                            self.postfetch.add(c)
+                            self.postfetch.append(c)
                     elif isinstance(c.default, schema.Sequence):
                         proc = self.process(c.default)
                         if proc is not None:
                             values.append((c, proc))
                             if not c.primary_key:
-                                self.postfetch.add(c)
+                                self.postfetch.append(c)
                 elif self.isupdate:
                     if isinstance(c.onupdate, schema.ColumnDefault):
                         if isinstance(c.onupdate.arg, sql.ClauseElement):
                             values.append((c, self.process(c.onupdate.arg.self_group())))
-                            self.postfetch.add(c)
+                            self.postfetch.append(c)
                         else:
                             values.append((c, create_bind_param(c, None)))
-                            self.prefetch.add(c)
+                            self.prefetch.append(c)
                     elif isinstance(c.onupdate, schema.PassiveDefault):
-                        self.postfetch.add(c)
+                        self.postfetch.append(c)
         return values
 
     def visit_delete(self, delete_stmt):
diff --git a/lib/sqlalchemy/sql/util.py b/lib/sqlalchemy/sql/util.py
index de5797059..5aa985f47 100644
--- a/lib/sqlalchemy/sql/util.py
+++ b/lib/sqlalchemy/sql/util.py
@@ -16,7 +16,7 @@ def sort_tables(tables, reverse=False):
     vis = TVisitor()
     for table in tables:
         vis.traverse(table)
-    sequence = topological.QueueDependencySorter( tuples, tables).sort(ignore_self_cycles=True, create_tree=False)
+    sequence = topological.sort(tuples, tables)
     if reverse:
         return util.reversed(sequence)
     else: