cpython: ddfcb0de564f (original) (raw)

Mercurial > cpython

changeset 73552:ddfcb0de564f 3.2

Issue #13333: The UTF-7 decoder now accepts lone surrogates (the encoder already accepts them). [#13333]

line wrap: on

line diff

--- a/Lib/test/test_unicode.py +++ b/Lib/test/test_unicode.py @@ -1091,10 +1091,18 @@ class UnicodeTest(string_tests.CommonTes for (x, y) in utfTests: self.assertEqual(x.encode('utf-7'), y)

•    # Unpaired surrogates not supported[](#l1.7)

•    self.assertRaises(UnicodeError, str, b'+3ADYAA-', 'utf-7')[](#l1.8)

•    # Unpaired surrogates are passed through[](#l1.9)

•    self.assertEqual('\uD801'.encode('utf-7'), b'+2AE-')[](#l1.10)

•    self.assertEqual('\uD801x'.encode('utf-7'), b'+2AE-x')[](#l1.11)

•    self.assertEqual('\uDC01'.encode('utf-7'), b'+3AE-')[](#l1.12)

•    self.assertEqual('\uDC01x'.encode('utf-7'), b'+3AE-x')[](#l1.13)

•    self.assertEqual(b'+2AE-'.decode('utf-7'), '\uD801')[](#l1.14)

•    self.assertEqual(b'+2AE-x'.decode('utf-7'), '\uD801x')[](#l1.15)

•    self.assertEqual(b'+3AE-'.decode('utf-7'), '\uDC01')[](#l1.16)

•    self.assertEqual(b'+3AE-x'.decode('utf-7'), '\uDC01x')[](#l1.17)

•    self.assertEqual(str(b'+3ADYAA-', 'utf-7', 'replace'), '\ufffd\ufffd')[](#l1.19)

•    self.assertEqual('\uD801\U000abcde'.encode('utf-7'), b'+2AHab9ze-')[](#l1.20)

•    self.assertEqual(b'+2AHab9ze-'.decode('utf-7'), '\uD801\U000abcde')[](#l1.21)

# Issue #2242: crash on some Windows/MSVC versions self.assertEqual(b'+\xc1'.decode('utf-7'), '\xc1')

--- a/Misc/NEWS +++ b/Misc/NEWS @@ -10,6 +10,9 @@ What's New in Python 3.2.3? Core and Builtins ----------------- +- Issue #13333: The UTF-7 decoder now accepts lone surrogates (the encoder

• already accepts them). +

• Issue #13342: input() used to ignore sys.stdin's and sys.stdout's unicode error handler in interactive mode (when calling into PyOS_Readline()).

--- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -2282,21 +2282,17 @@ PyObject *PyUnicode_DecodeUTF7Stateful(c *p++ = outCh; #endif surrogate = 0;

•                        continue;[](#l3.7)
                     }[](#l3.8)
                     else {[](#l3.9)

•                        *p++ = surrogate;[](#l3.10)
                         surrogate = 0;[](#l3.11)

•                        errmsg = "second surrogate missing";[](#l3.12)

•                        goto utf7Error;[](#l3.13)
                     }[](#l3.14)
                 }[](#l3.15)

•                else if (outCh >= 0xD800 && outCh <= 0xDBFF) {[](#l3.16)

•                if (outCh >= 0xD800 && outCh <= 0xDBFF) {[](#l3.17)
                     /* first surrogate */[](#l3.18)
                     surrogate = outCh;[](#l3.19)
                 }[](#l3.20)

•                else if (outCh >= 0xDC00 && outCh <= 0xDFFF) {[](#l3.21)

•                    errmsg = "unexpected second surrogate";[](#l3.22)

•                    goto utf7Error;[](#l3.23)

•                }[](#l3.24)
                 else {[](#l3.25)
                     *p++ = outCh;[](#l3.26)
                 }[](#l3.27)

@@ -2306,8 +2302,8 @@ PyObject *PyUnicode_DecodeUTF7Stateful(c inShift = 0; s++; if (surrogate) {

•                errmsg = "second surrogate missing at end of shift sequence";[](#l3.32)

•                goto utf7Error;[](#l3.33)

•                *p++ = surrogate;[](#l3.34)

•                surrogate = 0;[](#l3.35)
             }[](#l3.36)
             if (base64bits > 0) { /* left-over bits */[](#l3.37)
                 if (base64bits >= 6) {[](#l3.38)