-- fixed width calculations for alternations

-- fixed literal check in branch operator
   (this broke test_tokenize, as reported by Mark Favas)
-- added REPEAT_ONE operator (still not enabled, though)
-- added some debugging stuff (maxlevel)

Lib/sre_compile.py

@@ -270,6 +270,7 @@ def _compile_info(code, pattern, flags):
                     table[i+1] = table[table[i+1]-1]+1
             code.extend(table[1:]) # don't store first entry
     elif charset:
+        # FIXME: use charset optimizer!
         for char in charset:
             emit(OPCODES[LITERAL])
             emit(char)
@@ -283,7 +284,7 @@ try:
 except NameError:
     pass
 
-def _compile1(p, flags):
+def _code(p, flags):
 
     flags = p.pattern.flags | flags
     code = []
@@ -308,7 +309,7 @@ def compile(p, flags=0):
     else:
         pattern = None
 
-    code = _compile1(p, flags)
+    code = _code(p, flags)
 
     # print code
 

Lib/sre_parse.py

@@ -137,12 +137,12 @@ class SubPattern:
         lo = hi = 0L
         for op, av in self.data:
             if op is BRANCH:
-                l = sys.maxint
-                h = 0
+                i = sys.maxint
+                j = 0
                 for av in av[1]:
-                    i, j = av.getwidth()
-                    l = min(l, i)
-                    h = min(h, j)
+                    l, h = av.getwidth()
+                    i = min(i, l)
+                    j = min(j, h)
                 lo = lo + i
                 hi = hi + j
             elif op is CALL:

Modules/sre.h

@@ -74,6 +74,8 @@ typedef struct {
     SRE_REPEAT *repeat; /* current repeat context */
     /* hooks */
     SRE_TOLOWER_HOOK lower;
+    /* debugging */
+    int maxlevel;
 } SRE_STATE;
 
 typedef struct {