Merge part of the trunk changes into the p3yk branch. This merges from 43030

(branch-creation time) up to 43067. 43068 and 43069 contain a little
swapping action between re.py and sre.py, and this mightily confuses svn
merge, so later changes are going in separately.

This merge should break no additional tests.

The last-merged revision is going in a 'last_merge' property on '.' (the
branch directory.) Arbitrarily chosen, really; if there's a BCP for this, I
couldn't find it, but we can easily change it afterwards ;)

1 files changed, 50 insertions, 2 deletions

diff --git a/Lib/encodings/utf_16.py b/Lib/encodings/utf_16.py
index 95abb05..eff08f3 100644
--- a/Lib/encodings/utf_16.py
+++ b/Lib/encodings/utf_16.py
@@ -15,6 +15,47 @@ encode = codecs.utf_16_encode
 def decode(input, errors='strict'):
     return codecs.utf_16_decode(input, errors, True)
 
+class IncrementalEncoder(codecs.IncrementalEncoder):
+    def __init__(self, errors='strict'):
+        codecs.IncrementalEncoder.__init__(self, errors)
+        self.encoder = None
+
+    def encode(self, input, final=False):
+        if self.encoder is None:
+            result = codecs.utf_16_encode(input, self.errors)[0]
+            if sys.byteorder == 'little':
+                self.encoder = codecs.utf_16_le_encode
+            else:
+                self.encoder = codecs.utf_16_be_encode
+            return result
+        return self.encoder(input, self.errors)[0]
+
+    def reset(self):
+        codecs.IncrementalEncoder.reset(self)
+        self.encoder = None
+
+class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
+    def __init__(self, errors='strict'):
+        codecs.BufferedIncrementalDecoder.__init__(self, errors)
+        self.decoder = None
+
+    def _buffer_decode(self, input, errors, final):
+        if self.decoder is None:
+            (output, consumed, byteorder) = \
+                codecs.utf_16_ex_decode(input, errors, 0, final)
+            if byteorder == -1:
+                self.decoder = codecs.utf_16_le_decode
+            elif byteorder == 1:
+                self.decoder = codecs.utf_16_be_decode
+            elif consumed >= 2:
+                raise UnicodeError("UTF-16 stream does not start with BOM")
+            return (output, consumed)
+        return self.decoder(input, self.errors, final)
+
+    def reset(self):
+        codecs.BufferedIncrementalDecoder.reset(self)
+        self.decoder = None
+
 class StreamWriter(codecs.StreamWriter):
     def __init__(self, stream, errors='strict'):
         self.bom_written = False
@@ -52,5 +93,12 @@ class StreamReader(codecs.StreamReader):
 ### encodings module API
 
 def getregentry():
-
-    return (encode,decode,StreamReader,StreamWriter)
+    return codecs.CodecInfo(
+        name='utf-16',
+        encode=encode,
+        decode=decode,
+        incrementalencoder=IncrementalEncoder,
+        incrementaldecoder=IncrementalDecoder,
+        streamreader=StreamReader,
+        streamwriter=StreamWriter,
+    )