diff --git a/schema.py b/schema.py
index 85a552e..9c1fc1e 100644
--- a/schema.py
+++ b/schema.py
@@ -1,4 +1,14 @@
-__version__ = '0.2.1'
+__version__ = '0.3.0'
+
+
+def unique(seq):
+    seen = set()
+    unique = []
+    for x in seq:
+        if x not in seen:
+            seen.add(x)
+            unique.append(x)
+    return unique
 
 
 class SchemaError(Exception):
@@ -12,31 +22,45 @@ def __init__(self, autos, errors):
 
     @property
     def code(self):
-        def uniq(seq):
-            seen = set()
-            seen_add = seen.add
-            return [x for x in seq if x not in seen and not seen_add(x)]
-        a = uniq(i for i in self.autos if i is not None)
-        e = uniq(i for i in self.errors if i is not None)
+        a = unique(i for i in self.autos if i is not None)
+        e = unique(i for i in self.errors if i is not None)
         if e:
             return '\n'.join(e)
         return '\n'.join(a)
 
 
-class And(object):
+class SchemaCutError(SchemaError):
+
+    """This exception is used to stop attempting further validation when
+    a Cut() object is reached. It can also be manually raised, but the
+    recommended way is to include a Cut([error]) in your schema."""
+    pass
+
+
+class SchemaBase(object):
+    priority = 4  # default priority for "validatables"
+
+    def validate(self, data):
+        raise NotImplementedError("redefine this method in subclasses")
+
+
+class And(SchemaBase):
 
     def __init__(self, *args, **kw):
         self._args = args
         assert list(kw) in (['error'], [])
         self._error = kw.get('error')
+        priority = kw.get('priority', None)
+        if priority is not None:
+            self.priority = priority
 
     def __repr__(self):
         return '%s(%s)' % (self.__class__.__name__,
                            ', '.join(repr(a) for a in self._args))
 
     def validate(self, data):
-        for s in [Schema(s, error=self._error) for s in self._args]:
-            data = s.validate(data)
+        for s in self._args:
+            data = Schema(s, error=self._error).validate(data)
         return data
 
 
@@ -44,21 +68,26 @@ class Or(And):
 
     def validate(self, data):
         x = SchemaError([], [])
-        for s in [Schema(s, error=self._error) for s in self._args]:
+        for s in self._args:
             try:
-                return s.validate(data)
+                return Schema(s, error=self._error).validate(data)
+            except SchemaCutError:
+                raise
             except SchemaError as _x:
                 x = _x
         raise SchemaError(['%r did not validate %r' % (self, data)] + x.autos,
                           [self._error] + x.errors)
 
 
-class Use(object):
+class Use(SchemaBase):
 
-    def __init__(self, callable_, error=None):
-        assert callable(callable_)
+    def __init__(self, callable_, error=None, priority=None):
+        if not callable(callable_):
+            raise ValueError("callable argument required")
         self._callable = callable_
         self._error = error
+        if priority is not None:
+            self.priority = priority
 
     def __repr__(self):
         return '%s(%r)' % (self.__class__.__name__, self._callable)
@@ -73,15 +102,47 @@ def validate(self, data):
             raise SchemaError('%s(%r) raised %r' % (f, data, x), self._error)
 
 
+class Cut(SchemaBase):
+    """
+    Cut([msg]) allows terminating validation immediately with a given error
+    message. This is useful, e.g. when another pattern in the same schema,
+    say a dictionary with a 'object: object' rule would accept anything and
+    the error produced message would be difficult to understand or an
+    exception wouldn't even be raised.
+
+    Example:
+        Schema({Optional("foo"): float,
+                object: object}).validate({"foo": "a"})
+
+    This would not even raise an exception, and we want it to NOT match "foo"
+    with the 'object: object' rule, so we can simply add a custom priority to
+    the 'Optional("foo")' key and a cut after 'float', like this:
+        Schema({Optional("foo", priority=0): Or(float, Cut("foobar")),
+                object: object}).validate({"foo": "a"})
+
+    This way we get the desired behavior: the "foo" rule is tested before
+    "object", and the cut will make validation fail immediately.
+    """
+    def __init__(self, error=None, priority=None):
+        self._error = error
+        if priority is not None:
+            self.priority = priority
+
+    def __repr__(self):
+        return '%s(%r)' % (self.__class__.__name__, self._error)
+
+    def validate(self, data):
+        raise SchemaCutError([None], [self._error])
+
+
 def priority(s):
-    """Return priority for a give object."""
     if type(s) in (list, tuple, set, frozenset):
         return 6
     if type(s) is dict:
         return 5
-    if hasattr(s, 'validate'):
-        return 4
-    if type(s) is type:
+    if isinstance(s, SchemaBase):
+        return s.priority
+    if isinstance(s, type):
         return 3
     if callable(s):
         return 2
@@ -89,72 +150,39 @@ def priority(s):
         return 1
 
 
-class Schema(object):
+class Schema(SchemaBase):
 
-    def __init__(self, schema, error=None):
+    def __init__(self, schema, error=None, priority=None):
         self._schema = schema
         self._error = error
+        if priority is not None:
+            self.priority = priority
 
     def __repr__(self):
         return '%s(%r)' % (self.__class__.__name__, self._schema)
 
     def validate(self, data):
-        s = self._schema
         e = self._error
-        if type(s) in (list, tuple, set, frozenset):
-            data = Schema(type(s), error=e).validate(data)
-            return type(s)(Or(*s, error=e).validate(d) for d in data)
-        if type(s) is dict:
-            data = Schema(dict, error=e).validate(data)
-            new = type(data)()  # new - is a dict of the validated values
-            x = None
-            coverage = set()  # non-optional schema keys that were matched
-            # for each key and value find a schema entry matching them, if any
-            sorted_skeys = list(sorted(s, key=priority))
-            for key, value in data.items():
-                valid = False
-                skey = None
-                for skey in sorted_skeys:
-                    svalue = s[skey]
-                    try:
-                        nkey = Schema(skey, error=e).validate(key)
-                    except SchemaError:
-                        pass
-                    else:
-                        try:
-                            nvalue = Schema(svalue, error=e).validate(value)
-                        except SchemaError as _x:
-                            x = _x
-                            raise
-                        else:
-                            coverage.add(skey)
-                            valid = True
-                            break
-                if valid:
-                    new[nkey] = nvalue
-                elif skey is not None:
-                    if x is not None:
-                        raise SchemaError(['invalid value for key %r' % key] +
-                                          x.autos, [e] + x.errors)
-            coverage = set(k for k in coverage if type(k) is not Optional)
-            required = set(k for k in s if type(k) is not Optional)
-            if coverage != required:
-                raise SchemaError('missed keys %r' % (required - coverage), e)
-            if len(new) != len(data):
-                wrong_keys = set(data.keys()) - set(new.keys())
-                s_wrong_keys = ', '.join('%r' % k for k in sorted(wrong_keys))
-                raise SchemaError('wrong keys %s in %r' % (s_wrong_keys, data),
-                                  e)
-            return new
-        if hasattr(s, 'validate'):
+        s = self._schema
+        s_type = type(s)
+        if s_type in (list, tuple, set, frozenset):
+            Schema(s_type).validate(data)
+            return s_type(Or(*s, error=e).validate(d) for d in data)
+        if s_type is dict:
+            Schema(s_type).validate(data)
+            return self._dict_validate(data)
+        if isinstance(s, SchemaBase):
             try:
                 return s.validate(data)
             except SchemaError as x:
-                raise SchemaError([None] + x.autos, [e] + x.errors)
+                # we must propagate the exception without losing its type
+                # otherwise cuts won't work correctly if we mask them with
+                # regular SchemaError exceptions
+                raise type(x)([None] + x.autos, [e] + x.errors)
             except BaseException as x:
                 raise SchemaError('%r.validate(%r) raised %r' % (s, data, x),
                                   self._error)
-        if type(s) is type:
+        if isinstance(s, type):
             if isinstance(data, s):
                 return data
             else:
@@ -165,16 +193,61 @@ def validate(self, data):
                 if s(data):
                     return data
             except SchemaError as x:
-                raise SchemaError([None] + x.autos, [e] + x.errors)
+                raise type(x)([None] + x.autos, [e] + x.errors)
             except BaseException as x:
                 raise SchemaError('%s(%r) raised %r' % (f, data, x),
                                   self._error)
-            raise SchemaError('%s(%r) should evaluate to True' % (f, data), e)
+            raise SchemaError('%s(%r) should evalutate to True' % (f, data), e)
         if s == data:
             return data
         else:
             raise SchemaError('%r does not match %r' % (s, data), e)
 
+    def _dict_validate(self, data):
+        s = self._schema
+        e = self._error
+        new = type(data)()
+        coverage = set()  # non-optional schema keys that were matched
+        skeys = sorted(s.keys(), key=priority)
+        for key, value in data.items():
+            valid = False
+            x = SchemaError([], [])
+            for skey in skeys:
+                # attempt to match the current key to any of the schema's keys
+                try:
+                    nkey = Schema(skey, error=e).validate(key)
+                except SchemaError:
+                    continue
+                # the key has matched, let's attempt to match the value
+                svalue = s[skey]
+                try:
+                    nvalue = Schema(svalue, error=e).validate(value)
+                except SchemaCutError as y:
+                    x = type(y)(y.autos + x.autos, y.errors + x.errors)
+                    break
+                except SchemaError as y:
+                    x = type(y)(y.autos + x.autos, y.errors + x.errors)
+                    continue
+                # both key and value matched, we can stop the loop
+                coverage.add(skey)
+                valid = True
+                break
+
+            if valid:
+                new[nkey] = nvalue
+            elif len(x.errors) > 0 or len(x.autos) > 0:
+                raise x
+            else:
+                raise SchemaError('unable to match %r to any schema key' % key, e)
+
+        coverage = set(k for k in coverage if type(k) is not Optional)
+        required = set(k for k in s if type(k) is not Optional)
+        if coverage != required:
+            raise SchemaError('missed keys %r' % (required - coverage), e)
+        if len(new) != len(data):
+            raise SchemaError('wrong keys %r in %r' % (new, data), e)
+        return new
+
 
 class Optional(Schema):
 
diff --git a/test_schema.py b/test_schema.py
index 079185e..fe19342 100644
--- a/test_schema.py
+++ b/test_schema.py
@@ -3,7 +3,8 @@
 
 from pytest import raises
 
-from schema import Schema, Use, And, Or, Optional, SchemaError
+from schema import (Schema, Use, And, Or, Optional, Cut,
+                    SchemaError, SchemaCutError)
 
 
 try:
@@ -113,28 +114,55 @@ def test_dict():
         try:
             Schema({'key': 5}).validate({'n': 5})
         except SchemaError as e:
-            assert e.args[0] in ["missed keys set(['key'])",
-                                 "missed keys {'key'}"]  # Python 3 style
+            assert e.args[0] == "unable to match 'n' to any schema key"
             raise
     with SE:
         try:
             Schema({}).validate({'n': 5})
         except SchemaError as e:
-            assert e.args[0] == "wrong keys 'n' in {'n': 5}"
+            assert e.args[0] == "unable to match 'n' to any schema key"
             raise
     with SE:
         try:
             Schema({'key': 5}).validate({'key': 5, 'bad': 5})
         except SchemaError as e:
-            assert e.args[0] in ["wrong keys 'bad' in {'key': 5, 'bad': 5}",
-                                 "wrong keys 'bad' in {'bad': 5, 'key': 5}"]
+            assert e.args[0] == "unable to match 'bad' to any schema key"
             raise
     with SE:
         try:
             Schema({}).validate({'a': 5, 'b': 5})
         except SchemaError as e:
-            assert e.args[0] in ["wrong keys 'a', 'b' in {'a': 5, 'b': 5}",
-                                 "wrong keys 'a', 'b' in {'b': 5, 'a': 5}"]
+            assert e.args[0] in ["unable to match 'a' to any schema key",
+                                 "unable to match 'b' to any schema key"]
+            raise
+
+
+def test_cuts():
+    with raises(SchemaCutError):
+        try:
+            Schema(Or(1, Cut("value is not equal to 1"))).validate(2)
+        except SchemaCutError as e:
+            assert e.args[0] == "value is not equal to 1"
+            raise
+
+    with raises(SchemaCutError):
+        try:
+            Schema(
+                {"foo": Or(float, Cut("foo is not a float")),
+                 str: object}
+            ).validate({"foo": "a"})
+        except SchemaCutError as e:
+            assert e.args[0] == "foo is not a float"
+            raise
+
+    with raises(SchemaCutError):
+        try:
+            Schema(
+                {Optional("foo", priority=0): Or(float, Cut("foo is not a float")),
+                 object: object}
+            ).validate({"foo": "a"})
+        except SchemaCutError as e:
+            assert e.args[0] == "foo is not a float"
             raise