[Python-checkins] python/dist/src/Lib/test test_codeccallbacks.py,1.8,1.9

[doerwalter@users.sourceforge.net]

Update of /cvsroot/python/python/dist/src/Lib/test
In directory sc8-pr-cvs1:/tmp/cvs-serv3391/dist/src/Lib/test

Modified Files:
	test_codeccallbacks.py 
Log Message:
Change the treatment of positions returned by PEP293
error handers in the Unicode codecs: Negative
positions are treated as being relative to the end of
the input and out of bounds positions result in an
IndexError.

Also update the PEP and include an explanation of
this in the documentation for codecs.register_error.

Fixes a small bug in iconv_codecs: if the position
from the callback is negative *add* it to the size
instead of substracting it.

>From SF patch #677429.


Index: test_codeccallbacks.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/test/test_codeccallbacks.py,v
retrieving revision 1.8
retrieving revision 1.9
diff -C2 -d -r1.8 -r1.9
*** test_codeccallbacks.py	20 Jan 2003 02:34:07 -0000	1.8
--- test_codeccallbacks.py	31 Jan 2003 17:19:07 -0000	1.9
***************
*** 2,5 ****
--- 2,22 ----
  import sys, codecs, htmlentitydefs, unicodedata
  
+ class PosReturn:
+     # this can be used for configurable callbacks
+ 
+     def __init__(self):
+         self.pos = 0
+ 
+     def handle(self, exc):
+         oldpos = self.pos
+         realpos = oldpos
+         if realpos<0:
+            realpos = len(exc.object) + realpos
+         # if we don't advance this time, terminate on the next call
+         # otherwise we'd get an endless loop
+         if realpos <= exc.start:
+             self.pos = len(exc.object)
+         return (u"<?>", oldpos)
+ 
  class CodecCallbackTest(unittest.TestCase):
  
***************
*** 544,559 ****
          self.assertRaises(TypeError, "\xff".decode, "ascii", "test.baddecodereturn2")
  
!         pos = [-42]
!         def negposreturn(exc):
!             pos[0] += 1 # use list to work around scoping problem
!             return (u"?", pos[0])
!         codecs.register_error("test.negposreturn", negposreturn)
!         "\xff".decode("ascii", "test.negposreturn")
  
!         def hugeposreturn(exc):
!             return (u"?", 424242)
!         codecs.register_error("test.hugeposreturn", hugeposreturn)
!         "\xff".decode("ascii", "test.hugeposreturn")
!         "\\uyyyy".decode("raw-unicode-escape", "test.hugeposreturn")
  
          class D(dict):
--- 561,594 ----
          self.assertRaises(TypeError, "\xff".decode, "ascii", "test.baddecodereturn2")
  
!         handler = PosReturn()
!         codecs.register_error("test.posreturn", handler.handle)
  
!         # Valid negative position
!         handler.pos = -1
!         self.assertEquals("\xff0".decode("ascii", "test.posreturn"), u"<?>0")
! 
!         # Valid negative position
!         handler.pos = -2
!         self.assertEquals("\xff0".decode("ascii", "test.posreturn"), u"<?><?>")
! 
!         # Negative position out of bounds
!         handler.pos = -3
!         self.assertRaises(IndexError, "\xff0".decode, "ascii", "test.posreturn")
! 
!         # Valid positive position
!         handler.pos = 1
!         self.assertEquals("\xff0".decode("ascii", "test.posreturn"), u"<?>0")
! 
!         # Largest valid positive position (one beyond end of input
!         handler.pos = 2
!         self.assertEquals("\xff0".decode("ascii", "test.posreturn"), u"<?>")
! 
!         # Invalid positive position
!         handler.pos = 3
!         self.assertRaises(IndexError, "\xff0".decode, "ascii", "test.posreturn")
! 
!         # Restart at the "0"
!         handler.pos = 6
!         self.assertEquals("\\uyyyy0".decode("raw-unicode-escape", "test.posreturn"), u"<?>0")
  
          class D(dict):
***************
*** 580,599 ****
          self.assertRaises(TypeError, u"\xff".encode, "ascii", "test.badencodereturn2")
  
!         pos = [-42]
!         def negposreturn(exc):
!             pos[0] += 1 # use list to work around scoping problem
!             return (u"?", pos[0])
!         codecs.register_error("test.negposreturn", negposreturn)
!         u"\xff".encode("ascii", "test.negposreturn")
  
!         def hugeposreturn(exc):
!             return (u"?", 424242)
!         codecs.register_error("test.hugeposreturn", hugeposreturn)
!         u"\xff".encode("ascii", "test.hugeposreturn")
  
          class D(dict):
              def __getitem__(self, key):
                  raise ValueError
!         for err in ("strict", "replace", "xmlcharrefreplace", "backslashreplace", "test.hugeposreturn"):
              self.assertRaises(UnicodeError, codecs.charmap_encode, u"\xff", err, {0xff: None})
              self.assertRaises(ValueError, codecs.charmap_encode, u"\xff", err, D())
--- 615,651 ----
          self.assertRaises(TypeError, u"\xff".encode, "ascii", "test.badencodereturn2")
  
!         handler = PosReturn()
!         codecs.register_error("test.posreturn", handler.handle)
  
!         # Valid negative position
!         handler.pos = -1
!         self.assertEquals(u"\xff0".encode("ascii", "test.posreturn"), "<?>0")
! 
!         # Valid negative position
!         handler.pos = -2
!         self.assertEquals(u"\xff0".encode("ascii", "test.posreturn"), "<?><?>")
! 
!         # Negative position out of bounds
!         handler.pos = -3
!         self.assertRaises(IndexError, u"\xff0".encode, "ascii", "test.posreturn")
! 
!         # Valid positive position
!         handler.pos = 1
!         self.assertEquals(u"\xff0".encode("ascii", "test.posreturn"), "<?>0")
! 
!         # Largest valid positive position (one beyond end of input
!         handler.pos = 2
!         self.assertEquals(u"\xff0".encode("ascii", "test.posreturn"), "<?>")
! 
!         # Invalid positive position
!         handler.pos = 3
!         self.assertRaises(IndexError, u"\xff0".encode, "ascii", "test.posreturn")
! 
!         handler.pos = 0
  
          class D(dict):
              def __getitem__(self, key):
                  raise ValueError
!         for err in ("strict", "replace", "xmlcharrefreplace", "backslashreplace", "test.posreturn"):
              self.assertRaises(UnicodeError, codecs.charmap_encode, u"\xff", err, {0xff: None})
              self.assertRaises(ValueError, codecs.charmap_encode, u"\xff", err, D())