#include "Python.h"
#include "osdefs.h"
#include
#ifdef MS_WINDOWS
# include
#endif
#ifdef HAVE_LANGINFO_H
#include
#endif
#ifdef __APPLE__
extern wchar_t* _Py_DecodeUTF8_surrogateescape(const char *s, Py_ssize_t size);
#endif
PyObject *
_Py_device_encoding(int fd)
{
#if defined(MS_WINDOWS) || defined(MS_WIN64)
UINT cp;
#endif
if (!_PyVerify_fd(fd) || !isatty(fd)) {
Py_RETURN_NONE;
}
#if defined(MS_WINDOWS) || defined(MS_WIN64)
if (fd == 0)
cp = GetConsoleCP();
else if (fd == 1 || fd == 2)
cp = GetConsoleOutputCP();
else
cp = 0;
/* GetConsoleCP() and GetConsoleOutputCP() return 0 if the application
has no console */
if (cp != 0)
return PyUnicode_FromFormat("cp%u", (unsigned int)cp);
#elif defined(CODESET)
{
char *codeset = nl_langinfo(CODESET);
if (codeset != NULL && codeset[0] != 0)
return PyUnicode_FromString(codeset);
}
#endif
Py_RETURN_NONE;
}
#if !defined(__APPLE__) && !defined(MS_WINDOWS)
extern int _Py_normalize_encoding(const char *, char *, size_t);
/* Workaround FreeBSD and OpenIndiana locale encoding issue with the C locale.
On these operating systems, nl_langinfo(CODESET) announces an alias of the
ASCII encoding, whereas mbstowcs() and wcstombs() functions use the
ISO-8859-1 encoding. The problem is that os.fsencode() and os.fsdecode() use
locale.getpreferredencoding() codec. For example, if command line arguments
are decoded by mbstowcs() and encoded back by os.fsencode(), we get a
UnicodeEncodeError instead of retrieving the original byte string.
The workaround is enabled if setlocale(LC_CTYPE, NULL) returns "C",
nl_langinfo(CODESET) announces "ascii" (or an alias to ASCII), and at least
one byte in range 0x80-0xff can be decoded from the locale encoding. The
workaround is also enabled on error, for example if getting the locale
failed.
Values of locale_is_ascii:
1: the workaround is used: _Py_wchar2char() uses
encode_ascii_surrogateescape() and _Py_char2wchar() uses
decode_ascii_surrogateescape()
0: the workaround is not used: _Py_wchar2char() uses wcstombs() and
_Py_char2wchar() uses mbstowcs()
-1: unknown, need to call check_force_ascii() to get the value
*/
static int force_ascii = -1;
static int
check_force_ascii(void)
{
char *loc;
#if defined(HAVE_LANGINFO_H) && defined(CODESET)
char *codeset, **alias;
char encoding[100];
int is_ascii;
unsigned int i;
char* ascii_aliases[] = {
"ascii",
"646",
"ansi-x3.4-1968",
"ansi-x3-4-1968",
"ansi-x3.4-1986",
"cp367",
"csascii",
"ibm367",
"iso646-us",
"iso-646.irv-1991",
"iso-ir-6",
"us",
"us-ascii",
NULL
};
#endif
loc = setlocale(LC_CTYPE, NULL);
if (loc == NULL)
goto error;
if (strcmp(loc, "C") != 0) {
/* the LC_CTYPE locale is different than C */
return 0;
}
#if defined(HAVE_LANGINFO_H) && defined(CODESET)
codeset = nl_langinfo(CODESET);
if (!codeset || codeset[0] == '\0') {
/* CODESET is not set or empty */
goto error;
}
if (!_Py_normalize_encoding(codeset, encoding, sizeof(encoding)))
goto error;
is_ascii = 0;
for (alias=ascii_aliases; *alias != NULL; alias++) {
if (strcmp(encoding, *alias) == 0) {
is_ascii = 1;
break;
}
}
if (!is_ascii) {
/* nl_langinfo(CODESET) is not "ascii" or an alias of ASCII */
return 0;
}
for (i=0x80; i= 0xd800 && *out 128. This will still roundtrip
correctly in the locale's charset, which must be an ASCII superset. */
res = decode_ascii_surrogateescape(arg, size);
if (res == NULL)
goto oom;
#endif /* HAVE_MBRTOWC */
return res;
oom:
if (size != NULL)
*size = (size_t)-1;
return NULL;
#endif /* __APPLE__ */
}
/* Encode a (wide) character string to the locale encoding with the
surrogateescape error handler (characters in range U+DC80..U+DCFF are
converted to bytes 0x80..0xFF).
This function is the reverse of _Py_char2wchar().
Return a pointer to a newly allocated byte string (use PyMem_Free() to free
the memory), or NULL on encoding or memory allocation error.
If error_pos is not NULL: *error_pos is the index of the invalid character
on encoding error, or (size_t)-1 otherwise. */
char*
_Py_wchar2char(const wchar_t *text, size_t *error_pos)
{
#ifdef __APPLE__
Py_ssize_t len;
PyObject *unicode, *bytes = NULL;
char *cpath;
unicode = PyUnicode_FromWideChar(text, wcslen(text));
if (unicode == NULL)
return NULL;
bytes = _PyUnicode_AsUTF8String(unicode, "surrogateescape");
Py_DECREF(unicode);
if (bytes == NULL) {
PyErr_Clear();
if (error_pos != NULL)
*error_pos = (size_t)-1;
return NULL;
}
len = PyBytes_GET_SIZE(bytes);
cpath = PyMem_Malloc(len+1);
if (cpath == NULL) {
PyErr_Clear();
Py_DECREF(bytes);
if (error_pos != NULL)
*error_pos = (size_t)-1;
return NULL;
}
memcpy(cpath, PyBytes_AsString(bytes), len + 1);
Py_DECREF(bytes);
return cpath;
#else /* __APPLE__ */
const size_t len = wcslen(text);
char *result = NULL, *bytes = NULL;
size_t i, size, converted;
wchar_t c, buf[2];
#ifndef MS_WINDOWS
if (force_ascii == -1)
force_ascii = check_force_ascii();
if (force_ascii)
return encode_ascii_surrogateescape(text, error_pos);
#endif
/* The function works in two steps:
1. compute the length of the output buffer in bytes (size)
2. outputs the bytes */
size = 0;
buf[1] = 0;
while (1) {
for (i=0; i < len; i++) {
c = text[i];
if (c >= 0xdc80 && c st_mode = wstatbuf.st_mode;
return err;
#else
int ret;
PyObject *bytes = PyUnicode_EncodeFSDefault(path);
if (bytes == NULL)
return -2;
ret = stat(PyBytes_AS_STRING(bytes), statbuf);
Py_DECREF(bytes);
return ret;
#endif
}
#endif
/* Open a file. Use _wfopen() on Windows, encode the path to the locale
encoding and use fopen() otherwise. */
FILE *
_Py_wfopen(const wchar_t *path, const wchar_t *mode)
{
#ifndef MS_WINDOWS
FILE *f;
char *cpath;
char cmode[10];
size_t r;
r = wcstombs(cmode, mode, 10);
if (r == (size_t)-1 || r >= 10) {
errno = EINVAL;
return NULL;
}
cpath = _Py_wchar2char(path, NULL);
if (cpath == NULL)
return NULL;
f = fopen(cpath, cmode);
PyMem_Free(cpath);
return f;
#else
return _wfopen(path, mode);
#endif
}
/* Call _wfopen() on Windows, or encode the path to the filesystem encoding and
call fopen() otherwise.
Return the new file object on success, or NULL if the file cannot be open or
(if PyErr_Occurred()) on unicode error */
FILE*
_Py_fopen(PyObject *path, const char *mode)
{
#ifdef MS_WINDOWS
wchar_t *wpath;
wchar_t wmode[10];
int usize;
if (!PyUnicode_Check(path)) {
PyErr_Format(PyExc_TypeError,
"str file path expected under Windows, got %R",
Py_TYPE(path));
return NULL;
}
wpath = PyUnicode_AsUnicode(path);
if (wpath == NULL)
return NULL;
usize = MultiByteToWideChar(CP_ACP, 0, mode, -1, wmode, sizeof(wmode));
if (usize == 0)
return NULL;
return _wfopen(wpath, wmode);
#else
FILE *f;
PyObject *bytes;
if (!PyUnicode_FSConverter(path, &bytes))
return NULL;
f = fopen(PyBytes_AS_STRING(bytes), mode);
Py_DECREF(bytes);
return f;
#endif
}
#ifdef HAVE_READLINK
/* Read value of symbolic link. Encode the path to the locale encoding, decode
the result from the locale encoding. Return -1 on error. */
int
_Py_wreadlink(const wchar_t *path, wchar_t *buf, size_t bufsiz)
{
char *cpath;
char cbuf[MAXPATHLEN];
wchar_t *wbuf;
int res;
size_t r1;
cpath = _Py_wchar2char(path, NULL);
if (cpath == NULL) {
errno = EINVAL;
return -1;
}
res = (int)readlink(cpath, cbuf, Py_ARRAY_LENGTH(cbuf));
PyMem_Free(cpath);
if (res == -1)
return -1;
if (res == Py_ARRAY_LENGTH(cbuf)) {
errno = EINVAL;
return -1;
}
cbuf[res] = '\0'; /* buf will be null terminated */
wbuf = _Py_char2wchar(cbuf, &r1);
if (wbuf == NULL) {
errno = EINVAL;
return -1;
}
if (bufsiz