Issue #11461: Fix the incremental UTF-16 decoder. Original patch by

Amaury Forgeot d'Arc. Added tests for partial decoding of non-BMP
characters.

diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index f620b4cde370bad50f9973c4ed05678ffe43e4d8..5baf2252596b3ce4e13f63133ccc5d4715872cfd 100644 (file)
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -281,7 +281,7 @@ class UTF32Test(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"", # first byte of BOM read
                 u"", # second byte of BOM read
@@ -303,6 +303,10 @@ class UTF32Test(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -331,7 +335,7 @@ class UTF32LETest(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"",
                 u"",
@@ -349,6 +353,10 @@ class UTF32LETest(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -371,7 +379,7 @@ class UTF32BETest(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"",
                 u"",
@@ -389,6 +397,10 @@ class UTF32BETest(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -439,7 +451,7 @@ class UTF16Test(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"", # first byte of BOM read
                 u"", # second byte of BOM read => byteorder known
@@ -451,6 +463,10 @@ class UTF16Test(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -481,7 +497,7 @@ class UTF16LETest(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"",
                 u"\x00",
@@ -491,6 +507,10 @@ class UTF16LETest(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -514,7 +534,7 @@ class UTF16BETest(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u0100\uffff",
+            u"\x00\xff\u0100\uffff\U00010000",
             [
                 u"",
                 u"\x00",
@@ -524,6 +544,10 @@ class UTF16BETest(ReadTest):
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100",
                 u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff",
+                u"\x00\xff\u0100\uffff\U00010000",
             ]
         )
 
@@ -547,7 +571,7 @@ class UTF8Test(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\x00\xff\u07ff\u0800\uffff",
+            u"\x00\xff\u07ff\u0800\uffff\U00010000",
             [
                 u"\x00",
                 u"\x00",
@@ -560,6 +584,10 @@ class UTF8Test(ReadTest):
                 u"\x00\xff\u07ff\u0800",
                 u"\x00\xff\u07ff\u0800",
                 u"\x00\xff\u07ff\u0800\uffff",
+                u"\x00\xff\u07ff\u0800\uffff",
+                u"\x00\xff\u07ff\u0800\uffff",
+                u"\x00\xff\u07ff\u0800\uffff",
+                u"\x00\xff\u07ff\u0800\uffff\U00010000",
             ]
         )
 
@@ -619,7 +647,7 @@ class UTF8SigTest(ReadTest):
 
     def test_partial(self):
         self.check_partial(
-            u"\ufeff\x00\xff\u07ff\u0800\uffff",
+            u"\ufeff\x00\xff\u07ff\u0800\uffff\U00010000",
             [
                 u"",
                 u"",
@@ -638,6 +666,10 @@ class UTF8SigTest(ReadTest):
                 u"\ufeff\x00\xff\u07ff\u0800",
                 u"\ufeff\x00\xff\u07ff\u0800",
                 u"\ufeff\x00\xff\u07ff\u0800\uffff",
+                u"\ufeff\x00\xff\u07ff\u0800\uffff",
+                u"\ufeff\x00\xff\u07ff\u0800\uffff",
+                u"\ufeff\x00\xff\u07ff\u0800\uffff",
+                u"\ufeff\x00\xff\u07ff\u0800\uffff\U00010000",
             ]
         )
 

diff --git a/Misc/NEWS b/Misc/NEWS
index dd26444bd954fbde88c94e39aa11601866731945..8f5a1e34df820fd4be90543db39e70c661e7fa37 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -9,6 +9,9 @@ What's New in Python 2.7.4
 Core and Builtins
 -----------------
 
+- Issue #11461: Fix the incremental UTF-16 decoder. Original patch by
+  Amaury Forgeot d'Arc.
+
 - Issue #16367: Fix FileIO.readall() on Windows for files larger than 2 GB.
 
 - Issue #15516: Fix a bug in PyString_FromFormat where it failed to properly

diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 7713b5497e1d7f920ca43565084107bf8a5f3600..1c6e55d1bf8437af3c41b553fce7b9d505a69567 100644 (file)
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -2565,8 +2565,11 @@ PyUnicode_DecodeUTF16Stateful(const char *s,
 
         /* UTF-16 code pair: */
         if (e - q < 2) {
+            q -= 2;
+            if (consumed)
+                break;
             errmsg = "unexpected end of data";
-            startinpos = (((const char *)q)-2)-starts;
+            startinpos = ((const char *)q)-starts;
             endinpos = ((const char *)e)-starts;
             goto utf16Error;
         }