Issue #13333: The UTF-7 decoder now accepts lone surrogates
(the encoder already accepts them).
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 7316abf..8680726 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -2282,21 +2282,17 @@
*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;
}
@@ -2306,8 +2302,8 @@
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) {