Revision: 19295
Author:   [email protected]
Date:     Tue Feb 11 18:23:07 2014 UTC
Log:      Experimental parser: store literals as ints

[email protected]

BUG=

Review URL: https://codereview.chromium.org/157813004
http://code.google.com/p/v8/source/detail?r=19295

Modified:
 /branches/experimental/parser/tools/lexer_generator/automata_test.py
 /branches/experimental/parser/tools/lexer_generator/nfa_builder.py
 /branches/experimental/parser/tools/lexer_generator/regex_parser.py
 /branches/experimental/parser/tools/lexer_generator/transition_key_test.py
 /branches/experimental/parser/tools/lexer_generator/transition_keys.py

=======================================
--- /branches/experimental/parser/tools/lexer_generator/automata_test.py Tue Feb 11 12:22:19 2014 UTC +++ /branches/experimental/parser/tools/lexer_generator/automata_test.py Tue Feb 11 18:23:07 2014 UTC
@@ -91,9 +91,9 @@
         'terminal' : False,
         'action' : Action.empty_action() }
     mapping = { k : empty_node() for k in ['S_0', 'S_1', 'S_2', 'S_3'] }
-    key_a = TransitionKey.single_char(encoding, 'a')
-    key_b = TransitionKey.single_char(encoding, 'b')
-    key_c = TransitionKey.single_char(encoding, 'c')
+    key_a = TransitionKey.single_char(encoding, ord('a'))
+    key_b = TransitionKey.single_char(encoding, ord('b'))
+    key_c = TransitionKey.single_char(encoding, ord('c'))

     mapping['S_0']['transitions'][key_a] = 'S_1'
     mapping['S_0']['transitions'][key_b] = 'S_2'
=======================================
--- /branches/experimental/parser/tools/lexer_generator/nfa_builder.py Fri Feb 7 08:31:41 2014 UTC +++ /branches/experimental/parser/tools/lexer_generator/nfa_builder.py Tue Feb 11 18:23:07 2014 UTC
@@ -128,7 +128,7 @@
     state.add_unclosed_transition(key)
     return (state, [state])

-  def __literal(self, chars):
+  def __literal(self, *chars):
     terms = map(lambda c : Term('SINGLE_CHAR', c), chars)
     return self.__process(self.cat_terms(terms))

@@ -320,24 +320,21 @@

   @staticmethod
   def __flatten_literals(terms):
-    literal = None
+    acc = ()
     for term in terms:
       assert isinstance(term, Term)
       if not term:
         continue
       if term.name() == 'LITERAL':
-        if literal:
-          literal += term.args()[0]
-        else:
-          literal = term.args()[0]
+        acc += term.args()
       else:
-        if literal:
-          yield Term('LITERAL', literal)
-          literal = None
+        if acc:
+          yield Term('LITERAL', *acc)
+          acc = ()
         if term:
           yield term
-    if literal:
-      yield Term('LITERAL', literal)
+    if acc:
+      yield Term('LITERAL', *acc)

   @staticmethod
   def or_terms(terms):
=======================================
--- /branches/experimental/parser/tools/lexer_generator/regex_parser.py Tue Feb 11 12:22:19 2014 UTC +++ /branches/experimental/parser/tools/lexer_generator/regex_parser.py Tue Feb 11 18:23:07 2014 UTC
@@ -238,7 +238,7 @@

   def p_literal(self, p):
     '''literal : LITERAL'''
-    p[0] = Term('LITERAL', p[1])
+    p[0] = Term('LITERAL', ord(p[1]))

   def p_any(self, p):
     '''any : ANY'''
@@ -262,10 +262,10 @@
                      | CHARACTER_CLASS maybe_class_content
     '''
     if len(p) == 5:
-      left = Term("RANGE", p[1], p[3])
+      left = Term("RANGE", ord(p[1]), ord(p[3]))
     else:
       if len(p[1]) == 1:
-        left = Term('LITERAL', p[1])
+        left = Term('LITERAL', ord(p[1]))
       else:
         left = Term('CHARACTER_CLASS', p[1][1:-1])
     p[0] = self.__cat(left, p[len(p)-1])
=======================================
--- /branches/experimental/parser/tools/lexer_generator/transition_key_test.py Tue Feb 11 12:22:19 2014 UTC +++ /branches/experimental/parser/tools/lexer_generator/transition_key_test.py Tue Feb 11 18:23:07 2014 UTC
@@ -36,8 +36,8 @@
   __equal_pairs = [
     (TransitionKey.epsilon(), TransitionKey.epsilon()),
     (TransitionKey.any(__encoding), TransitionKey.any(__encoding)),
-    (TransitionKey.single_char(__encoding, 'a'),
-     TransitionKey.single_char(__encoding, 'a')),
+    (TransitionKey.single_char(__encoding, ord('a')),
+     TransitionKey.single_char(__encoding, ord('a'))),
   ]

   def test_eq(self):
=======================================
--- /branches/experimental/parser/tools/lexer_generator/transition_keys.py Tue Feb 11 12:22:19 2014 UTC +++ /branches/experimental/parser/tools/lexer_generator/transition_keys.py Tue Feb 11 18:23:07 2014 UTC
@@ -25,6 +25,7 @@
 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

+from types import IntType
 from itertools import chain
 from action import Term
 from string import printable
@@ -144,8 +145,7 @@
   @staticmethod
   def omega():
     '''Always matches.'''
-    return TransitionKey.__cached_key(None, 'omega',
-      lambda : Term("OMEGA_KEY"))
+ return TransitionKey.__cached_key(None, 'omega', lambda : Term("OMEGA_KEY"))

   @staticmethod
   def any(encoding):
@@ -154,13 +154,14 @@
         lambda : encoding.all_components_iter())

   @staticmethod
-  def single_char(encoding, char):  # TODO(dcarney): char should be int
+  def single_char(encoding, char):
     '''Returns a TransitionKey for a single-character transition.'''
- return TransitionKey(encoding, Term("NUMERIC_RANGE_KEY", ord(char), ord(char)))
+    return TransitionKey.range(encoding, char, char)

   @staticmethod
   def range(encoding, a, b):
     '''Returns a TransitionKey for a single-character transition.'''
+    assert type(a) == IntType and type(b) == IntType
     return TransitionKey(encoding, Term("NUMERIC_RANGE_KEY", a, b))

   @staticmethod
@@ -176,10 +177,9 @@
     key = term.name()
     args = term.args()
     if key == 'RANGE':
- components.append(Term('NUMERIC_RANGE_KEY', ord(args[0]), ord(args[1])))
+      components.append(Term('NUMERIC_RANGE_KEY', args[0], args[1]))
     elif key == 'LITERAL':
-      for char in args[0]:  # TODO(dcarney): don't use strings for literals
-        components.append(Term('NUMERIC_RANGE_KEY', ord(char), ord(char)))
+      components += map(lambda x : Term('NUMERIC_RANGE_KEY', x, x), args)
     elif key == 'CAT':
       for x in args:
         TransitionKey.__process_term(encoding, x, components, key_map)

--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
--- You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to