[Python-checkins] CVS: python/dist/src/Lib sre.py,1.22,1.23 sre_compile.py,1.28,1.29 sre_constants.py,1.19,1.20 sre_parse.py,1.28,1.29

[Fredrik Lundh]

Update of /cvsroot/python/python/dist/src/Lib
In directory slayer.i.sourceforge.net:/tmp/cvs-serv10513/Lib

Modified Files:
	sre.py sre_compile.py sre_constants.py sre_parse.py 
Log Message:


final 0.9.8 updates:

-- added REPEAT_ONE operator
-- added ANY_ALL operator (used to represent "(?s).")


Index: sre.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/sre.py,v
retrieving revision 1.22
retrieving revision 1.23
diff -C2 -r1.22 -r1.23
*** sre.py	2000/08/01 18:20:06	1.22
--- sre.py	2000/08/01 22:47:48	1.23
***************
*** 99,103 ****
      except KeyError:
          pass
!     p = sre_compile.compile(pattern, flags)
      if len(_cache) >= _MAXCACHE:
          _cache.clear()
--- 99,106 ----
      except KeyError:
          pass
!     try:
!         p = sre_compile.compile(pattern, flags)
!     except error, v:
!         raise error, v # invalid expression
      if len(_cache) >= _MAXCACHE:
          _cache.clear()

Index: sre_compile.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/sre_compile.py,v
retrieving revision 1.28
retrieving revision 1.29
diff -C2 -r1.28 -r1.29
*** sre_compile.py	2000/08/01 21:05:40	1.28
--- sre_compile.py	2000/08/01 22:47:48	1.29
***************
*** 74,77 ****
--- 74,84 ----
      return charset
  
+ def _simple(av):
+     # check if av is a "simple" operator
+     lo, hi = av[2].getwidth()
+     if lo == 0:
+         raise error, "nothing to repeat"
+     return lo == hi == 1 and av[2][0][0] != SUBPATTERN
+ 
  def _compile(code, pattern, flags):
      # internal: compile a (sub)pattern
***************
*** 117,124 ****
          elif op is ANY:
              if flags & SRE_FLAG_DOTALL:
!                 emit(OPCODES[op])
              else:
!                 emit(OPCODES[CATEGORY])
!                 emit(CHCODES[CATEGORY_NOT_LINEBREAK])
          elif op in (REPEAT, MIN_REPEAT, MAX_REPEAT):
              if flags & SRE_FLAG_TEMPLATE:
--- 124,130 ----
          elif op is ANY:
              if flags & SRE_FLAG_DOTALL:
!                 emit(OPCODES[ANY_ALL])
              else:
!                 emit(OPCODES[ANY])
          elif op in (REPEAT, MIN_REPEAT, MAX_REPEAT):
              if flags & SRE_FLAG_TEMPLATE:
***************
*** 131,158 ****
                  emit(OPCODES[SUCCESS])
                  code[skip] = len(code) - skip
              else:
!                 lo, hi = av[2].getwidth()
!                 if lo == 0:
!                     raise error, "nothing to repeat"
!                 if 0 and lo == hi == 1 and op is MAX_REPEAT:
!                     # FIXME: <fl> fast and wrong (but we'll fix that)
!                     emit(OPCODES[REPEAT_ONE])
!                     skip = len(code); emit(0)
!                     emit(av[0])
!                     emit(av[1])
!                     _compile(code, av[2], flags)
!                     emit(OPCODES[SUCCESS])
!                     code[skip] = len(code) - skip
                  else:
!                     emit(OPCODES[REPEAT])
!                     skip = len(code); emit(0)
!                     emit(av[0])
!                     emit(av[1])
!                     _compile(code, av[2], flags)
!                     code[skip] = len(code) - skip
!                     if op == MAX_REPEAT:
!                         emit(OPCODES[MAX_UNTIL])
!                     else:
!                         emit(OPCODES[MIN_UNTIL])
          elif op is SUBPATTERN:
              if av[0]:
--- 137,159 ----
                  emit(OPCODES[SUCCESS])
                  code[skip] = len(code) - skip
+             elif _simple(av) and op == MAX_REPEAT:
+                 emit(OPCODES[REPEAT_ONE])
+                 skip = len(code); emit(0)
+                 emit(av[0])
+                 emit(av[1])
+                 _compile(code, av[2], flags)
+                 emit(OPCODES[SUCCESS])
+                 code[skip] = len(code) - skip
              else:
!                 emit(OPCODES[REPEAT])
!                 skip = len(code); emit(0)
!                 emit(av[0])
!                 emit(av[1])
!                 _compile(code, av[2], flags)
!                 code[skip] = len(code) - skip
!                 if op == MAX_REPEAT:
!                     emit(OPCODES[MAX_UNTIL])
                  else:
!                     emit(OPCODES[MIN_UNTIL])
          elif op is SUBPATTERN:
              if av[0]:

Index: sre_constants.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/sre_constants.py,v
retrieving revision 1.19
retrieving revision 1.20
diff -C2 -r1.19 -r1.20
*** sre_constants.py	2000/08/01 18:20:06	1.19
--- sre_constants.py	2000/08/01 22:47:48	1.20
***************
*** 21,24 ****
--- 21,25 ----
  
  ANY = "any"
+ ANY_ALL = "any_all"
  ASSERT = "assert"
  ASSERT_NOT = "assert_not"
***************
*** 82,86 ****
      FAILURE, SUCCESS,
  
!     ANY,
      ASSERT, ASSERT_NOT,
      AT,
--- 83,87 ----
      FAILURE, SUCCESS,
  
!     ANY, ANY_ALL,
      ASSERT, ASSERT_NOT,
      AT,

Index: sre_parse.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/sre_parse.py,v
retrieving revision 1.28
retrieving revision 1.29
diff -C2 -r1.28 -r1.29
*** sre_parse.py	2000/08/01 21:05:40	1.28
--- sre_parse.py	2000/08/01 22:47:48	1.29
***************
*** 143,147 ****
                      l, h = av.getwidth()
                      i = min(i, l)
!                     j = min(j, h)
                  lo = lo + i
                  hi = hi + j
--- 143,147 ----
                      l, h = av.getwidth()
                      i = min(i, l)
!                     j = max(j, h)
                  lo = lo + i
                  hi = hi + j