diff options
| author | Antoine Pitrou <solipsis@pitrou.net> | 2011-11-15 00:49:40 (GMT) |
|---|---|---|
| committer | Antoine Pitrou <solipsis@pitrou.net> | 2011-11-15 00:49:40 (GMT) |
| commit | 30402549de68a5303a5e2995dca7375d3d17966f (patch) | |
| tree | d58a964054da32b13020849f51191a6264f6c7e3 /Objects | |
| parent | 16ed86831bc44cc54abf6af634b6e9be91d8a3b3 (diff) | |
| download | cpython-30402549de68a5303a5e2995dca7375d3d17966f.zip cpython-30402549de68a5303a5e2995dca7375d3d17966f.tar.gz cpython-30402549de68a5303a5e2995dca7375d3d17966f.tar.bz2 | |
Issue #13333: The UTF-7 decoder now accepts lone surrogates
(the encoder already accepts them).
Diffstat (limited to 'Objects')
| -rw-r--r-- | Objects/unicodeobject.c | 14 |
1 files changed, 5 insertions, 9 deletions
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c index 4d6864d..5ce879d 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -1628,21 +1628,17 @@ PyObject *PyUnicode_DecodeUTF7Stateful(const char *s, *p++ = outCh; #endif surrogate = 0; + continue; } else { + *p++ = surrogate; surrogate = 0; - errmsg = "second surrogate missing"; - goto utf7Error; } } - else if (outCh >= 0xD800 && outCh <= 0xDBFF) { + if (outCh >= 0xD800 && outCh <= 0xDBFF) { /* first surrogate */ surrogate = outCh; } - else if (outCh >= 0xDC00 && outCh <= 0xDFFF) { - errmsg = "unexpected second surrogate"; - goto utf7Error; - } else { *p++ = outCh; } @@ -1652,8 +1648,8 @@ PyObject *PyUnicode_DecodeUTF7Stateful(const char *s, inShift = 0; s++; if (surrogate) { - errmsg = "second surrogate missing at end of shift sequence"; - goto utf7Error; + *p++ = surrogate; + surrogate = 0; } if (base64bits > 0) { /* left-over bits */ if (base64bits >= 6) { |