Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-10-31 13:41:24 +00:00
Code Issues Projects Releases Packages Wiki Activity

Bug #1572832: fix a bug in ISO-2022 codecs which may cause segfault

Browse source 
when encoding non-BMP unicode characters.  (Submitted by Ray Chason)
This commit is contained in:
Hye-Shik Chang 2006-10-08 13:48:34 +00:00
parent 846f73a530
commit b788346573
3 changed files with 35 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

10
Lib/test/test_multibytecodec.py
View file

@ -208,6 +208,16 @@ def test_iso2022_jp_g0(self):
e = u'\u3406'.encode(encoding) e = u'\u3406'.encode(encoding)
self.failIf(filter(lambda x: x >= '\x80', e)) self.failIf(filter(lambda x: x >= '\x80', e))
def test_bug1572832(self):
if sys.maxunicode >= 0x10000:
myunichr = unichr
else:
myunichr = lambda x: unichr(0xD7C0+(x>>10)) + unichr(0xDC00+(x&0x3FF))
for x in xrange(0x10000, 0x110000):
# Any ISO 2022 codec will cause the segfault
myunichr(x).encode('iso_2022_jp', 'ignore')
def test_main(): def test_main():
suite = unittest.TestSuite() suite = unittest.TestSuite()
suite.addTest(unittest.makeSuite(Test_MultibyteCodec)) suite.addTest(unittest.makeSuite(Test_MultibyteCodec))

3
Misc/NEWS
View file

@ -123,6 +123,9 @@ Library
Extension Modules Extension Modules
----------------- -----------------
- Bug #1572832: fix a bug in ISO-2022 codecs which may cause segfault
when encoding non-BMP unicode characters.
- Bug #1556784: allow format strings longer than 127 characters in - Bug #1556784: allow format strings longer than 127 characters in
datetime's strftime function. datetime's strftime function.

36
Modules/cjkcodecs/_codecs_iso2022.c
View file

@ -592,9 +592,11 @@ ksx1001_encoder(const ucs4_t *data, Py_ssize_t *length)
{ {
DBCHAR coded; DBCHAR coded;
assert(*length == 1); assert(*length == 1);
TRYMAP_ENC(cp949, coded, *data) if (*data < 0x10000) {
if (!(coded & 0x8000)) TRYMAP_ENC(cp949, coded, *data)
return coded; if (!(coded & 0x8000))
return coded;
}
return MAP_UNMAPPABLE; return MAP_UNMAPPABLE;
} }
@ -628,11 +630,13 @@ jisx0208_encoder(const ucs4_t *data, Py_ssize_t *length)
{ {
DBCHAR coded; DBCHAR coded;
assert(*length == 1); assert(*length == 1);
if (*data == 0xff3c) /* F/W REVERSE SOLIDUS */ if (*data < 0x10000) {
return 0x2140; if (*data == 0xff3c) /* F/W REVERSE SOLIDUS */
else TRYMAP_ENC(jisxcommon, coded, *data) { return 0x2140;
if (!(coded & 0x8000)) else TRYMAP_ENC(jisxcommon, coded, *data) {
return coded; if (!(coded & 0x8000))
return coded;
}
} }
return MAP_UNMAPPABLE; return MAP_UNMAPPABLE;
} }
@ -665,9 +669,11 @@ jisx0212_encoder(const ucs4_t *data, Py_ssize_t *length)
{ {
DBCHAR coded; DBCHAR coded;
assert(*length == 1); assert(*length == 1);
TRYMAP_ENC(jisxcommon, coded, *data) { if (*data < 0x10000) {
if (coded & 0x8000) TRYMAP_ENC(jisxcommon, coded, *data) {
return coded & 0x7fff; if (coded & 0x8000)
return coded & 0x7fff;
}
} }
return MAP_UNMAPPABLE; return MAP_UNMAPPABLE;
} }
@ -970,9 +976,11 @@ gb2312_encoder(const ucs4_t *data, Py_ssize_t *length)
{ {
DBCHAR coded; DBCHAR coded;
assert(*length == 1); assert(*length == 1);
TRYMAP_ENC(gbcommon, coded, *data) { if (*data < 0x10000) {
if (!(coded & 0x8000)) TRYMAP_ENC(gbcommon, coded, *data) {
return coded; if (!(coded & 0x8000))
return coded;
}
} }
return MAP_UNMAPPABLE; return MAP_UNMAPPABLE;
} }