mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-07-07 19:35:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 8

gh-126004: Fix positions handling in codecs.xmlcharrefreplace_errors (#127675)

Browse source 
This fixes how `PyCodec_XMLCharRefReplaceErrors` handles the `start` and `end`
attributes of `UnicodeError` objects via the `_PyUnicodeError_GetParams` helper.
This commit is contained in:
Bénédikt Tran 2025-01-23 11:42:38 +01:00 • committed by GitHub
parent a10f99375e
commit 70dcc847df
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 111 additions and 94 deletions
Download patch file Download diff file Expand all files Collapse all files

195
Python/codecs.c
View file

@ -755,100 +755,113 @@ PyObject *PyCodec_ReplaceErrors(PyObject *exc)
PyObject *PyCodec_XMLCharRefReplaceErrors(PyObject *exc)
{
if (PyObject_TypeCheck(exc, (PyTypeObject *)PyExc_UnicodeEncodeError)) {
PyObject *restuple;
PyObject *object;
Py_ssize_t i;
Py_ssize_t start;
Py_ssize_t end;
PyObject *res;
Py_UCS1 *outp;
Py_ssize_t ressize;
Py_UCS4 ch;
if (PyUnicodeEncodeError_GetStart(exc, &start))
return NULL;
if (PyUnicodeEncodeError_GetEnd(exc, &end))
return NULL;
if (!(object = PyUnicodeEncodeError_GetObject(exc)))
return NULL;
if (end - start > PY_SSIZE_T_MAX / (2+7+1))
end = start + PY_SSIZE_T_MAX / (2+7+1);
for (i = start, ressize = 0; i < end; ++i) {
/* object is guaranteed to be "ready" */
ch = PyUnicode_READ_CHAR(object, i);
if (ch<10)
ressize += 2+1+1;
else if (ch<100)
ressize += 2+2+1;
else if (ch<1000)
ressize += 2+3+1;
else if (ch<10000)
ressize += 2+4+1;
else if (ch<100000)
ressize += 2+5+1;
else if (ch<1000000)
ressize += 2+6+1;
else
ressize += 2+7+1;
}
/* allocate replacement */
res = PyUnicode_New(ressize, 127);
if (res == NULL) {
Py_DECREF(object);
return NULL;
}
outp = PyUnicode_1BYTE_DATA(res);
/* generate replacement */
for (i = start; i < end; ++i) {
int digits;
int base;
ch = PyUnicode_READ_CHAR(object, i);
*outp++ = '&';
*outp++ = '#';
if (ch<10) {
digits = 1;
base = 1;
}
else if (ch<100) {
digits = 2;
base = 10;
}
else if (ch<1000) {
digits = 3;
base = 100;
}
else if (ch<10000) {
digits = 4;
base = 1000;
}
else if (ch<100000) {
digits = 5;
base = 10000;
}
else if (ch<1000000) {
digits = 6;
base = 100000;
}
else {
digits = 7;
base = 1000000;
}
while (digits-->0) {
*outp++ = '0' + ch/base;
ch %= base;
base /= 10;
}
*outp++ = ';';
}
assert(_PyUnicode_CheckConsistency(res, 1));
restuple = Py_BuildValue("(Nn)", res, end);
Py_DECREF(object);
return restuple;
}
else {
if (!PyObject_TypeCheck(exc, (PyTypeObject *)PyExc_UnicodeEncodeError)) {
wrong_exception_type(exc);
return NULL;
}
PyObject *obj;
Py_ssize_t objlen, start, end, slen;
if (_PyUnicodeError_GetParams(exc,
&obj, &objlen,
&start, &end, &slen, false) < 0)
{
return NULL;
}
// The number of characters that each character 'ch' contributes
// in the result is 2 + k + 1, where k = min{t >= 1 | 10^t > ch}
// and will be formatted as "&#" + DIGITS + ";". Since the Unicode
// range is below 10^7, each "block" requires at most 2 + 7 + 1
// characters.
if (slen > PY_SSIZE_T_MAX / (2 + 7 + 1)) {
end = start + PY_SSIZE_T_MAX / (2 + 7 + 1);
end = Py_MIN(end, objlen);
slen = Py_MAX(0, end - start);
}
Py_ssize_t ressize = 0;
for (Py_ssize_t i = start; i < end; ++i) {
/* object is guaranteed to be "ready" */
Py_UCS4 ch = PyUnicode_READ_CHAR(obj, i);
if (ch < 10) {
ressize += 2 + 1 + 1;
}
else if (ch < 100) {
ressize += 2 + 2 + 1;
}
else if (ch < 1000) {
ressize += 2 + 3 + 1;
}
else if (ch < 10000) {
ressize += 2 + 4 + 1;
}
else if (ch < 100000) {
ressize += 2 + 5 + 1;
}
else if (ch < 1000000) {
ressize += 2 + 6 + 1;
}
else {
assert(ch < 10000000);
ressize += 2 + 7 + 1;
}
}
/* allocate replacement */
PyObject *res = PyUnicode_New(ressize, 127);
if (res == NULL) {
Py_DECREF(obj);
return NULL;
}
Py_UCS1 *outp = PyUnicode_1BYTE_DATA(res);
/* generate replacement */
for (Py_ssize_t i = start; i < end; ++i) {
int digits, base;
Py_UCS4 ch = PyUnicode_READ_CHAR(obj, i);
if (ch < 10) {
digits = 1;
base = 1;
}
else if (ch < 100) {
digits = 2;
base = 10;
}
else if (ch < 1000) {
digits = 3;
base = 100;
}
else if (ch < 10000) {
digits = 4;
base = 1000;
}
else if (ch < 100000) {
digits = 5;
base = 10000;
}
else if (ch < 1000000) {
digits = 6;
base = 100000;
}
else {
assert(ch < 10000000);
digits = 7;
base = 1000000;
}
*outp++ = '&';
*outp++ = '#';
while (digits-- > 0) {
assert(base >= 1);
*outp++ = '0' + ch / base;
ch %= base;
base /= 10;
}
*outp++ = ';';
}
assert(_PyUnicode_CheckConsistency(res, 1));
PyObject *restuple = Py_BuildValue("(Nn)", res, end);
Py_DECREF(obj);
return restuple;
}
PyObject *PyCodec_BackslashReplaceErrors(PyObject *exc)