cpython: 0f1c2e2b6bc2 (original) (raw)

Mercurial > cpython

changeset 82051:0f1c2e2b6bc2 3.2

Issue #17043: The unicode-internal decoder no longer read past the end of input buffer. [#17043]

Serhiy Storchaka storchaka@gmail.com
date	Thu, 07 Feb 2013 16:23:21 +0200
parents	148e6ebfe854
children	fec2976c8503 55a89352e220
files	Misc/NEWS Objects/unicodeobject.c
diffstat	2 files changed, 27 insertions(+), 27 deletions(-)[+] [-] Misc/NEWS 3 Objects/unicodeobject.c 51

line wrap: on

line diff

--- a/Misc/NEWS +++ b/Misc/NEWS @@ -10,6 +10,9 @@ What's New in Python 3.2.4 Core and Builtins ----------------- +- Issue #17043: The unicode-internal decoder no longer read past the end of

input buffer. +

Issue #16979: Fix error handling bugs in the unicode-escape-decode decoder.
Issue #10156: In the interpreter's initialization phase, unicode globals

--- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -4392,37 +4392,34 @@ PyObject *_PyUnicode_DecodeUnicodeIntern end = s + size; while (s < end) {

   if (end-s < Py_UNICODE_SIZE) {[](#l2.7)

```
       endinpos = end-starts;[](#l2.8)
```

       reason = "truncated input";[](#l2.9)

```
       goto error;[](#l2.10)
```

   }[](#l2.11)
   memcpy(p, s, sizeof(Py_UNICODE));[](#l2.12)

+#ifdef Py_UNICODE_WIDE /* We have to sanity check the raw data, otherwise doom looms for some malformed UCS-4 data. */

```
   if ([](#l2.16)
```

-#ifdef Py_UNICODE_WIDE

       *p > unimax || *p < 0 ||[](#l2.18)

   if (*p > unimax || *p < 0) {[](#l2.19)

       endinpos = s - starts + Py_UNICODE_SIZE;[](#l2.20)

       reason = "illegal code point (> 0x10FFFF)";[](#l2.21)

```
       goto error;[](#l2.22)
```
```
   }[](#l2.23)
```

#endif

       end-s < Py_UNICODE_SIZE[](#l2.25)

```
       )[](#l2.26)
```
```
   {[](#l2.27)
```

       startinpos = s - starts;[](#l2.28)

       if (end-s < Py_UNICODE_SIZE) {[](#l2.29)

           endinpos = end-starts;[](#l2.30)

           reason = "truncated input";[](#l2.31)

```
       }[](#l2.32)
```
```
       else {[](#l2.33)
```

           endinpos = s - starts + Py_UNICODE_SIZE;[](#l2.34)

           reason = "illegal code point (> 0x10FFFF)";[](#l2.35)

```
       }[](#l2.36)
```

       outpos = p - PyUnicode_AS_UNICODE(v);[](#l2.37)

       if (unicode_decode_call_errorhandler([](#l2.38)

               errors, &errorHandler,[](#l2.39)

               "unicode_internal", reason,[](#l2.40)

               &starts, &end, &startinpos, &endinpos, &exc, &s,[](#l2.41)

               &v, &outpos, &p)) {[](#l2.42)

```
           goto onError;[](#l2.43)
```
```
       }[](#l2.44)
```
```
   }[](#l2.45)
```
```
   else {[](#l2.46)
```
```
       p++;[](#l2.47)
```
```
       s += Py_UNICODE_SIZE;[](#l2.48)
```

```
   p++;[](#l2.49)
```
```
   s += Py_UNICODE_SIZE;[](#l2.50)
```
```
   continue;[](#l2.51)
```

error:
```
   startinpos = s - starts;[](#l2.54)
```

   outpos = p - PyUnicode_AS_UNICODE(v);[](#l2.55)

   if (unicode_decode_call_errorhandler([](#l2.56)

           errors, &errorHandler,[](#l2.57)

           "unicode_internal", reason,[](#l2.58)

           &starts, &end, &startinpos, &endinpos, &exc, &s,[](#l2.59)

           &v, &outpos, &p)) {[](#l2.60)

       goto onError;[](#l2.61)
   }[](#l2.62)

}