/*
Unicode implementation based on original code by Fredrik Lundh,
modified by Marc-Andre Lemburg .
Major speed upgrades to the method implementations at the Reykjavik
NeedForSpeed sprint, by Fredrik Lundh and Andrew Dalke.
Copyright (c) Corporation for National Research Initiatives.
--------------------------------------------------------------------
The original string type implementation is:
Copyright (c) 1999 by Secret Labs AB
Copyright (c) 1999 by Fredrik Lundh
By obtaining, using, and/or copying this software and/or its
associated documentation, you agree that you have read, understood,
and will comply with the following terms and conditions:
Permission to use, copy, modify, and distribute this software and its
associated documentation for any purpose and without fee is hereby
granted, provided that the above copyright notice appears in all
copies, and that both that copyright notice and this permission notice
appear in supporting documentation, and that the name of Secret Labs
AB or the author not be used in advertising or publicity pertaining to
distribution of the software without specific, written prior
permission.
SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO
THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
FITNESS. IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR
ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
--------------------------------------------------------------------
*/
#include "Python.h"
#include "pycore_abstract.h" // _PyIndex_Check()
#include "pycore_bytes_methods.h" // _Py_bytes_lower()
#include "pycore_bytesobject.h" // _PyBytes_RepeatBuffer()
#include "pycore_ceval.h" // _PyEval_GetBuiltin()
#include "pycore_codecs.h" // _PyCodec_Lookup()
#include "pycore_critical_section.h" // Py_*_CRITICAL_SECTION_SEQUENCE_FAST
#include "pycore_format.h" // F_LJUST
#include "pycore_initconfig.h" // _PyStatus_OK()
#include "pycore_interp.h" // PyInterpreterState.fs_codec
#include "pycore_long.h" // _PyLong_FormatWriter()
#include "pycore_object.h" // _PyObject_GC_TRACK(), _Py_FatalRefcountError()
#include "pycore_pathconfig.h" // _Py_DumpPathConfig()
#include "pycore_pyerrors.h" // _PyUnicodeTranslateError_Create()
#include "pycore_pyhash.h" // _Py_HashSecret_t
#include "pycore_pylifecycle.h" // _Py_SetFileSystemEncoding()
#include "pycore_pystate.h" // _PyInterpreterState_GET()
#include "pycore_ucnhash.h" // _PyUnicode_Name_CAPI
#include "pycore_unicodectype.h" // _PyUnicode_IsXidStart
#include "pycore_unicodeobject.h" // struct _Py_unicode_state
#include "pycore_unicodeobject_generated.h" // _PyUnicode_InitStaticStrings()
#include "stringlib/eq.h" // unicode_eq()
#include // ptrdiff_t
#ifdef MS_WINDOWS
#include
#endif
#ifdef HAVE_ICONV
#include // iconv_open()
#endif
#ifdef HAVE_NON_UNICODE_WCHAR_T_REPRESENTATION
# include "pycore_fileutils.h" // _Py_LocaleUsesNonUnicodeWchar()
#endif
/* Uncomment to display statistics on interned strings at exit
in _PyUnicode_ClearInterned(). */
/* #define INTERNED_STATS 1 */
/*[clinic input]
class str "PyObject *" "&PyUnicode_Type"
[clinic start generated code]*/
/*[clinic end generated code: output=da39a3ee5e6b4b0d input=4884c934de622cf6]*/
/*[python input]
class Py_UCS4_converter(CConverter):
type = 'Py_UCS4'
converter = 'convert_uc'
def c_default_init(self):
import libclinic
self.c_default = libclinic.c_unichar_repr(self.default)
[python start generated code]*/
/*[python end generated code: output=da39a3ee5e6b4b0d input=22f057b68fd9a65a]*/
/* --- Globals ------------------------------------------------------------
NOTE: In the interpreter's initialization phase, some globals are currently
initialized dynamically as needed. In the process Unicode objects may
be created before the Unicode type is ready.
*/
#define MAX_UNICODE _Py_MAX_UNICODE
#define ensure_unicode _PyUnicode_EnsureUnicode
#ifdef Py_DEBUG
# define _PyUnicode_CHECK(op) _PyUnicode_CheckConsistency(op, 0)
#else
# define _PyUnicode_CHECK(op) PyUnicode_Check(op)
#endif
static inline char* _PyUnicode_UTF8(PyObject *op)
{
return FT_ATOMIC_LOAD_PTR_ACQUIRE(_PyCompactUnicodeObject_CAST(op)->utf8);
}
static inline char* PyUnicode_UTF8(PyObject *op)
{
assert(_PyUnicode_CHECK(op));
if (PyUnicode_IS_COMPACT_ASCII(op)) {
return ((char*)(_PyASCIIObject_CAST(op) + 1));
}
else {
return _PyUnicode_UTF8(op);
}
}
static inline void PyUnicode_SET_UTF8(PyObject *op, char *utf8)
{
FT_ATOMIC_STORE_PTR_RELEASE(_PyCompactUnicodeObject_CAST(op)->utf8, utf8);
}
static inline Py_ssize_t PyUnicode_UTF8_LENGTH(PyObject *op)
{
assert(_PyUnicode_CHECK(op));
if (PyUnicode_IS_COMPACT_ASCII(op)) {
return _PyASCIIObject_CAST(op)->length;
}
else {
return _PyCompactUnicodeObject_CAST(op)->utf8_length;
}
}
static inline void PyUnicode_SET_UTF8_LENGTH(PyObject *op, Py_ssize_t length)
{
_PyCompactUnicodeObject_CAST(op)->utf8_length = length;
}
#define _PyUnicode_LENGTH(op) \
(_PyASCIIObject_CAST(op)->length)
#define _PyUnicode_STATE(op) \
(_PyASCIIObject_CAST(op)->state)
#define _PyUnicode_HASH(op) \
(_PyASCIIObject_CAST(op)->hash)
#define PyUnicode_HASH PyUnstable_Unicode_GET_CACHED_HASH
static inline void PyUnicode_SET_HASH(PyObject *op, Py_hash_t hash)
{
FT_ATOMIC_STORE_SSIZE_RELAXED(_PyASCIIObject_CAST(op)->hash, hash);
}
#define _PyUnicode_DATA_ANY(op) \
(_PyUnicodeObject_CAST(op)->data.any)
static inline int _PyUnicode_SHARE_UTF8(PyObject *op)
{
assert(_PyUnicode_CHECK(op));
assert(!PyUnicode_IS_COMPACT_ASCII(op));
return (_PyUnicode_UTF8(op) == PyUnicode_DATA(op));
}
/* true if the Unicode object has an allocated UTF-8 memory block
(not shared with other data) */
static inline int _PyUnicode_HAS_UTF8_MEMORY(PyObject *op)
{
return (!PyUnicode_IS_COMPACT_ASCII(op)
&& _PyUnicode_UTF8(op) != NULL
&& _PyUnicode_UTF8(op) != PyUnicode_DATA(op));
}
#define LATIN1 _Py_LATIN1_CHR
/* Forward declaration */
static PyObject *
unicode_encode_utf8(PyObject *unicode, _Py_error_handler error_handler,
const char *errors);
static PyObject *
unicode_decode_utf8(const char *s, Py_ssize_t size,
_Py_error_handler error_handler, const char *errors,
Py_ssize_t *consumed);
#ifdef Py_DEBUG
static inline int unicode_is_finalizing(void);
static int unicode_is_singleton(PyObject *unicode);
#endif
// Return a reference to the immortal empty string singleton.
PyObject*
_PyUnicode_GetEmpty(void)
{
_Py_DECLARE_STR(empty, "");
return &_Py_STR(empty);
}
/* This dictionary holds per-interpreter interned strings.
* See InternalDocs/string_interning.md for details.
*/
static inline PyObject *get_interned_dict(PyInterpreterState *interp)
{
return _Py_INTERP_CACHED_OBJECT(interp, interned_strings);
}
/* This hashtable holds statically allocated interned strings.
* See InternalDocs/string_interning.md for details.
*/
#define INTERNED_STRINGS _PyRuntime.cached_objects.interned_strings
/* Get number of all interned strings for the current interpreter. */
Py_ssize_t
_PyUnicode_InternedSize(void)
{
PyObject *dict = get_interned_dict(_PyInterpreterState_GET());
return _Py_hashtable_len(INTERNED_STRINGS) + PyDict_GET_SIZE(dict);
}
/* Get number of immortal interned strings for the current interpreter. */
Py_ssize_t
_PyUnicode_InternedSize_Immortal(void)
{
PyObject *dict = get_interned_dict(_PyInterpreterState_GET());
PyObject *key, *value;
Py_ssize_t pos = 0;
Py_ssize_t count = 0;
// It's tempting to keep a count and avoid a loop here. But, this function
// is intended for refleak tests. It spends extra work to report the true
// value, to help detect bugs in optimizations.
while (PyDict_Next(dict, &pos, &key, &value)) {
assert(PyUnicode_CHECK_INTERNED(key) != SSTATE_INTERNED_IMMORTAL_STATIC);
if (PyUnicode_CHECK_INTERNED(key) == SSTATE_INTERNED_IMMORTAL) {
count++;
}
}
return _Py_hashtable_len(INTERNED_STRINGS) + count;
}
static Py_hash_t unicode_hash(PyObject *);
static Py_uhash_t
hashtable_unicode_hash(const void *key)
{
return unicode_hash((PyObject *)key);
}
static int
hashtable_unicode_compare(const void *key1, const void *key2)
{
PyObject *obj1 = (PyObject *)key1;
PyObject *obj2 = (PyObject *)key2;
if (obj1 != NULL && obj2 != NULL) {
return unicode_eq(obj1, obj2);
}
else {
return obj1 == obj2;
}
}
/* Return true if this interpreter should share the main interpreter's
intern_dict. That's important for interpreters which load basic
single-phase init extension modules (m_size == -1). There could be interned
immortal strings that are shared between interpreters, due to the
PyDict_Update(mdict, m_copy) call in import_find_extension().
It's not safe to deallocate those strings until all interpreters that
potentially use them are freed. By storing them in the main interpreter, we
ensure they get freed after all other interpreters are freed.
*/
static bool
has_shared_intern_dict(PyInterpreterState *interp)
{
PyInterpreterState *main_interp = _PyInterpreterState_Main();
return interp != main_interp && interp->feature_flags & Py_RTFLAGS_USE_MAIN_OBMALLOC;
}
static int
init_interned_dict(PyInterpreterState *interp)
{
assert(get_interned_dict(interp) == NULL);
PyObject *interned;
if (has_shared_intern_dict(interp)) {
interned = get_interned_dict(_PyInterpreterState_Main());
Py_INCREF(interned);
}
else {
interned = PyDict_New();
if (interned == NULL) {
return -1;
}
}
_Py_INTERP_CACHED_OBJECT(interp, interned_strings) = interned;
return 0;
}
static void
clear_interned_dict(PyInterpreterState *interp)
{
PyObject *interned = get_interned_dict(interp);
if (interned != NULL) {
if (!has_shared_intern_dict(interp)) {
// only clear if the dict belongs to this interpreter
PyDict_Clear(interned);
}
Py_DECREF(interned);
_Py_INTERP_CACHED_OBJECT(interp, interned_strings) = NULL;
}
}
static PyStatus
init_global_interned_strings(PyInterpreterState *interp)
{
assert(INTERNED_STRINGS == NULL);
_Py_hashtable_allocator_t hashtable_alloc = {PyMem_RawMalloc, PyMem_RawFree};
INTERNED_STRINGS = _Py_hashtable_new_full(
hashtable_unicode_hash,
hashtable_unicode_compare,
// Objects stored here are immortal and statically allocated,
// so we don't need key_destroy_func & value_destroy_func:
NULL,
NULL,
&hashtable_alloc
);
if (INTERNED_STRINGS == NULL) {
PyErr_Clear();
return _PyStatus_ERR("failed to create global interned dict");
}
/* Intern statically allocated string identifiers, deepfreeze strings,
* and one-byte latin-1 strings.
* This must be done before any module initialization so that statically
* allocated string identifiers are used instead of heap allocated strings.
* Deepfreeze uses the interned identifiers if present to save space
* else generates them and they are interned to speed up dict lookups.
*/
_PyUnicode_InitStaticStrings(interp);
for (int i = 0; i < 256; i++) {
PyObject *s = LATIN1(i);
_PyUnicode_InternStatic(interp, &s);
assert(s == LATIN1(i));
}
#ifdef Py_DEBUG
assert(_PyUnicode_CheckConsistency(&_Py_STR(empty), 1));
for (int i = 0; i < 256; i++) {
assert(_PyUnicode_CheckConsistency(LATIN1(i), 1));
}
#endif
return _PyStatus_OK();
}
static void clear_global_interned_strings(void)
{
if (INTERNED_STRINGS != NULL) {
_Py_hashtable_destroy(INTERNED_STRINGS);
INTERNED_STRINGS = NULL;
}
}
#define _Py_RETURN_UNICODE_EMPTY() \
do { \
return _PyUnicode_GetEmpty();\
} while (0)
/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
0, 0, 0, 0, 0, 0, 0, 0,
/* case 0x0009: * CHARACTER TABULATION */
/* case 0x000A: * LINE FEED */
/* case 0x000B: * LINE TABULATION */
/* case 0x000C: * FORM FEED */
/* case 0x000D: * CARRIAGE RETURN */
0, 1, 1, 1, 1, 1, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
/* case 0x001C: * FILE SEPARATOR */
/* case 0x001D: * GROUP SEPARATOR */
/* case 0x001E: * RECORD SEPARATOR */
/* case 0x001F: * UNIT SEPARATOR */
0, 0, 0, 0, 1, 1, 1, 1,
/* case 0x0020: * SPACE */
1, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0
};
/* forward */
static PyObject* get_latin1_char(unsigned char ch);
static PyObject *
_PyUnicode_FromUCS1(const Py_UCS1 *s, Py_ssize_t size);
static PyObject *
_PyUnicode_FromUCS2(const Py_UCS2 *s, Py_ssize_t size);
static PyObject *
_PyUnicode_FromUCS4(const Py_UCS4 *s, Py_ssize_t size);
static PyObject *
unicode_encode_call_errorhandler(const char *errors,
PyObject **errorHandler,const char *encoding, const char *reason,
PyObject *unicode, PyObject **exceptionObject,
Py_ssize_t startpos, Py_ssize_t endpos, Py_ssize_t *newpos);
static void
raise_encode_exception(PyObject **exceptionObject,
const char *encoding,
PyObject *unicode,
Py_ssize_t startpos, Py_ssize_t endpos,
const char *reason);
/* Same for linebreaks */
static const unsigned char ascii_linebreak[] = {
0, 0, 0, 0, 0, 0, 0, 0,
/* 0x000A, * LINE FEED */
/* 0x000B, * LINE TABULATION */
/* 0x000C, * FORM FEED */
/* 0x000D, * CARRIAGE RETURN */
0, 0, 1, 1, 1, 1, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
/* 0x001C, * FILE SEPARATOR */
/* 0x001D, * GROUP SEPARATOR */
/* 0x001E, * RECORD SEPARATOR */
0, 0, 0, 0, 1, 1, 1, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0
};
static int convert_uc(PyObject *obj, void *addr);
struct encoding_map;
#include "clinic/unicodeobject.c.h"
_Py_error_handler
_Py_GetErrorHandler(const char *errors)
{
if (errors == NULL || strcmp(errors, "strict") == 0) {
return _Py_ERROR_STRICT;
}
if (strcmp(errors, "surrogateescape") == 0) {
return _Py_ERROR_SURROGATEESCAPE;
}
if (strcmp(errors, "replace") == 0) {
return _Py_ERROR_REPLACE;
}
if (strcmp(errors, "ignore") == 0) {
return _Py_ERROR_IGNORE;
}
if (strcmp(errors, "backslashreplace") == 0) {
return _Py_ERROR_BACKSLASHREPLACE;
}
if (strcmp(errors, "surrogatepass") == 0) {
return _Py_ERROR_SURROGATEPASS;
}
if (strcmp(errors, "xmlcharrefreplace") == 0) {
return _Py_ERROR_XMLCHARREFREPLACE;
}
return _Py_ERROR_OTHER;
}
static _Py_error_handler
get_error_handler_wide(const wchar_t *errors)
{
if (errors == NULL || wcscmp(errors, L"strict") == 0) {
return _Py_ERROR_STRICT;
}
if (wcscmp(errors, L"surrogateescape") == 0) {
return _Py_ERROR_SURROGATEESCAPE;
}
if (wcscmp(errors, L"replace") == 0) {
return _Py_ERROR_REPLACE;
}
if (wcscmp(errors, L"ignore") == 0) {
return _Py_ERROR_IGNORE;
}
if (wcscmp(errors, L"backslashreplace") == 0) {
return _Py_ERROR_BACKSLASHREPLACE;
}
if (wcscmp(errors, L"surrogatepass") == 0) {
return _Py_ERROR_SURROGATEPASS;
}
if (wcscmp(errors, L"xmlcharrefreplace") == 0) {
return _Py_ERROR_XMLCHARREFREPLACE;
}
return _Py_ERROR_OTHER;
}
static inline int
unicode_check_encoding_errors(const char *encoding, const char *errors)
{
if (encoding == NULL && errors == NULL) {
return 0;
}
PyInterpreterState *interp = _PyInterpreterState_GET();
#ifndef Py_DEBUG
/* In release mode, only check in development mode (-X dev) */
if (!_PyInterpreterState_GetConfig(interp)->dev_mode) {
return 0;
}
#else
/* Always check in debug mode */
#endif
/* Avoid calling _PyCodec_Lookup() and PyCodec_LookupError() before the
codec registry is ready: before_PyUnicode_InitEncodings() is called. */
if (!interp->unicode.fs_codec.encoding) {
return 0;
}
/* Disable checks during Python finalization. For example, it allows to
* call PyObject_Dump() during finalization for debugging purpose.
*/
if (_PyInterpreterState_GetFinalizing(interp) != NULL) {
return 0;
}
if (encoding != NULL
// Fast path for the most common built-in encodings. Even if the codec
// is cached, _PyCodec_Lookup() decodes the bytes string from UTF-8 to
// create a temporary Unicode string (the key in the cache).
&& strcmp(encoding, "utf-8") != 0
&& strcmp(encoding, "utf8") != 0
&& strcmp(encoding, "ascii") != 0)
{
PyObject *handler = _PyCodec_Lookup(encoding);
if (handler == NULL) {
return -1;
}
Py_DECREF(handler);
}
if (errors != NULL
// Fast path for the most common built-in error handlers.
&& strcmp(errors, "strict") != 0
&& strcmp(errors, "ignore") != 0
&& strcmp(errors, "replace") != 0
&& strcmp(errors, "surrogateescape") != 0
&& strcmp(errors, "surrogatepass") != 0)
{
PyObject *handler = PyCodec_LookupError(errors);
if (handler == NULL) {
return -1;
}
Py_DECREF(handler);
}
return 0;
}
int
_PyUnicode_CheckConsistency(PyObject *op, int check_content)
{
#define CHECK(expr) \
do { if (!(expr)) { _PyObject_ASSERT_FAILED_MSG(op, Py_STRINGIFY(expr)); } } while (0)
#ifdef Py_GIL_DISABLED
# define CHECK_IF_GIL(expr) (void)(expr)
# define CHECK_IF_FT(expr) CHECK(expr)
#else
# define CHECK_IF_GIL(expr) CHECK(expr)
# define CHECK_IF_FT(expr) (void)(expr)
#endif
assert(op != NULL);
CHECK(PyUnicode_Check(op));
PyASCIIObject *ascii = _PyASCIIObject_CAST(op);
int kind = ascii->state.kind;
if (ascii->state.ascii == 1 && ascii->state.compact == 1) {
CHECK(kind == PyUnicode_1BYTE_KIND);
}
else {
PyCompactUnicodeObject *compact = _PyCompactUnicodeObject_CAST(op);
void *data;
if (ascii->state.compact == 1) {
data = compact + 1;
CHECK(kind == PyUnicode_1BYTE_KIND
|| kind == PyUnicode_2BYTE_KIND
|| kind == PyUnicode_4BYTE_KIND);
CHECK(ascii->state.ascii == 0);
CHECK(_PyUnicode_UTF8(op) != data);
}
else {
PyUnicodeObject *unicode = _PyUnicodeObject_CAST(op);
data = unicode->data.any;
CHECK(kind == PyUnicode_1BYTE_KIND
|| kind == PyUnicode_2BYTE_KIND
|| kind == PyUnicode_4BYTE_KIND);
CHECK(ascii->state.compact == 0);
CHECK(data != NULL);
if (ascii->state.ascii) {
CHECK(_PyUnicode_UTF8(op) == data);
CHECK(compact->utf8_length == ascii->length);
}
else {
CHECK(_PyUnicode_UTF8(op) != data);
}
}
#ifndef Py_GIL_DISABLED
if (_PyUnicode_UTF8(op) == NULL)
CHECK(compact->utf8_length == 0);
#endif
}
/* check that the best kind is used: O(n) operation */
if (check_content) {
Py_ssize_t i;
Py_UCS4 maxchar = 0;
const void *data;
Py_UCS4 ch;
data = PyUnicode_DATA(ascii);
for (i=0; i < ascii->length; i++)
{
ch = PyUnicode_READ(kind, data, i);
if (ch > maxchar)
maxchar = ch;
}
if (kind == PyUnicode_1BYTE_KIND) {
if (ascii->state.ascii == 0) {
CHECK(maxchar >= 128);
CHECK(maxchar = 0x100);
CHECK(maxchar = 0x10000);
CHECK(maxchar length) == 0);
}
/* Check interning state */
#ifdef Py_DEBUG
// Note that we do not check `_Py_IsImmortal(op)` in the GIL-enabled build
// since stable ABI extensions can make immortal strings mortal (but with a
// high enough refcount).
switch (PyUnicode_CHECK_INTERNED(op)) {
case SSTATE_NOT_INTERNED:
if (ascii->state.statically_allocated) {
// This state is for two exceptions:
// - strings are currently checked before they're interned
// - the 256 one-latin1-character strings
// are static but use SSTATE_NOT_INTERNED
}
else {
CHECK_IF_GIL(!_Py_IsImmortal(op));
}
break;
case SSTATE_INTERNED_MORTAL:
CHECK(!ascii->state.statically_allocated);
CHECK_IF_GIL(!_Py_IsImmortal(op));
break;
case SSTATE_INTERNED_IMMORTAL:
CHECK(!ascii->state.statically_allocated);
CHECK_IF_FT(_Py_IsImmortal(op));
break;
case SSTATE_INTERNED_IMMORTAL_STATIC:
CHECK(ascii->state.statically_allocated);
CHECK_IF_FT(_Py_IsImmortal(op));
break;
default:
Py_UNREACHABLE();
}
#endif
return 1;
#undef CHECK
}
PyObject*
_PyUnicode_Result(PyObject *unicode)
{
assert(_PyUnicode_CHECK(unicode));
Py_ssize_t length = PyUnicode_GET_LENGTH(unicode);
if (length == 0) {
PyObject *empty = _PyUnicode_GetEmpty();
if (unicode != empty) {
Py_DECREF(unicode);
}
return empty;
}
if (length == 1) {
int kind = PyUnicode_KIND(unicode);
if (kind == PyUnicode_1BYTE_KIND) {
const Py_UCS1 *data = PyUnicode_1BYTE_DATA(unicode);
Py_UCS1 ch = data[0];
PyObject *latin1_char = LATIN1(ch);
if (unicode != latin1_char) {
Py_DECREF(unicode);
}
return latin1_char;
}
}
assert(_PyUnicode_CheckConsistency(unicode, 1));
return unicode;
}
#define unicode_result _PyUnicode_Result
static PyObject*
unicode_result_unchanged(PyObject *unicode)
{
if (PyUnicode_CheckExact(unicode)) {
return Py_NewRef(unicode);
}
else
/* Subtype -- return genuine unicode string with the same value. */
return _PyUnicode_Copy(unicode);
}
/* Implementation of the "backslashreplace" error handler for 8-bit encodings:
ASCII, Latin1, UTF-8, etc. */
static char*
backslashreplace(PyBytesWriter *writer, char *str,
PyObject *unicode, Py_ssize_t collstart, Py_ssize_t collend)
{
Py_ssize_t size, i;
Py_UCS4 ch;
int kind;
const void *data;
kind = PyUnicode_KIND(unicode);
data = PyUnicode_DATA(unicode);
size = 0;
/* determine replacement size */
for (i = collstart; i < collend; ++i) {
Py_ssize_t incr;
ch = PyUnicode_READ(kind, data, i);
if (ch < 0x100)
incr = 2+2;
else if (ch < 0x10000)
incr = 2+4;
else {
assert(ch PY_SSIZE_T_MAX - incr) {
PyErr_SetString(PyExc_OverflowError,
"encoded result is too long for a Python string");
return NULL;
}
size += incr;
}
str = PyBytesWriter_GrowAndUpdatePointer(writer, size, str);
if (str == NULL) {
return NULL;
}
/* generate replacement */
for (i = collstart; i < collend; ++i) {
ch = PyUnicode_READ(kind, data, i);
*str++ = '\\';
if (ch >= 0x00010000) {
*str++ = 'U';
*str++ = Py_hexdigits[(ch>>28)&0xf];
*str++ = Py_hexdigits[(ch>>24)&0xf];
*str++ = Py_hexdigits[(ch>>20)&0xf];
*str++ = Py_hexdigits[(ch>>16)&0xf];
*str++ = Py_hexdigits[(ch>>12)&0xf];
*str++ = Py_hexdigits[(ch>>8)&0xf];
}
else if (ch >= 0x100) {
*str++ = 'u';
*str++ = Py_hexdigits[(ch>>12)&0xf];
*str++ = Py_hexdigits[(ch>>8)&0xf];
}
else
*str++ = 'x';
*str++ = Py_hexdigits[(ch>>4)&0xf];
*str++ = Py_hexdigits[ch&0xf];
}
return str;
}
/* Implementation of the "xmlcharrefreplace" error handler for 8-bit encodings:
ASCII, Latin1, UTF-8, etc. */
static char*
xmlcharrefreplace(PyBytesWriter *writer, char *str,
PyObject *unicode, Py_ssize_t collstart, Py_ssize_t collend)
{
Py_ssize_t size, i;
Py_UCS4 ch;
int kind;
const void *data;
kind = PyUnicode_KIND(unicode);
data = PyUnicode_DATA(unicode);
size = 0;
/* determine replacement size */
for (i = collstart; i < collend; ++i) {
Py_ssize_t incr;
ch = PyUnicode_READ(kind, data, i);
if (ch < 10)
incr = 2+1+1;
else if (ch < 100)
incr = 2+2+1;
else if (ch < 1000)
incr = 2+3+1;
else if (ch < 10000)
incr = 2+4+1;
else if (ch < 100000)
incr = 2+5+1;
else if (ch < 1000000)
incr = 2+6+1;
else {
assert(ch PY_SSIZE_T_MAX - incr) {
PyErr_SetString(PyExc_OverflowError,
"encoded result is too long for a Python string");
return NULL;
}
size += incr;
}
str = PyBytesWriter_GrowAndUpdatePointer(writer, size, str);
if (str == NULL) {
return NULL;
}
/* generate replacement */
for (i = collstart; i < collend; ++i) {
size = sprintf(str, "&#%d;", PyUnicode_READ(kind, data, i));
if (size < 0) {
return NULL;
}
str += size;
}
return str;
}
/* --- Bloom Filters ----------------------------------------------------- */
/* stuff to implement simple "bloom filters" for Unicode characters.
to keep things simple, we use a single bitmask, using the least 5
bits from each unicode characters as the bit index. */
/* the linebreak mask is set up by _PyUnicode_Init() below */
#if LONG_BIT >= 128
#define BLOOM_WIDTH 128
#elif LONG_BIT >= 64
#define BLOOM_WIDTH 64
#elif LONG_BIT >= 32
#define BLOOM_WIDTH 32
#else
#error "LONG_BIT is smaller than 32"
#endif
#define BLOOM_MASK unsigned long
static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
#define BLOOM(mask, ch) ((mask & (1UL 0)
return ucs2lib_find_char((const Py_UCS2 *) s, size, (Py_UCS2) ch);
else
return ucs2lib_rfind_char((const Py_UCS2 *) s, size, (Py_UCS2) ch);
case PyUnicode_4BYTE_KIND:
if (direction > 0)
return ucs4lib_find_char((const Py_UCS4 *) s, size, ch);
else
return ucs4lib_rfind_char((const Py_UCS4 *) s, size, ch);
default:
Py_UNREACHABLE();
}
}
#ifdef Py_DEBUG
/* Fill the data of a Unicode string with invalid characters to detect bugs
earlier.
_PyUnicode_CheckConsistency(str, 1) detects invalid characters, at least for
ASCII and UCS-4 strings. U+00FF is invalid in ASCII and U+FFFFFFFF is an
invalid character in Unicode 6.0. */
static void
unicode_fill_invalid(PyObject *unicode, Py_ssize_t old_length)
{
int kind = PyUnicode_KIND(unicode);
Py_UCS1 *data = PyUnicode_1BYTE_DATA(unicode);
Py_ssize_t length = _PyUnicode_LENGTH(unicode);
if (length ((PY_SSIZE_T_MAX - struct_size) / char_size - 1)) {
PyErr_NoMemory();
return NULL;
}
new_size = (struct_size + (length + 1) * char_size);
if (_PyUnicode_HAS_UTF8_MEMORY(unicode)) {
PyMem_Free(_PyUnicode_UTF8(unicode));
PyUnicode_SET_UTF8_LENGTH(unicode, 0);
PyUnicode_SET_UTF8(unicode, NULL);
}
#ifdef Py_TRACE_REFS
_Py_ForgetReference(unicode);
#endif
_PyReftracerTrack(unicode, PyRefTracer_DESTROY);
new_unicode = (PyObject *)PyObject_Realloc(unicode, new_size);
if (new_unicode == NULL) {
_Py_NewReferenceNoTotal(unicode);
PyErr_NoMemory();
return NULL;
}
unicode = new_unicode;
_Py_NewReferenceNoTotal(unicode);
_PyUnicode_LENGTH(unicode) = length;
#ifdef Py_DEBUG
unicode_fill_invalid(unicode, old_length);
#endif
PyUnicode_WRITE(PyUnicode_KIND(unicode), PyUnicode_DATA(unicode),
length, 0);
assert(_PyUnicode_CheckConsistency(unicode, 0));
return unicode;
}
static int
resize_inplace(PyObject *unicode, Py_ssize_t length)
{
assert(!PyUnicode_IS_COMPACT(unicode));
assert(Py_REFCNT(unicode) == 1);
Py_ssize_t new_size;
Py_ssize_t char_size;
int share_utf8;
void *data;
#ifdef Py_DEBUG
Py_ssize_t old_length = _PyUnicode_LENGTH(unicode);
#endif
data = _PyUnicode_DATA_ANY(unicode);
char_size = PyUnicode_KIND(unicode);
share_utf8 = _PyUnicode_SHARE_UTF8(unicode);
if (length > (PY_SSIZE_T_MAX / char_size - 1)) {
PyErr_NoMemory();
return -1;
}
new_size = (length + 1) * char_size;
if (!share_utf8 && _PyUnicode_HAS_UTF8_MEMORY(unicode))
{
PyMem_Free(_PyUnicode_UTF8(unicode));
PyUnicode_SET_UTF8_LENGTH(unicode, 0);
PyUnicode_SET_UTF8(unicode, NULL);
}
data = (PyObject *)PyObject_Realloc(data, new_size);
if (data == NULL) {
PyErr_NoMemory();
return -1;
}
_PyUnicode_DATA_ANY(unicode) = data;
if (share_utf8) {
PyUnicode_SET_UTF8_LENGTH(unicode, length);
PyUnicode_SET_UTF8(unicode, data);
}
_PyUnicode_LENGTH(unicode) = length;
PyUnicode_WRITE(PyUnicode_KIND(unicode), data, length, 0);
#ifdef Py_DEBUG
unicode_fill_invalid(unicode, old_length);
#endif
/* check for integer overflow */
if (length > PY_SSIZE_T_MAX / (Py_ssize_t)sizeof(wchar_t) - 1) {
PyErr_NoMemory();
return -1;
}
assert(_PyUnicode_CheckConsistency(unicode, 0));
return 0;
}
static const char*
unicode_kind_name(PyObject *unicode)
{
/* don't check consistency: unicode_kind_name() is called from
_PyUnicode_Dump() */
if (!PyUnicode_IS_COMPACT(unicode))
{
switch (PyUnicode_KIND(unicode))
{
case PyUnicode_1BYTE_KIND:
if (PyUnicode_IS_ASCII(unicode))
return "legacy ascii";
else
return "legacy latin1";
case PyUnicode_2BYTE_KIND:
return "legacy UCS2";
case PyUnicode_4BYTE_KIND:
return "legacy UCS4";
default:
return "";
}
}
switch (PyUnicode_KIND(unicode)) {
case PyUnicode_1BYTE_KIND:
if (PyUnicode_IS_ASCII(unicode))
return "ascii";
else
return "latin1";
case PyUnicode_2BYTE_KIND:
return "UCS2";
case PyUnicode_4BYTE_KIND:
return "UCS4";
default:
return "";
}
}
#ifdef Py_DEBUG
/* Functions wrapping macros for use in debugger */
const char *_PyUnicode_utf8(void *unicode_raw){
PyObject *unicode = _PyObject_CAST(unicode_raw);
return PyUnicode_UTF8(unicode);
}
const void *_PyUnicode_compact_data(void *unicode_raw) {
PyObject *unicode = _PyObject_CAST(unicode_raw);
return _PyUnicode_COMPACT_DATA(unicode);
}
const void *_PyUnicode_data(void *unicode_raw) {
PyObject *unicode = _PyObject_CAST(unicode_raw);
printf("obj %p\n", (void*)unicode);
printf("compact %d\n", PyUnicode_IS_COMPACT(unicode));
printf("compact ascii %d\n", PyUnicode_IS_COMPACT_ASCII(unicode));
printf("ascii op %p\n", (void*)(_PyASCIIObject_CAST(unicode) + 1));
printf("compact op %p\n", (void*)(_PyCompactUnicodeObject_CAST(unicode) + 1));
printf("compact data %p\n", _PyUnicode_COMPACT_DATA(unicode));
return PyUnicode_DATA(unicode);
}
void
_PyUnicode_Dump(PyObject *op)
{
PyASCIIObject *ascii = _PyASCIIObject_CAST(op);
PyCompactUnicodeObject *compact = _PyCompactUnicodeObject_CAST(op);
PyUnicodeObject *unicode = _PyUnicodeObject_CAST(op);
const void *data;
if (ascii->state.compact)
{
if (ascii->state.ascii)
data = (ascii + 1);
else
data = (compact + 1);
}
else
data = unicode->data.any;
printf("%s: len=%zu, ", unicode_kind_name(op), ascii->length);
if (!ascii->state.ascii) {
printf("utf8=%p (%zu)", (void *)compact->utf8, compact->utf8_length);
}
printf(", data=%p\n", data);
}
#endif
PyObject *
PyUnicode_New(Py_ssize_t size, Py_UCS4 maxchar)
{
/* Optimization for empty strings */
if (size == 0) {
return _PyUnicode_GetEmpty();
}
PyObject *obj;
PyCompactUnicodeObject *unicode;
void *data;
int kind;
int is_ascii;
Py_ssize_t char_size;
Py_ssize_t struct_size;
is_ascii = 0;
struct_size = sizeof(PyCompactUnicodeObject);
if (maxchar < 128) {
kind = PyUnicode_1BYTE_KIND;
char_size = 1;
is_ascii = 1;
struct_size = sizeof(PyASCIIObject);
}
else if (maxchar < 256) {
kind = PyUnicode_1BYTE_KIND;
char_size = 1;
}
else if (maxchar < 65536) {
kind = PyUnicode_2BYTE_KIND;
char_size = 2;
}
else {
if (maxchar > MAX_UNICODE) {
PyErr_SetString(PyExc_SystemError,
"invalid maximum character passed to PyUnicode_New");
return NULL;
}
kind = PyUnicode_4BYTE_KIND;
char_size = 4;
}
/* Ensure we won't overflow the size. */
if (size < 0) {
PyErr_SetString(PyExc_SystemError,
"Negative size passed to PyUnicode_New");
return NULL;
}
if (size > ((PY_SSIZE_T_MAX - struct_size) / char_size - 1))
return PyErr_NoMemory();
/* Duplicated allocation code from _PyObject_New() instead of a call to
* PyObject_New() so we are able to allocate space for the object and
* it's data buffer.
*/
obj = (PyObject *) PyObject_Malloc(struct_size + (size + 1) * char_size);
if (obj == NULL) {
return PyErr_NoMemory();
}
_PyObject_Init(obj, &PyUnicode_Type);
unicode = (PyCompactUnicodeObject *)obj;
if (is_ascii)
data = ((PyASCIIObject*)obj) + 1;
else
data = unicode + 1;
_PyUnicode_LENGTH(unicode) = size;
_PyUnicode_HASH(unicode) = -1;
_PyUnicode_STATE(unicode).interned = 0;
_PyUnicode_STATE(unicode).kind = kind;
_PyUnicode_STATE(unicode).compact = 1;
_PyUnicode_STATE(unicode).ascii = is_ascii;
_PyUnicode_STATE(unicode).statically_allocated = 0;
if (is_ascii) {
((char*)data)[size] = 0;
}
else if (kind == PyUnicode_1BYTE_KIND) {
((char*)data)[size] = 0;
unicode->utf8 = NULL;
unicode->utf8_length = 0;
}
else {
unicode->utf8 = NULL;
unicode->utf8_length = 0;
if (kind == PyUnicode_2BYTE_KIND)
((Py_UCS2*)data)[size] = 0;
else /* kind == PyUnicode_4BYTE_KIND */
((Py_UCS4*)data)[size] = 0;
}
#ifdef Py_DEBUG
unicode_fill_invalid((PyObject*)unicode, 0);
#endif
assert(_PyUnicode_CheckConsistency((PyObject*)unicode, 0));
return obj;
}
static int
unicode_check_modifiable(PyObject *unicode)
{
if (!_PyUnicode_IsModifiable(unicode)) {
PyErr_SetString(PyExc_SystemError,
"Cannot modify a string currently used");
return -1;
}
return 0;
}
static int
_copy_characters(PyObject *to, Py_ssize_t to_start,
PyObject *from, Py_ssize_t from_start,
Py_ssize_t how_many, int check_maxchar)
{
int from_kind, to_kind;
const void *from_data;
void *to_data;
assert(0 (size_t)PyUnicode_GET_LENGTH(to)) {
PyErr_SetString(PyExc_IndexError, "string index out of range");
return -1;
}
if (how_many < 0) {
PyErr_SetString(PyExc_SystemError, "how_many cannot be negative");
return -1;
}
how_many = Py_MIN(PyUnicode_GET_LENGTH(from)-from_start, how_many);
if (to_start + how_many > PyUnicode_GET_LENGTH(to)) {
PyErr_Format(PyExc_SystemError,
"Cannot write %zi characters at %zi "
"in a string of %zi characters",
how_many, to_start, PyUnicode_GET_LENGTH(to));
return -1;
}
if (how_many == 0)
return 0;
if (unicode_check_modifiable(to))
return -1;
err = _copy_characters(to, to_start, from, from_start, how_many, 1);
if (err) {
PyErr_Format(PyExc_SystemError,
"Cannot copy %s characters "
"into a string of %s characters",
unicode_kind_name(from),
unicode_kind_name(to));
return -1;
}
return how_many;
}
/* Find the maximum code point and count the number of surrogate pairs so a
correct string length can be computed before converting a string to UCS4.
This function counts single surrogates as a character and not as a pair.
Return 0 on success, or -1 on error. */
static int
find_maxchar_surrogates(const wchar_t *begin, const wchar_t *end,
Py_UCS4 *maxchar, Py_ssize_t *num_surrogates)
{
const wchar_t *iter;
Py_UCS4 ch;
assert(num_surrogates != NULL && maxchar != NULL);
*num_surrogates = 0;
*maxchar = 0;
for (iter = begin; iter < end; ) {
#if SIZEOF_WCHAR_T == 2
if (Py_UNICODE_IS_HIGH_SURROGATE(iter[0])
&& (iter+1) < end
&& Py_UNICODE_IS_LOW_SURROGATE(iter[1]))
{
ch = Py_UNICODE_JOIN_SURROGATES(iter[0], iter[1]);
++(*num_surrogates);
iter += 2;
}
else
#endif
{
ch = *iter;
iter++;
}
if (ch > *maxchar) {
*maxchar = ch;
if (*maxchar > MAX_UNICODE) {
PyErr_Format(PyExc_ValueError,
"character U+%x is not in range [U+0000; U+%x]",
ch, MAX_UNICODE);
return -1;
}
}
}
return 0;
}
static void
unicode_dealloc(PyObject *unicode)
{
#ifdef Py_DEBUG
if (!unicode_is_finalizing() && unicode_is_singleton(unicode)) {
_Py_FatalRefcountError("deallocating an Unicode singleton");
}
#endif
if (_PyUnicode_STATE(unicode).statically_allocated) {
/* This should never get called, but we also don't want to SEGV if
* we accidentally decref an immortal string out of existence. Since
* the string is an immortal object, just re-set the reference count.
*/
#ifdef Py_DEBUG
Py_UNREACHABLE();
#endif
_Py_SetImmortal(unicode);
return;
}
switch (_PyUnicode_STATE(unicode).interned) {
case SSTATE_NOT_INTERNED:
break;
case SSTATE_INTERNED_MORTAL:
/* Remove the object from the intern dict.
* Before doing so, we set the refcount to 2: the key and value
* in the interned_dict.
*/
assert(Py_REFCNT(unicode) == 0);
Py_SET_REFCNT(unicode, 2);
#ifdef Py_REF_DEBUG
/* let's be pedantic with the ref total */
_Py_IncRefTotal(_PyThreadState_GET());
_Py_IncRefTotal(_PyThreadState_GET());
#endif
PyInterpreterState *interp = _PyInterpreterState_GET();
PyObject *interned = get_interned_dict(interp);
assert(interned != NULL);
PyObject *popped;
int r = PyDict_Pop(interned, unicode, &popped);
if (r == -1) {
PyErr_FormatUnraisable("Exception ignored while "
"removing an interned string %R",
unicode);
// We don't know what happened to the string. It's probably
// best to leak it:
// - if it was popped, there are no more references to it
// so it can't cause trouble (except wasted memory)
// - if it wasn't popped, it'll remain interned
_Py_SetImmortal(unicode);
_PyUnicode_STATE(unicode).interned = SSTATE_INTERNED_IMMORTAL;
return;
}
if (r == 0) {
// The interned string was not found in the interned_dict.
#ifdef Py_DEBUG
Py_UNREACHABLE();
#endif
_Py_SetImmortal(unicode);
return;
}
// Successfully popped.
assert(popped == unicode);
// Only our `popped` reference should be left; remove it too.
assert(Py_REFCNT(unicode) == 1);
Py_SET_REFCNT(unicode, 0);
#ifdef Py_REF_DEBUG
/* let's be pedantic with the ref total */
_Py_DecRefTotal(_PyThreadState_GET());
#endif
break;
default:
// As with `statically_allocated` above.
#ifdef Py_REF_DEBUG
Py_UNREACHABLE();
#endif
_Py_SetImmortal(unicode);
return;
}
if (_PyUnicode_HAS_UTF8_MEMORY(unicode)) {
PyMem_Free(_PyUnicode_UTF8(unicode));
}
if (!PyUnicode_IS_COMPACT(unicode) && _PyUnicode_DATA_ANY(unicode)) {
PyMem_Free(_PyUnicode_DATA_ANY(unicode));
}
Py_TYPE(unicode)->tp_free(unicode);
}
#ifdef Py_DEBUG
static int
unicode_is_singleton(PyObject *unicode)
{
if (unicode == &_Py_STR(empty)) {
return 1;
}
PyASCIIObject *ascii = _PyASCIIObject_CAST(unicode);
if (ascii->length == 1) {
Py_UCS4 ch = PyUnicode_READ_CHAR(unicode, 0);
if (ch < 256 && LATIN1(ch) == unicode) {
return 1;
}
}
return 0;
}
#endif
int
_PyUnicode_IsModifiable(PyObject *unicode)
{
assert(_PyUnicode_CHECK(unicode));
if (!_PyObject_IsUniquelyReferenced(unicode))
return 0;
if (PyUnicode_HASH(unicode) != -1)
return 0;
if (PyUnicode_CHECK_INTERNED(unicode))
return 0;
if (!PyUnicode_CheckExact(unicode))
return 0;
#ifdef Py_DEBUG
/* singleton refcount is greater than 1 */
assert(!unicode_is_singleton(unicode));
#endif
return 1;
}
static int
unicode_resize(PyObject **p_unicode, Py_ssize_t length)
{
PyObject *unicode;
Py_ssize_t old_length;
assert(p_unicode != NULL);
unicode = *p_unicode;
assert(unicode != NULL);
assert(PyUnicode_Check(unicode));
assert(0 data + writer->pos * kind;
unicode_write_widechar(kind, data, str, size, num_surrogates);
writer->pos += size - num_surrogates;
return 0;
}
PyObject *
PyUnicode_FromStringAndSize(const char *u, Py_ssize_t size)
{
if (size < 0) {
PyErr_SetString(PyExc_SystemError,
"Negative size passed to PyUnicode_FromStringAndSize");
return NULL;
}
if (u != NULL) {
return PyUnicode_DecodeUTF8Stateful(u, size, NULL, NULL);
}
if (size > 0) {
PyErr_SetString(PyExc_SystemError,
"NULL string with positive size with NULL passed to PyUnicode_FromStringAndSize");
return NULL;
}
return _PyUnicode_GetEmpty();
}
PyObject *
PyUnicode_FromString(const char *u)
{
size_t size = strlen(u);
if (size > PY_SSIZE_T_MAX) {
PyErr_SetString(PyExc_OverflowError, "input too long");
return NULL;
}
return PyUnicode_DecodeUTF8Stateful(u, (Py_ssize_t)size, NULL, NULL);
}
PyObject *
_PyUnicode_FromId(_Py_Identifier *id)
{
PyMutex_Lock((PyMutex *)&id->mutex);
PyInterpreterState *interp = _PyInterpreterState_GET();
struct _Py_unicode_ids *ids = &interp->unicode.ids;
Py_ssize_t index = _Py_atomic_load_ssize(&id->index);
if (index < 0) {
struct _Py_unicode_runtime_ids *rt_ids = &interp->runtime->unicode_state.ids;
PyMutex_Lock(&rt_ids->mutex);
// Check again to detect concurrent access. Another thread can have
// initialized the index while this thread waited for the lock.
index = _Py_atomic_load_ssize(&id->index);
if (index < 0) {
assert(rt_ids->next_index < PY_SSIZE_T_MAX);
index = rt_ids->next_index;
rt_ids->next_index++;
_Py_atomic_store_ssize(&id->index, index);
}
PyMutex_Unlock(&rt_ids->mutex);
}
assert(index >= 0);
PyObject *obj;
if (index < ids->size) {
obj = ids->array[index];
if (obj) {
// Return a borrowed reference
goto end;
}
}
obj = PyUnicode_DecodeUTF8Stateful(id->string, strlen(id->string),
NULL, NULL);
if (!obj) {
goto end;
}
_PyUnicode_InternImmortal(interp, &obj);
if (index >= ids->size) {
// Overallocate to reduce the number of realloc
Py_ssize_t new_size = Py_MAX(index * 2, 16);
Py_ssize_t item_size = sizeof(ids->array[0]);
PyObject **new_array = PyMem_Realloc(ids->array, new_size * item_size);
if (new_array == NULL) {
PyErr_NoMemory();
obj = NULL;
goto end;
}
memset(&new_array[ids->size], 0, (new_size - ids->size) * item_size);
ids->array = new_array;
ids->size = new_size;
}
// The array stores a strong reference
ids->array[index] = obj;
end:
PyMutex_Unlock((PyMutex *)&id->mutex);
// Return a borrowed reference
return obj;
}
static void
unicode_clear_identifiers(struct _Py_unicode_state *state)
{
struct _Py_unicode_ids *ids = &state->ids;
for (Py_ssize_t i=0; i < ids->size; i++) {
Py_XDECREF(ids->array[i]);
}
ids->size = 0;
PyMem_Free(ids->array);
ids->array = NULL;
// Don't reset _PyRuntime next_index: _Py_Identifier.id remains valid
// after Py_Finalize().
}
/* Internal function, doesn't check maximum character */
PyObject*
_PyUnicode_FromASCII(const char *buffer, Py_ssize_t size)
{
const unsigned char *s = (const unsigned char *)buffer;
PyObject *unicode;
if (size == 1) {
#ifdef Py_DEBUG
assert((unsigned char)s[0] < 128);
#endif
return get_latin1_char(s[0]);
}
unicode = PyUnicode_New(size, 127);
if (!unicode)
return NULL;
memcpy(PyUnicode_1BYTE_DATA(unicode), s, size);
assert(_PyUnicode_CheckConsistency(unicode, 1));
return unicode;
}
static Py_UCS4
kind_maxchar_limit(int kind)
{
switch (kind) {
case PyUnicode_1BYTE_KIND:
return 0x80;
case PyUnicode_2BYTE_KIND:
return 0x100;
case PyUnicode_4BYTE_KIND:
return 0x10000;
default:
Py_UNREACHABLE();
}
}
static PyObject*
_PyUnicode_FromUCS1(const Py_UCS1* u, Py_ssize_t size)
{
PyObject *res;
unsigned char max_char;
if (size == 0) {
_Py_RETURN_UNICODE_EMPTY();
}
assert(size > 0);
if (size == 1) {
return get_latin1_char(u[0]);
}
max_char = ucs1lib_find_max_char(u, u + size);
res = PyUnicode_New(size, max_char);
if (!res)
return NULL;
memcpy(PyUnicode_1BYTE_DATA(res), u, size);
assert(_PyUnicode_CheckConsistency(res, 1));
return res;
}
static PyObject*
_PyUnicode_FromUCS2(const Py_UCS2 *u, Py_ssize_t size)
{
PyObject *res;
Py_UCS2 max_char;
if (size == 0)
_Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1)
return unicode_char(u[0]);
max_char = ucs2lib_find_max_char(u, u + size);
res = PyUnicode_New(size, max_char);
if (!res)
return NULL;
if (max_char >= 256)
memcpy(PyUnicode_2BYTE_DATA(res), u, sizeof(Py_UCS2)*size);
else {
_PyUnicode_CONVERT_BYTES(
Py_UCS2, Py_UCS1, u, u + size, PyUnicode_1BYTE_DATA(res));
}
assert(_PyUnicode_CheckConsistency(res, 1));
return res;
}
static PyObject*
_PyUnicode_FromUCS4(const Py_UCS4 *u, Py_ssize_t size)
{
PyObject *res;
Py_UCS4 max_char;
if (size == 0)
_Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1)
return unicode_char(u[0]);
max_char = ucs4lib_find_max_char(u, u + size);
res = PyUnicode_New(size, max_char);
if (!res)
return NULL;
if (max_char < 256)
_PyUnicode_CONVERT_BYTES(Py_UCS4, Py_UCS1, u, u + size,
PyUnicode_1BYTE_DATA(res));
else if (max_char < 0x10000)
_PyUnicode_CONVERT_BYTES(Py_UCS4, Py_UCS2, u, u + size,
PyUnicode_2BYTE_DATA(res));
else
memcpy(PyUnicode_4BYTE_DATA(res), u, sizeof(Py_UCS4)*size);
assert(_PyUnicode_CheckConsistency(res, 1));
return res;
}
int
PyUnicodeWriter_WriteUCS4(PyUnicodeWriter *pub_writer,
const Py_UCS4 *str,
Py_ssize_t size)
{
_PyUnicodeWriter *writer = (_PyUnicodeWriter*)pub_writer;
if (size < 0) {
PyErr_SetString(PyExc_ValueError,
"size must be positive");
return -1;
}
if (size == 0) {
return 0;
}
Py_UCS4 max_char = ucs4lib_find_max_char(str, str + size);
if (_PyUnicodeWriter_Prepare(writer, size, max_char) < 0) {
return -1;
}
int kind = writer->kind;
void *data = (Py_UCS1*)writer->data + writer->pos * kind;
if (kind == PyUnicode_1BYTE_KIND) {
_PyUnicode_CONVERT_BYTES(Py_UCS4, Py_UCS1,
str, str + size,
data);
}
else if (kind == PyUnicode_2BYTE_KIND) {
_PyUnicode_CONVERT_BYTES(Py_UCS4, Py_UCS2,
str, str + size,
data);
}
else {
memcpy(data, str, size * sizeof(Py_UCS4));
}
writer->pos += size;
return 0;
}
PyObject*
PyUnicode_FromKindAndData(int kind, const void *buffer, Py_ssize_t size)
{
if (size < 0) {
PyErr_SetString(PyExc_ValueError, "size must be positive");
return NULL;
}
switch (kind) {
case PyUnicode_1BYTE_KIND:
return _PyUnicode_FromUCS1(buffer, size);
case PyUnicode_2BYTE_KIND:
return _PyUnicode_FromUCS2(buffer, size);
case PyUnicode_4BYTE_KIND:
return _PyUnicode_FromUCS4(buffer, size);
default:
PyErr_SetString(PyExc_SystemError, "invalid kind");
return NULL;
}
}
Py_UCS4
_PyUnicode_FindMaxChar(PyObject *unicode, Py_ssize_t start, Py_ssize_t end)
{
int kind;
const void *startptr, *endptr;
assert(0 = 0x10000)
return;
}
else
Py_UNREACHABLE();
copy = PyUnicode_New(len, max_char);
if (copy != NULL)
_PyUnicode_FastCopyCharacters(copy, 0, unicode, 0, len);
Py_DECREF(unicode);
*p_unicode = copy;
}
PyObject*
_PyUnicode_Copy(PyObject *unicode)
{
Py_ssize_t length;
PyObject *copy;
if (!PyUnicode_Check(unicode)) {
PyErr_BadInternalCall();
return NULL;
}
length = PyUnicode_GET_LENGTH(unicode);
copy = PyUnicode_New(length, PyUnicode_MAX_CHAR_VALUE(unicode));
if (!copy)
return NULL;
assert(PyUnicode_KIND(copy) == PyUnicode_KIND(unicode));
memcpy(PyUnicode_DATA(copy), PyUnicode_DATA(unicode),
length * PyUnicode_KIND(unicode));
assert(_PyUnicode_CheckConsistency(copy, 1));
return copy;
}
/* Widen Unicode objects to larger buffers. Don't write terminating null
character. Return NULL on error. */
static void*
unicode_askind(int skind, void const *data, Py_ssize_t len, int kind)
{
void *result;
assert(skind < kind);
switch (kind) {
case PyUnicode_2BYTE_KIND:
result = PyMem_New(Py_UCS2, len);
if (!result)
return PyErr_NoMemory();
assert(skind == PyUnicode_1BYTE_KIND);
_PyUnicode_CONVERT_BYTES(
Py_UCS1, Py_UCS2,
(const Py_UCS1 *)data,
((const Py_UCS1 *)data) + len,
result);
return result;
case PyUnicode_4BYTE_KIND:
result = PyMem_New(Py_UCS4, len);
if (!result)
return PyErr_NoMemory();
if (skind == PyUnicode_2BYTE_KIND) {
_PyUnicode_CONVERT_BYTES(
Py_UCS2, Py_UCS4,
(const Py_UCS2 *)data,
((const Py_UCS2 *)data) + len,
result);
}
else {
assert(skind == PyUnicode_1BYTE_KIND);
_PyUnicode_CONVERT_BYTES(
Py_UCS1, Py_UCS4,
(const Py_UCS1 *)data,
((const Py_UCS1 *)data) + len,
result);
}
return result;
default:
Py_UNREACHABLE();
return NULL;
}
}
static Py_UCS4*
as_ucs4(PyObject *string, Py_UCS4 *target, Py_ssize_t targetsize,
int copy_null)
{
int kind;
const void *data;
Py_ssize_t len, targetlen;
kind = PyUnicode_KIND(string);
data = PyUnicode_DATA(string);
len = PyUnicode_GET_LENGTH(string);
targetlen = len;
if (copy_null)
targetlen++;
if (!target) {
target = PyMem_New(Py_UCS4, targetlen);
if (!target) {
PyErr_NoMemory();
return NULL;
}
}
else {
if (targetsize < targetlen) {
PyErr_Format(PyExc_SystemError,
"string is longer than the buffer");
if (copy_null && 0 < targetsize)
target[0] = 0;
return NULL;
}
}
if (kind == PyUnicode_1BYTE_KIND) {
const Py_UCS1 *start = (const Py_UCS1 *) data;
_PyUnicode_CONVERT_BYTES(Py_UCS1, Py_UCS4, start, start + len, target);
}
else if (kind == PyUnicode_2BYTE_KIND) {
const Py_UCS2 *start = (const Py_UCS2 *) data;
_PyUnicode_CONVERT_BYTES(Py_UCS2, Py_UCS4, start, start + len, target);
}
else if (kind == PyUnicode_4BYTE_KIND) {
memcpy(target, data, len * sizeof(Py_UCS4));
}
else {
Py_UNREACHABLE();
}
if (copy_null)
target[len] = 0;
return target;
}
Py_UCS4*
PyUnicode_AsUCS4(PyObject *string, Py_UCS4 *target, Py_ssize_t targetsize,
int copy_null)
{
if (target == NULL || targetsize < 0) {
PyErr_BadInternalCall();
return NULL;
}
return as_ucs4(string, target, targetsize, copy_null);
}
Py_UCS4*
PyUnicode_AsUCS4Copy(PyObject *string)
{
return as_ucs4(string, NULL, 0, 1);
}
/* maximum number of characters required for output of %jo or %jd or %p.
We need at most ceil(log8(256)*sizeof(intmax_t)) digits,
plus 1 for the sign, plus 2 for the 0x prefix (for %p),
plus 1 for the terminal NUL. */
#define MAX_INTMAX_CHARS (5 + (sizeof(intmax_t)*8-1) / 3)
static int
unicode_fromformat_write_str(_PyUnicodeWriter *writer, PyObject *str,
Py_ssize_t width, Py_ssize_t precision, int flags)
{
Py_ssize_t length, fill, arglen;
Py_UCS4 maxchar;
length = PyUnicode_GET_LENGTH(str);
if ((precision == -1 || precision >= length)
&& width writer->maxchar)
maxchar = _PyUnicode_FindMaxChar(str, 0, length);
else
maxchar = writer->maxchar;
if (_PyUnicodeWriter_Prepare(writer, arglen, maxchar) == -1)
return -1;
fill = Py_MAX(width - length, 0);
if (fill && !(flags & F_LJUST)) {
if (PyUnicode_Fill(writer->buffer, writer->pos, fill, ' ') == -1)
return -1;
writer->pos += fill;
}
_PyUnicode_FastCopyCharacters(writer->buffer, writer->pos,
str, 0, length);
writer->pos += length;
if (fill && (flags & F_LJUST)) {
if (PyUnicode_Fill(writer->buffer, writer->pos, fill, ' ') == -1)
return -1;
writer->pos += fill;
}
return 0;
}
static int
unicode_fromformat_write_utf8(_PyUnicodeWriter *writer, const char *str,
Py_ssize_t width, Py_ssize_t precision, int flags)
{
/* UTF-8 */
Py_ssize_t *pconsumed = NULL;
Py_ssize_t length;
if (precision == -1) {
length = strlen(str);
}
else {
length = 0;
while (length < precision && str[length]) {
length++;
}
if (length == precision) {
/* The input string is not NUL-terminated. If it ends with an
* incomplete UTF-8 sequence, truncate the string just before it.
* Incomplete sequences in the middle and sequences which cannot
* be valid prefixes are still treated as errors and replaced
* with \xfffd. */
pconsumed = &length;
}
}
if (width < 0) {
return _PyUnicode_DecodeUTF8Writer(writer, str, length,
_Py_ERROR_REPLACE, "replace", pconsumed);
}
PyObject *unicode = PyUnicode_DecodeUTF8Stateful(str, length,
"replace", pconsumed);
if (unicode == NULL)
return -1;
int res = unicode_fromformat_write_str(writer, unicode,
width, -1, flags);
Py_DECREF(unicode);
return res;
}
static int
unicode_fromformat_write_wcstr(_PyUnicodeWriter *writer, const wchar_t *str,
Py_ssize_t width, Py_ssize_t precision, int flags)
{
Py_ssize_t length;
if (precision == -1) {
length = wcslen(str);
}
else {
length = 0;
while (length < precision && str[length]) {
length++;
}
}
if (width < 0) {
return PyUnicodeWriter_WriteWideChar((PyUnicodeWriter*)writer,
str, length);
}
PyObject *unicode = PyUnicode_FromWideChar(str, length);
if (unicode == NULL)
return -1;
int res = unicode_fromformat_write_str(writer, unicode, width, -1, flags);
Py_DECREF(unicode);
return res;
}
#define F_LONG 1
#define F_LONGLONG 2
#define F_SIZE 3
#define F_PTRDIFF 4
#define F_INTMAX 5
static const char*
unicode_fromformat_arg(_PyUnicodeWriter *writer,
const char *f, va_list *vargs)
{
const char *p;
Py_ssize_t len;
int flags = 0;
Py_ssize_t width;
Py_ssize_t precision;
p = f;
f++;
if (*f == '%') {
if (_PyUnicodeWriter_WriteCharInline(writer, '%') < 0)
return NULL;
f++;
return f;
}
/* Parse flags. Example: "%-i" => flags=F_LJUST. */
/* Flags '+', ' ' and '#' are not particularly useful.
* They are not worth the implementation and maintenance costs.
* In addition, '#' should add "0" for "o" conversions for compatibility
* with printf, but it would confuse Python users. */
while (1) {
switch (*f++) {
case '-': flags |= F_LJUST; continue;
case '0': flags |= F_ZERO; continue;
case '#': flags |= F_ALT; continue;
}
f--;
break;
}
/* parse the width.precision part, e.g. "%2.5s" => width=2, precision=5 */
width = -1;
if (*f == '*') {
width = va_arg(*vargs, int);
if (width < 0) {
flags |= F_LJUST;
width = -width;
}
f++;
}
else if (Py_ISDIGIT((unsigned)*f)) {
width = *f - '0';
f++;
while (Py_ISDIGIT((unsigned)*f)) {
if (width > (PY_SSIZE_T_MAX - ((int)*f - '0')) / 10) {
PyErr_SetString(PyExc_ValueError,
"width too big");
return NULL;
}
width = (width * 10) + (*f - '0');
f++;
}
}
precision = -1;
if (*f == '.') {
f++;
if (*f == '*') {
precision = va_arg(*vargs, int);
if (precision < 0) {
precision = -2;
}
f++;
}
else if (Py_ISDIGIT((unsigned)*f)) {
precision = (*f - '0');
f++;
while (Py_ISDIGIT((unsigned)*f)) {
if (precision > (PY_SSIZE_T_MAX - ((int)*f - '0')) / 10) {
PyErr_SetString(PyExc_ValueError,
"precision too big");
return NULL;
}
precision = (precision * 10) + (*f - '0');
f++;
}
}
}
int sizemod = 0;
if (*f == 'l') {
if (f[1] == 'l') {
sizemod = F_LONGLONG;
f += 2;
}
else {
sizemod = F_LONG;
++f;
}
}
else if (*f == 'z') {
sizemod = F_SIZE;
++f;
}
else if (*f == 't') {
sizemod = F_PTRDIFF;
++f;
}
else if (*f == 'j') {
sizemod = F_INTMAX;
++f;
}
if (f[0] != '\0' && f[1] == '\0')
writer->overallocate = 0;
switch (*f) {
case 'd': case 'i': case 'o': case 'u': case 'x': case 'X':
break;
case 'c': case 'p':
if (sizemod || width >= 0 || precision >= 0) goto invalid_format;
break;
case 's':
case 'V':
if (sizemod && sizemod != F_LONG) goto invalid_format;
break;
default:
if (sizemod) goto invalid_format;
break;
}
switch (*f) {
case 'c':
{
int ordinal = va_arg(*vargs, int);
if (ordinal < 0 || ordinal > MAX_UNICODE) {
PyErr_SetString(PyExc_OverflowError,
"character argument not in range(0x110000)");
return NULL;
}
if (_PyUnicodeWriter_WriteCharInline(writer, ordinal) < 0)
return NULL;
break;
}
case 'd': case 'i':
case 'o': case 'u': case 'x': case 'X':
{
char buffer[MAX_INTMAX_CHARS];
// Fill buffer using sprinf, with one of many possible format
// strings, like "%llX" for `long long` in hexadecimal.
// The type/size is in `sizemod`; the format is in `*f`.
// Use macros with nested switches to keep the sprintf format strings
// as compile-time literals, avoiding warnings and maybe allowing
// optimizations.
// `SPRINT` macro does one sprintf
// Example usage: SPRINT("l", "X", unsigned long) expands to
// sprintf(buffer, "%" "l" "X", va_arg(*vargs, unsigned long))
#define SPRINT(SIZE_SPEC, FMT_CHAR, TYPE) \
sprintf(buffer, "%" SIZE_SPEC FMT_CHAR, va_arg(*vargs, TYPE))
// One inner switch to handle all format variants
#define DO_SPRINTS(SIZE_SPEC, SIGNED_TYPE, UNSIGNED_TYPE) \
switch (*f) { \
case 'o': len = SPRINT(SIZE_SPEC, "o", UNSIGNED_TYPE); break; \
case 'u': len = SPRINT(SIZE_SPEC, "u", UNSIGNED_TYPE); break; \
case 'x': len = SPRINT(SIZE_SPEC, "x", UNSIGNED_TYPE); break; \
case 'X': len = SPRINT(SIZE_SPEC, "X", UNSIGNED_TYPE); break; \
default: len = SPRINT(SIZE_SPEC, "d", SIGNED_TYPE); break; \
}
// Outer switch to handle all the sizes/types
switch (sizemod) {
case F_LONG: DO_SPRINTS("l", long, unsigned long); break;
case F_LONGLONG: DO_SPRINTS("ll", long long, unsigned long long); break;
case F_SIZE: DO_SPRINTS("z", Py_ssize_t, size_t); break;
case F_PTRDIFF: DO_SPRINTS("t", ptrdiff_t, ptrdiff_t); break;
case F_INTMAX: DO_SPRINTS("j", intmax_t, uintmax_t); break;
default: DO_SPRINTS("", int, unsigned int); break;
}
#undef SPRINT
#undef DO_SPRINTS
assert(len >= 0);
int sign = (buffer[0] == '-');
len -= sign;
precision = Py_MAX(precision, len);
width = Py_MAX(width, precision + sign);
if ((flags & F_ZERO) && !(flags & F_LJUST)) {
precision = width - sign;
}
Py_ssize_t spacepad = Py_MAX(width - precision - sign, 0);
Py_ssize_t zeropad = Py_MAX(precision - len, 0);
if (_PyUnicodeWriter_Prepare(writer, width, 127) == -1)
return NULL;
if (spacepad && !(flags & F_LJUST)) {
if (PyUnicode_Fill(writer->buffer, writer->pos, spacepad, ' ') == -1)
return NULL;
writer->pos += spacepad;
}
if (sign) {
if (_PyUnicodeWriter_WriteChar(writer, '-') == -1)
return NULL;
}
if (zeropad) {
if (PyUnicode_Fill(writer->buffer, writer->pos, zeropad, '0') == -1)
return NULL;
writer->pos += zeropad;
}
if (_PyUnicodeWriter_WriteASCIIString(writer, &buffer[sign], len) < 0)
return NULL;
if (spacepad && (flags & F_LJUST)) {
if (PyUnicode_Fill(writer->buffer, writer->pos, spacepad, ' ') == -1)
return NULL;
writer->pos += spacepad;
}
break;
}
case 'p':
{
char number[MAX_INTMAX_CHARS];
len = sprintf(number, "%p", va_arg(*vargs, void*));
assert(len >= 0);
/* %p is ill-defined: ensure leading 0x. */
if (number[1] == 'X')
number[1] = 'x';
else if (number[1] != 'x') {
memmove(number + 2, number,
strlen(number) + 1);
number[0] = '0';
number[1] = 'x';
len += 2;
}
if (_PyUnicodeWriter_WriteASCIIString(writer, number, len) < 0)
return NULL;
break;
}
case 's':
{
if (sizemod) {
const wchar_t *s = va_arg(*vargs, const wchar_t*);
if (unicode_fromformat_write_wcstr(writer, s, width, precision, flags) < 0)
return NULL;
}
else {
/* UTF-8 */
const char *s = va_arg(*vargs, const char*);
if (unicode_fromformat_write_utf8(writer, s, width, precision, flags) < 0)
return NULL;
}
break;
}
case 'U':
{
PyObject *obj = va_arg(*vargs, PyObject *);
assert(obj && _PyUnicode_CHECK(obj));
if (unicode_fromformat_write_str(writer, obj, width, precision, flags) == -1)
return NULL;
break;
}
case 'V':
{
PyObject *obj = va_arg(*vargs, PyObject *);
const char *str;
const wchar_t *wstr;
if (sizemod) {
wstr = va_arg(*vargs, const wchar_t*);
}
else {
str = va_arg(*vargs, const char *);
}
if (obj) {
assert(_PyUnicode_CHECK(obj));
if (unicode_fromformat_write_str(writer, obj, width, precision, flags) == -1)
return NULL;
}
else if (sizemod) {
assert(wstr != NULL);
if (unicode_fromformat_write_wcstr(writer, wstr, width, precision, flags) < 0)
return NULL;
}
else {
assert(str != NULL);
if (unicode_fromformat_write_utf8(writer, str, width, precision, flags) < 0)
return NULL;
}
break;
}
case 'S':
{
PyObject *obj = va_arg(*vargs, PyObject *);
PyObject *str;
assert(obj);
str = PyObject_Str(obj);
if (!str)
return NULL;
if (unicode_fromformat_write_str(writer, str, width, precision, flags) == -1) {
Py_DECREF(str);
return NULL;
}
Py_DECREF(str);
break;
}
case 'R':
{
PyObject *obj = va_arg(*vargs, PyObject *);
PyObject *repr;
assert(obj);
repr = PyObject_Repr(obj);
if (!repr)
return NULL;
if (unicode_fromformat_write_str(writer, repr, width, precision, flags) == -1) {
Py_DECREF(repr);
return NULL;
}
Py_DECREF(repr);
break;
}
case 'A':
{
PyObject *obj = va_arg(*vargs, PyObject *);
PyObject *ascii;
assert(obj);
ascii = PyObject_ASCII(obj);
if (!ascii)
return NULL;
if (unicode_fromformat_write_str(writer, ascii, width, precision, flags) == -1) {
Py_DECREF(ascii);
return NULL;
}
Py_DECREF(ascii);
break;
}
case 'T':
{
PyObject *obj = va_arg(*vargs, PyObject *);
PyTypeObject *type = (PyTypeObject *)Py_NewRef(Py_TYPE(obj));
PyObject *type_name;
if (flags & F_ALT) {
type_name = _PyType_GetFullyQualifiedName(type, ':');
}
else {
type_name = PyType_GetFullyQualifiedName(type);
}
Py_DECREF(type);
if (!type_name) {
return NULL;
}
if (unicode_fromformat_write_str(writer, type_name,
width, precision, flags) == -1) {
Py_DECREF(type_name);
return NULL;
}
Py_DECREF(type_name);
break;
}
case 'N':
{
PyObject *type_raw = va_arg(*vargs, PyObject *);
assert(type_raw != NULL);
if (!PyType_Check(type_raw)) {
PyErr_SetString(PyExc_TypeError, "%N argument must be a type");
return NULL;
}
PyTypeObject *type = (PyTypeObject*)type_raw;
PyObject *type_name;
if (flags & F_ALT) {
type_name = _PyType_GetFullyQualifiedName(type, ':');
}
else {
type_name = PyType_GetFullyQualifiedName(type);
}
if (!type_name) {
return NULL;
}
if (unicode_fromformat_write_str(writer, type_name,
width, precision, flags) == -1) {
Py_DECREF(type_name);
return NULL;
}
Py_DECREF(type_name);
break;
}
default:
invalid_format:
PyErr_Format(PyExc_SystemError, "invalid format string: %s", p);
return NULL;
}
f++;
return f;
}
static int
unicode_from_format(_PyUnicodeWriter *writer, const char *format, va_list vargs)
{
Py_ssize_t len = strlen(format);
writer->min_length += len + 100;
writer->overallocate = 1;
// Copy varags to be able to pass a reference to a subfunction.
va_list vargs2;
va_copy(vargs2, vargs);
// _PyUnicodeWriter_WriteASCIIString() below requires the format string
// to be encoded to ASCII.
int is_ascii = (ucs1lib_find_max_char((Py_UCS1*)format, (Py_UCS1*)format + len) < 128);
if (!is_ascii) {
Py_ssize_t i;
for (i=0; i < len && (unsigned char)format[i] overallocate = 0;
}
if (_PyUnicodeWriter_WriteASCIIString(writer, f, len) < 0) {
goto fail;
}
f += len;
}
}
va_end(vargs2);
return 0;
fail:
va_end(vargs2);
return -1;
}
PyObject *
PyUnicode_FromFormatV(const char *format, va_list vargs)
{
_PyUnicodeWriter writer;
_PyUnicodeWriter_Init(&writer);
if (unicode_from_format(&writer, format, vargs) < 0) {
_PyUnicodeWriter_Dealloc(&writer);
return NULL;
}
return _PyUnicodeWriter_Finish(&writer);
}
PyObject *
PyUnicode_FromFormat(const char *format, ...)
{
PyObject* ret;
va_list vargs;
va_start(vargs, format);
ret = PyUnicode_FromFormatV(format, vargs);
va_end(vargs);
return ret;
}
int
PyUnicodeWriter_Format(PyUnicodeWriter *writer, const char *format, ...)
{
va_list vargs;
va_start(vargs, format);
int res = _PyUnicodeWriter_FormatV(writer, format, vargs);
va_end(vargs);
return res;
}
int
_PyUnicodeWriter_FormatV(PyUnicodeWriter *writer, const char *format,
va_list vargs)
{
_PyUnicodeWriter *_writer = (_PyUnicodeWriter*)writer;
Py_ssize_t old_pos = _writer->pos;
int res = unicode_from_format(_writer, format, vargs);
if (res < 0) {
_writer->pos = old_pos;
}
return res;
}
static Py_ssize_t
unicode_get_widechar_size(PyObject *unicode)
{
Py_ssize_t res;
assert(unicode != NULL);
assert(_PyUnicode_CHECK(unicode));
res = _PyUnicode_LENGTH(unicode);
#if SIZEOF_WCHAR_T == 2
if (PyUnicode_KIND(unicode) == PyUnicode_4BYTE_KIND) {
const Py_UCS4 *s = PyUnicode_4BYTE_DATA(unicode);
const Py_UCS4 *end = s + res;
for (; s < end; ++s) {
if (*s > 0xFFFF) {
++res;
}
}
}
#endif
return res;
}
static void
unicode_copy_as_widechar(PyObject *unicode, wchar_t *w, Py_ssize_t size)
{
assert(unicode != NULL);
assert(_PyUnicode_CHECK(unicode));
if (PyUnicode_KIND(unicode) == sizeof(wchar_t)) {
memcpy(w, PyUnicode_DATA(unicode), size * sizeof(wchar_t));
return;
}
if (PyUnicode_KIND(unicode) == PyUnicode_1BYTE_KIND) {
const Py_UCS1 *s = PyUnicode_1BYTE_DATA(unicode);
for (; size--; ++s, ++w) {
*w = *s;
}
}
else {
#if SIZEOF_WCHAR_T == 4
assert(PyUnicode_KIND(unicode) == PyUnicode_2BYTE_KIND);
const Py_UCS2 *s = PyUnicode_2BYTE_DATA(unicode);
for (; size--; ++s, ++w) {
*w = *s;
}
#else
assert(PyUnicode_KIND(unicode) == PyUnicode_4BYTE_KIND);
const Py_UCS4 *s = PyUnicode_4BYTE_DATA(unicode);
for (; size--; ++s, ++w) {
Py_UCS4 ch = *s;
if (ch > 0xFFFF) {
assert(ch res) {
size = res + 1;
}
else {
res = size;
}
unicode_copy_as_widechar(unicode, w, size);
#ifdef HAVE_NON_UNICODE_WCHAR_T_REPRESENTATION
/* Oracle Solaris uses non-Unicode internal wchar_t form for
non-Unicode locales and hence needs conversion first. */
if (_Py_LocaleUsesNonUnicodeWchar()) {
if (_Py_EncodeNonUnicodeWchar_InPlace(w, size) < 0) {
return -1;
}
}
#endif
return res;
}
wchar_t*
PyUnicode_AsWideCharString(PyObject *unicode,
Py_ssize_t *size)
{
wchar_t *buffer;
Py_ssize_t buflen;
if (unicode == NULL) {
PyErr_BadInternalCall();
return NULL;
}
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return NULL;
}
buflen = unicode_get_widechar_size(unicode);
buffer = (wchar_t *) PyMem_New(wchar_t, (buflen + 1));
if (buffer == NULL) {
PyErr_NoMemory();
return NULL;
}
unicode_copy_as_widechar(unicode, buffer, buflen + 1);
#ifdef HAVE_NON_UNICODE_WCHAR_T_REPRESENTATION
/* Oracle Solaris uses non-Unicode internal wchar_t form for
non-Unicode locales and hence needs conversion first. */
if (_Py_LocaleUsesNonUnicodeWchar()) {
if (_Py_EncodeNonUnicodeWchar_InPlace(buffer, (buflen + 1)) < 0) {
return NULL;
}
}
#endif
if (size != NULL) {
*size = buflen;
}
else if (wcslen(buffer) != (size_t)buflen) {
PyMem_Free(buffer);
PyErr_SetString(PyExc_ValueError,
"embedded null character");
return NULL;
}
return buffer;
}
#endif /* HAVE_WCHAR_H */
int
_PyUnicode_WideCharString_Converter(PyObject *obj, void *ptr)
{
wchar_t **p = (wchar_t **)ptr;
if (obj == NULL) {
PyMem_Free(*p);
*p = NULL;
return 1;
}
if (PyUnicode_Check(obj)) {
*p = PyUnicode_AsWideCharString(obj, NULL);
if (*p == NULL) {
return 0;
}
return Py_CLEANUP_SUPPORTED;
}
PyErr_Format(PyExc_TypeError,
"argument must be str, not %.50s",
Py_TYPE(obj)->tp_name);
return 0;
}
int
_PyUnicode_WideCharString_Opt_Converter(PyObject *obj, void *ptr)
{
wchar_t **p = (wchar_t **)ptr;
if (obj == NULL) {
PyMem_Free(*p);
*p = NULL;
return 1;
}
if (obj == Py_None) {
*p = NULL;
return 1;
}
if (PyUnicode_Check(obj)) {
*p = PyUnicode_AsWideCharString(obj, NULL);
if (*p == NULL) {
return 0;
}
return Py_CLEANUP_SUPPORTED;
}
PyErr_Format(PyExc_TypeError,
"argument must be str or None, not %.50s",
Py_TYPE(obj)->tp_name);
return 0;
}
PyObject *
PyUnicode_FromOrdinal(int ordinal)
{
if (ordinal < 0 || ordinal > MAX_UNICODE) {
PyErr_SetString(PyExc_ValueError,
"chr() arg not in range(0x110000)");
return NULL;
}
return unicode_char((Py_UCS4)ordinal);
}
PyObject *
PyUnicode_FromObject(PyObject *obj)
{
/* XXX Perhaps we should make this API an alias of
PyObject_Str() instead ?! */
if (PyUnicode_CheckExact(obj)) {
return Py_NewRef(obj);
}
if (PyUnicode_Check(obj)) {
/* For a Unicode subtype that's not a Unicode object,
return a true Unicode object with the same data. */
return _PyUnicode_Copy(obj);
}
PyErr_Format(PyExc_TypeError,
"Can't convert '%.100s' object to str implicitly",
Py_TYPE(obj)->tp_name);
return NULL;
}
PyObject *
PyUnicode_FromEncodedObject(PyObject *obj,
const char *encoding,
const char *errors)
{
Py_buffer buffer;
PyObject *v;
if (obj == NULL) {
PyErr_BadInternalCall();
return NULL;
}
/* Decoding bytes objects is the most common case and should be fast */
if (PyBytes_Check(obj)) {
if (PyBytes_GET_SIZE(obj) == 0) {
if (unicode_check_encoding_errors(encoding, errors) < 0) {
return NULL;
}
_Py_RETURN_UNICODE_EMPTY();
}
return PyUnicode_Decode(
PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
encoding, errors);
}
if (PyUnicode_Check(obj)) {
PyErr_SetString(PyExc_TypeError,
"decoding str is not supported");
return NULL;
}
/* Retrieve a bytes buffer view through the PEP 3118 buffer interface */
if (PyObject_GetBuffer(obj, &buffer, PyBUF_SIMPLE) < 0) {
PyErr_Format(PyExc_TypeError,
"decoding to str: need a bytes-like object, %.80s found",
Py_TYPE(obj)->tp_name);
return NULL;
}
if (buffer.len == 0) {
PyBuffer_Release(&buffer);
if (unicode_check_encoding_errors(encoding, errors) < 0) {
return NULL;
}
_Py_RETURN_UNICODE_EMPTY();
}
v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
PyBuffer_Release(&buffer);
return v;
}
/* Normalize an encoding name like encodings.normalize_encoding()
but allow to convert to lowercase if *to_lower* is true.
Return 1 on success, or 0 on error (encoding is longer than lower_len-1). */
int
_Py_normalize_encoding(const char *encoding,
char *lower,
size_t lower_len,
int to_lower)
{
const char *e;
char *l;
char *l_end;
int punct;
assert(encoding != NULL);
e = encoding;
l = lower;
l_end = &lower[lower_len - 1];
punct = 0;
while (1) {
char c = *e;
if (c == 0) {
break;
}
if (Py_ISALNUM(c) || c == '.') {
if (punct && l != lower) {
if (l == l_end) {
return 0;
}
*l++ = '_';
}
punct = 0;
if (l == l_end) {
return 0;
}
*l++ = to_lower ? Py_TOLOWER(c) : c;
}
else {
punct = 1;
}
e++;
}
*l = '\0';
return 1;
}
PyObject *
PyUnicode_Decode(const char *s,
Py_ssize_t size,
const char *encoding,
const char *errors)
{
PyObject *buffer = NULL, *unicode;
Py_buffer info;
char buflower[11]; /* strlen("iso-8859-1\0") == 11, longest shortcut */
if (unicode_check_encoding_errors(encoding, errors) < 0) {
return NULL;
}
if (size == 0) {
_Py_RETURN_UNICODE_EMPTY();
}
if (encoding == NULL) {
return PyUnicode_DecodeUTF8Stateful(s, size, errors, NULL);
}
/* Shortcuts for common default encodings */
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower), 1)) {
char *lower = buflower;
/* Fast paths */
if (lower[0] == 'u' && lower[1] == 't' && lower[2] == 'f') {
lower += 3;
if (*lower == '_') {
/* Match "utf8" and "utf_8" */
lower++;
}
if (lower[0] == '8' && lower[1] == 0) {
return PyUnicode_DecodeUTF8Stateful(s, size, errors, NULL);
}
else if (lower[0] == '1' && lower[1] == '6' && lower[2] == 0) {
return PyUnicode_DecodeUTF16(s, size, errors, 0);
}
else if (lower[0] == '3' && lower[1] == '2' && lower[2] == 0) {
return PyUnicode_DecodeUTF32(s, size, errors, 0);
}
}
else {
if (strcmp(lower, "ascii") == 0
|| strcmp(lower, "us_ascii") == 0) {
return PyUnicode_DecodeASCII(s, size, errors);
}
#ifdef MS_WINDOWS
else if (strcmp(lower, "mbcs") == 0) {
return PyUnicode_DecodeMBCS(s, size, errors);
}
#endif
else if (strcmp(lower, "latin1") == 0
|| strcmp(lower, "latin_1") == 0
|| strcmp(lower, "iso_8859_1") == 0
|| strcmp(lower, "iso8859_1") == 0) {
return PyUnicode_DecodeLatin1(s, size, errors);
}
}
}
/* Decode via the codec registry */
buffer = NULL;
if (PyBuffer_FillInfo(&info, NULL, (void *)s, size, 1, PyBUF_FULL_RO) < 0)
goto onError;
buffer = PyMemoryView_FromBuffer(&info);
if (buffer == NULL)
goto onError;
unicode = _PyCodec_DecodeText(buffer, encoding, errors);
if (unicode == NULL)
goto onError;
if (!PyUnicode_Check(unicode)) {
PyErr_Format(PyExc_TypeError,
"'%.400s' decoder returned '%.400s' instead of 'str'; "
"use codecs.decode() to decode to arbitrary types",
encoding,
Py_TYPE(unicode)->tp_name);
Py_DECREF(unicode);
goto onError;
}
Py_DECREF(buffer);
return unicode_result(unicode);
onError:
Py_XDECREF(buffer);
return NULL;
}
PyAPI_FUNC(PyObject *)
PyUnicode_AsDecodedObject(PyObject *unicode,
const char *encoding,
const char *errors)
{
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return NULL;
}
if (encoding == NULL)
encoding = PyUnicode_GetDefaultEncoding();
/* Decode via the codec registry */
return PyCodec_Decode(unicode, encoding, errors);
}
PyAPI_FUNC(PyObject *)
PyUnicode_AsDecodedUnicode(PyObject *unicode,
const char *encoding,
const char *errors)
{
PyObject *v;
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
goto onError;
}
if (encoding == NULL)
encoding = PyUnicode_GetDefaultEncoding();
/* Decode via the codec registry */
v = PyCodec_Decode(unicode, encoding, errors);
if (v == NULL)
goto onError;
if (!PyUnicode_Check(v)) {
PyErr_Format(PyExc_TypeError,
"'%.400s' decoder returned '%.400s' instead of 'str'; "
"use codecs.decode() to decode to arbitrary types",
encoding,
Py_TYPE(unicode)->tp_name);
Py_DECREF(v);
goto onError;
}
return unicode_result(v);
onError:
return NULL;
}
PyAPI_FUNC(PyObject *)
PyUnicode_AsEncodedObject(PyObject *unicode,
const char *encoding,
const char *errors)
{
PyObject *v;
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
goto onError;
}
if (encoding == NULL)
encoding = PyUnicode_GetDefaultEncoding();
/* Encode via the codec registry */
v = PyCodec_Encode(unicode, encoding, errors);
if (v == NULL)
goto onError;
return v;
onError:
return NULL;
}
static PyObject *
unicode_encode_locale(PyObject *unicode, _Py_error_handler error_handler,
int current_locale)
{
Py_ssize_t wlen;
wchar_t *wstr = PyUnicode_AsWideCharString(unicode, &wlen);
if (wstr == NULL) {
return NULL;
}
if ((size_t)wlen != wcslen(wstr)) {
PyErr_SetString(PyExc_ValueError, "embedded null character");
PyMem_Free(wstr);
return NULL;
}
char *str;
size_t error_pos;
const char *reason;
int res = _Py_EncodeLocaleEx(wstr, &str, &error_pos, &reason,
current_locale, error_handler);
PyMem_Free(wstr);
if (res != 0) {
if (res == -2) {
PyObject *exc;
exc = PyObject_CallFunction(PyExc_UnicodeEncodeError, "sOnns",
"locale", unicode,
(Py_ssize_t)error_pos,
(Py_ssize_t)(error_pos+1),
reason);
if (exc != NULL) {
PyCodec_StrictErrors(exc);
Py_DECREF(exc);
}
}
else if (res == -3) {
PyErr_SetString(PyExc_ValueError, "unsupported error handler");
}
else {
PyErr_NoMemory();
}
return NULL;
}
PyObject *bytes = PyBytes_FromString(str);
PyMem_RawFree(str);
return bytes;
}
PyObject *
PyUnicode_EncodeLocale(PyObject *unicode, const char *errors)
{
_Py_error_handler error_handler = _Py_GetErrorHandler(errors);
return unicode_encode_locale(unicode, error_handler, 1);
}
PyObject *
PyUnicode_EncodeFSDefault(PyObject *unicode)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
struct _Py_unicode_fs_codec *fs_codec = &interp->unicode.fs_codec;
if (fs_codec->utf8) {
return unicode_encode_utf8(unicode,
fs_codec->error_handler,
fs_codec->errors);
}
#ifndef _Py_FORCE_UTF8_FS_ENCODING
else if (fs_codec->encoding) {
return PyUnicode_AsEncodedString(unicode,
fs_codec->encoding,
fs_codec->errors);
}
#endif
else {
/* Before _PyUnicode_InitEncodings() is called, the Python codec
machinery is not ready and so cannot be used:
use wcstombs() in this case. */
const PyConfig *config = _PyInterpreterState_GetConfig(interp);
const wchar_t *filesystem_errors = config->filesystem_errors;
assert(filesystem_errors != NULL);
_Py_error_handler errors = get_error_handler_wide(filesystem_errors);
assert(errors != _Py_ERROR_UNKNOWN);
#ifdef _Py_FORCE_UTF8_FS_ENCODING
return unicode_encode_utf8(unicode, errors, NULL);
#else
return unicode_encode_locale(unicode, errors, 0);
#endif
}
}
PyObject *
PyUnicode_AsEncodedString(PyObject *unicode,
const char *encoding,
const char *errors)
{
PyObject *v;
char buflower[11]; /* strlen("iso_8859_1\0") == 11, longest shortcut */
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return NULL;
}
if (unicode_check_encoding_errors(encoding, errors) < 0) {
return NULL;
}
if (encoding == NULL) {
return _PyUnicode_AsUTF8String(unicode, errors);
}
/* Shortcuts for common default encodings */
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower), 1)) {
char *lower = buflower;
/* Fast paths */
if (lower[0] == 'u' && lower[1] == 't' && lower[2] == 'f') {
lower += 3;
if (*lower == '_') {
/* Match "utf8" and "utf_8" */
lower++;
}
if (lower[0] == '8' && lower[1] == 0) {
return _PyUnicode_AsUTF8String(unicode, errors);
}
else if (lower[0] == '1' && lower[1] == '6' && lower[2] == 0) {
return _PyUnicode_EncodeUTF16(unicode, errors, 0);
}
else if (lower[0] == '3' && lower[1] == '2' && lower[2] == 0) {
return _PyUnicode_EncodeUTF32(unicode, errors, 0);
}
}
else {
if (strcmp(lower, "ascii") == 0
|| strcmp(lower, "us_ascii") == 0) {
return _PyUnicode_AsASCIIString(unicode, errors);
}
#ifdef MS_WINDOWS
else if (strcmp(lower, "mbcs") == 0) {
return PyUnicode_EncodeCodePage(CP_ACP, unicode, errors);
}
#endif
else if (strcmp(lower, "latin1") == 0 ||
strcmp(lower, "latin_1") == 0 ||
strcmp(lower, "iso_8859_1") == 0 ||
strcmp(lower, "iso8859_1") == 0) {
return _PyUnicode_AsLatin1String(unicode, errors);
}
}
}
/* Encode via the codec registry */
v = _PyCodec_EncodeText(unicode, encoding, errors);
if (v == NULL)
return NULL;
/* The normal path */
if (PyBytes_Check(v))
return v;
/* If the codec returns a buffer, raise a warning and convert to bytes */
if (PyByteArray_Check(v)) {
int error;
PyObject *b;
error = PyErr_WarnFormat(PyExc_RuntimeWarning, 1,
"encoder %s returned bytearray instead of bytes; "
"use codecs.encode() to encode to arbitrary types",
encoding);
if (error) {
Py_DECREF(v);
return NULL;
}
b = PyBytes_FromStringAndSize(PyByteArray_AS_STRING(v),
PyByteArray_GET_SIZE(v));
Py_DECREF(v);
return b;
}
PyErr_Format(PyExc_TypeError,
"'%.400s' encoder returned '%.400s' instead of 'bytes'; "
"use codecs.encode() to encode to arbitrary types",
encoding,
Py_TYPE(v)->tp_name);
Py_DECREF(v);
return NULL;
}
PyAPI_FUNC(PyObject *)
PyUnicode_AsEncodedUnicode(PyObject *unicode,
const char *encoding,
const char *errors)
{
PyObject *v;
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
goto onError;
}
if (encoding == NULL)
encoding = PyUnicode_GetDefaultEncoding();
/* Encode via the codec registry */
v = PyCodec_Encode(unicode, encoding, errors);
if (v == NULL)
goto onError;
if (!PyUnicode_Check(v)) {
PyErr_Format(PyExc_TypeError,
"'%.400s' encoder returned '%.400s' instead of 'str'; "
"use codecs.encode() to encode to arbitrary types",
encoding,
Py_TYPE(v)->tp_name);
Py_DECREF(v);
goto onError;
}
return v;
onError:
return NULL;
}
static PyObject*
unicode_decode_locale(const char *str, Py_ssize_t len,
_Py_error_handler errors, int current_locale)
{
if (str[len] != '\0' || (size_t)len != strlen(str)) {
PyErr_SetString(PyExc_ValueError, "embedded null byte");
return NULL;
}
wchar_t *wstr;
size_t wlen;
const char *reason;
int res = _Py_DecodeLocaleEx(str, &wstr, &wlen, &reason,
current_locale, errors);
if (res != 0) {
if (res == -2) {
PyObject *exc;
exc = PyObject_CallFunction(PyExc_UnicodeDecodeError, "sy#nns",
"locale", str, len,
(Py_ssize_t)wlen,
(Py_ssize_t)(wlen + 1),
reason);
if (exc != NULL) {
PyCodec_StrictErrors(exc);
Py_DECREF(exc);
}
}
else if (res == -3) {
PyErr_SetString(PyExc_ValueError, "unsupported error handler");
}
else {
PyErr_NoMemory();
}
return NULL;
}
PyObject *unicode = PyUnicode_FromWideChar(wstr, wlen);
PyMem_RawFree(wstr);
return unicode;
}
PyObject*
PyUnicode_DecodeLocaleAndSize(const char *str, Py_ssize_t len,
const char *errors)
{
_Py_error_handler error_handler = _Py_GetErrorHandler(errors);
return unicode_decode_locale(str, len, error_handler, 1);
}
PyObject*
PyUnicode_DecodeLocale(const char *str, const char *errors)
{
Py_ssize_t size = (Py_ssize_t)strlen(str);
_Py_error_handler error_handler = _Py_GetErrorHandler(errors);
return unicode_decode_locale(str, size, error_handler, 1);
}
PyObject*
PyUnicode_DecodeFSDefault(const char *s) {
Py_ssize_t size = (Py_ssize_t)strlen(s);
return PyUnicode_DecodeFSDefaultAndSize(s, size);
}
PyObject*
PyUnicode_DecodeFSDefaultAndSize(const char *s, Py_ssize_t size)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
struct _Py_unicode_fs_codec *fs_codec = &interp->unicode.fs_codec;
if (fs_codec->utf8) {
return unicode_decode_utf8(s, size,
fs_codec->error_handler,
fs_codec->errors,
NULL);
}
#ifndef _Py_FORCE_UTF8_FS_ENCODING
else if (fs_codec->encoding) {
return PyUnicode_Decode(s, size,
fs_codec->encoding,
fs_codec->errors);
}
#endif
else {
/* Before _PyUnicode_InitEncodings() is called, the Python codec
machinery is not ready and so cannot be used:
use mbstowcs() in this case. */
const PyConfig *config = _PyInterpreterState_GetConfig(interp);
const wchar_t *filesystem_errors = config->filesystem_errors;
assert(filesystem_errors != NULL);
_Py_error_handler errors = get_error_handler_wide(filesystem_errors);
assert(errors != _Py_ERROR_UNKNOWN);
#ifdef _Py_FORCE_UTF8_FS_ENCODING
return unicode_decode_utf8(s, size, errors, NULL, NULL);
#else
return unicode_decode_locale(s, size, errors, 0);
#endif
}
}
int
PyUnicode_FSConverter(PyObject* arg, void* addr)
{
PyObject *path = NULL;
PyObject *output = NULL;
Py_ssize_t size;
const char *data;
if (arg == NULL) {
Py_DECREF(*(PyObject**)addr);
*(PyObject**)addr = NULL;
return 1;
}
path = PyOS_FSPath(arg);
if (path == NULL) {
return 0;
}
if (PyBytes_Check(path)) {
output = path;
}
else { // PyOS_FSPath() guarantees its returned value is bytes or str.
output = PyUnicode_EncodeFSDefault(path);
Py_DECREF(path);
if (!output) {
return 0;
}
assert(PyBytes_Check(output));
}
size = PyBytes_GET_SIZE(output);
data = PyBytes_AS_STRING(output);
if ((size_t)size != strlen(data)) {
PyErr_SetString(PyExc_ValueError, "embedded null byte");
Py_DECREF(output);
return 0;
}
*(PyObject**)addr = output;
return Py_CLEANUP_SUPPORTED;
}
int
PyUnicode_FSDecoder(PyObject* arg, void* addr)
{
if (arg == NULL) {
Py_DECREF(*(PyObject**)addr);
*(PyObject**)addr = NULL;
return 1;
}
PyObject *path = PyOS_FSPath(arg);
if (path == NULL) {
return 0;
}
PyObject *output = NULL;
if (PyUnicode_Check(path)) {
output = path;
}
else if (PyBytes_Check(path)) {
output = PyUnicode_DecodeFSDefaultAndSize(PyBytes_AS_STRING(path),
PyBytes_GET_SIZE(path));
Py_DECREF(path);
if (!output) {
return 0;
}
}
else {
PyErr_Format(PyExc_TypeError,
"path should be string, bytes, or os.PathLike, not %.200s",
Py_TYPE(arg)->tp_name);
Py_DECREF(path);
return 0;
}
if (findchar(PyUnicode_DATA(output), PyUnicode_KIND(output),
PyUnicode_GET_LENGTH(output), 0, 1) >= 0) {
PyErr_SetString(PyExc_ValueError, "embedded null character");
Py_DECREF(output);
return 0;
}
*(PyObject**)addr = output;
return Py_CLEANUP_SUPPORTED;
}
static int unicode_fill_utf8(PyObject *unicode);
static int
unicode_ensure_utf8(PyObject *unicode)
{
int err = 0;
if (PyUnicode_UTF8(unicode) == NULL) {
Py_BEGIN_CRITICAL_SECTION(unicode);
if (PyUnicode_UTF8(unicode) == NULL) {
err = unicode_fill_utf8(unicode);
}
Py_END_CRITICAL_SECTION();
}
return err;
}
const char *
PyUnicode_AsUTF8AndSize(PyObject *unicode, Py_ssize_t *psize)
{
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
if (psize) {
*psize = -1;
}
return NULL;
}
if (unicode_ensure_utf8(unicode) == -1) {
if (psize) {
*psize = -1;
}
return NULL;
}
if (psize) {
*psize = PyUnicode_UTF8_LENGTH(unicode);
}
return PyUnicode_UTF8(unicode);
}
const char *
PyUnicode_AsUTF8(PyObject *unicode)
{
return PyUnicode_AsUTF8AndSize(unicode, NULL);
}
const char *
_PyUnicode_AsUTF8NoNUL(PyObject *unicode)
{
Py_ssize_t size;
const char *s = PyUnicode_AsUTF8AndSize(unicode, &size);
if (s && strlen(s) != (size_t)size) {
PyErr_SetString(PyExc_ValueError, "embedded null character");
return NULL;
}
return s;
}
/*
PyUnicode_GetSize() has been deprecated since Python 3.3
because it returned length of Py_UNICODE.
But this function is part of stable abi, because it doesn't
include Py_UNICODE in signature and it was not excluded from
stable ABI in PEP 384.
*/
PyAPI_FUNC(Py_ssize_t)
PyUnicode_GetSize(PyObject *unicode)
{
PyErr_SetString(PyExc_RuntimeError,
"PyUnicode_GetSize has been removed.");
return -1;
}
Py_ssize_t
PyUnicode_GetLength(PyObject *unicode)
{
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return -1;
}
return PyUnicode_GET_LENGTH(unicode);
}
Py_UCS4
PyUnicode_ReadChar(PyObject *unicode, Py_ssize_t index)
{
const void *data;
int kind;
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return (Py_UCS4)-1;
}
if (index < 0 || index >= PyUnicode_GET_LENGTH(unicode)) {
PyErr_SetString(PyExc_IndexError, "string index out of range");
return (Py_UCS4)-1;
}
data = PyUnicode_DATA(unicode);
kind = PyUnicode_KIND(unicode);
return PyUnicode_READ(kind, data, index);
}
int
PyUnicode_WriteChar(PyObject *unicode, Py_ssize_t index, Py_UCS4 ch)
{
if (!PyUnicode_Check(unicode) || !PyUnicode_IS_COMPACT(unicode)) {
PyErr_BadArgument();
return -1;
}
if (index < 0 || index >= PyUnicode_GET_LENGTH(unicode)) {
PyErr_SetString(PyExc_IndexError, "string index out of range");
return -1;
}
if (unicode_check_modifiable(unicode))
return -1;
if (ch > PyUnicode_MAX_CHAR_VALUE(unicode)) {
PyErr_SetString(PyExc_ValueError, "character out of range");
return -1;
}
PyUnicode_WRITE(PyUnicode_KIND(unicode), PyUnicode_DATA(unicode),
index, ch);
return 0;
}
const char *
PyUnicode_GetDefaultEncoding(void)
{
return "utf-8";
}
/* create or adjust a UnicodeDecodeError */
static void
make_decode_exception(PyObject **exceptionObject,
const char *encoding,
const char *input, Py_ssize_t length,
Py_ssize_t startpos, Py_ssize_t endpos,
const char *reason)
{
if (*exceptionObject == NULL) {
*exceptionObject = PyUnicodeDecodeError_Create(
encoding, input, length, startpos, endpos, reason);
}
else {
if (PyUnicodeDecodeError_SetStart(*exceptionObject, startpos))
goto onError;
if (PyUnicodeDecodeError_SetEnd(*exceptionObject, endpos))
goto onError;
if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
goto onError;
}
return;
onError:
Py_CLEAR(*exceptionObject);
}
#ifdef MS_WINDOWS
static int
widechar_resize(wchar_t **buf, Py_ssize_t *size, Py_ssize_t newsize)
{
if (newsize > *size) {
wchar_t *newbuf = *buf;
if (PyMem_Resize(newbuf, wchar_t, newsize) == NULL) {
PyErr_NoMemory();
return -1;
}
*buf = newbuf;
}
*size = newsize;
return 0;
}
/* error handling callback helper:
build arguments, call the callback and check the arguments,
if no exception occurred, copy the replacement to the output
and adjust various state variables.
return 0 on success, -1 on error
*/
static int
unicode_decode_call_errorhandler_wchar(
const char *errors, PyObject **errorHandler,
const char *encoding, const char *reason,
const char **input, const char **inend, Py_ssize_t *startinpos,
Py_ssize_t *endinpos, PyObject **exceptionObject, const char **inptr,
wchar_t **buf, Py_ssize_t *bufsize, Py_ssize_t *outpos)
{
static const char *argparse = "Un;decoding error handler must return (str, int) tuple";
PyObject *restuple = NULL;
PyObject *repunicode = NULL;
Py_ssize_t outsize;
Py_ssize_t insize;
Py_ssize_t requiredsize;
Py_ssize_t newpos;
PyObject *inputobj = NULL;
Py_ssize_t repwlen;
if (*errorHandler == NULL) {
*errorHandler = PyCodec_LookupError(errors);
if (*errorHandler == NULL)
goto onError;
}
make_decode_exception(exceptionObject,
encoding,
*input, *inend - *input,
*startinpos, *endinpos,
reason);
if (*exceptionObject == NULL)
goto onError;
restuple = PyObject_CallOneArg(*errorHandler, *exceptionObject);
if (restuple == NULL)
goto onError;
if (!PyTuple_Check(restuple)) {
PyErr_SetString(PyExc_TypeError, &argparse[3]);
goto onError;
}
if (!PyArg_ParseTuple(restuple, argparse, &repunicode, &newpos))
goto onError;
/* Copy back the bytes variables, which might have been modified by the
callback */
inputobj = PyUnicodeDecodeError_GetObject(*exceptionObject);
if (!inputobj)
goto onError;
*input = PyBytes_AS_STRING(inputobj);
insize = PyBytes_GET_SIZE(inputobj);
*inend = *input + insize;
/* we can DECREF safely, as the exception has another reference,
so the object won't go away. */
Py_DECREF(inputobj);
if (newposmin_length += replen - 1;
need_to_grow = 1;
}
new_inptr = *input + newpos;
if (*inend - new_inptr > remain) {
/* We don't know the decoding algorithm here so we make the worst
assumption that one byte decodes to one unicode character.
If unfortunately one byte could decode to more unicode characters,
the decoder may write out-of-bound then. Is it possible for the
algorithms using this function? */
writer->min_length += *inend - new_inptr - remain;
need_to_grow = 1;
}
if (need_to_grow) {
writer->overallocate = 1;
if (_PyUnicodeWriter_Prepare(writer, writer->min_length - writer->pos,
PyUnicode_MAX_CHAR_VALUE(repunicode)) == -1)
goto onError;
}
if (_PyUnicodeWriter_WriteStr(writer, repunicode) == -1)
goto onError;
*endinpos = newpos;
*inptr = new_inptr;
/* we made it! */
Py_DECREF(restuple);
return 0;
onError:
Py_XDECREF(restuple);
return -1;
}
/* --- UTF-7 Codec -------------------------------------------------------- */
/* See RFC2152 for details. We encode conservatively and decode liberally. */
/* Three simple macros defining base-64. */
/* Is c a base-64 character? */
#define IS_BASE64(c) \
(((c) >= 'A' && (c) = 'a' && (c) = '0' && (c) = 'A' && (c) = 'a' && (c) = '0' && (c) ? */
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0,
/* @ A B C D E F G H I J K L M N O */
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* P Q R S T U V W X Y Z [ \ ] ^ _ */
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3, 1, 1, 1,
/* ` a b c d e f g h i j k l m n o */
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* p q r s t u v w x y z { | } ~ del */
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 3, 3,
};
/* ENCODE_DIRECT: this character should be encoded as itself. The
* answer depends on whether we are encoding set O as itself, and also
* on whether we are encoding whitespace as itself. RFC 2152 makes it
* clear that the answers to these questions vary between
* applications, so this code needs to be flexible. */
#define ENCODE_DIRECT(c) \
((c) < 128 && (c) > 0 && ((utf7_category[(c)] != 3)))
PyObject *
PyUnicode_DecodeUTF7(const char *s,
Py_ssize_t size,
const char *errors)
{
return PyUnicode_DecodeUTF7Stateful(s, size, errors, NULL);
}
/* The decoder. The only state we preserve is our read position,
* i.e. how many characters we have consumed. So if we end in the
* middle of a shift sequence we have to back off the read position
* and the output to the beginning of the sequence, otherwise we lose
* all the shift state (seen bits, number of bits seen, high
* surrogate). */
PyObject *
PyUnicode_DecodeUTF7Stateful(const char *s,
Py_ssize_t size,
const char *errors,
Py_ssize_t *consumed)
{
const char *starts = s;
Py_ssize_t startinpos;
Py_ssize_t endinpos;
const char *e;
_PyUnicodeWriter writer;
const char *errmsg = "";
int inShift = 0;
Py_ssize_t shiftOutStart;
unsigned int base64bits = 0;
unsigned long base64buffer = 0;
Py_UCS4 surrogate = 0;
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
if (size == 0) {
if (consumed)
*consumed = 0;
_Py_RETURN_UNICODE_EMPTY();
}
/* Start off assuming it's all ASCII. Widen later as necessary. */
_PyUnicodeWriter_Init(&writer);
writer.min_length = size;
shiftOutStart = 0;
e = s + size;
while (s < e) {
Py_UCS4 ch;
restart:
ch = (unsigned char) *s;
if (inShift) { /* in a base-64 section */
if (IS_BASE64(ch)) { /* consume a base-64 character */
base64buffer = (base64buffer = 16) {
/* we have enough bits for a UTF-16 value */
Py_UCS4 outCh = (Py_UCS4)(base64buffer >> (base64bits-16));
base64bits -= 16;
base64buffer &= (1 = 6) {
/* We've seen at least one base-64 character */
s++;
errmsg = "partial character in shift sequence";
goto utf7Error;
}
else {
/* Some bits remain; they should be zero */
if (base64buffer != 0) {
s++;
errmsg = "non-zero padding bits in shift sequence";
goto utf7Error;
}
}
}
if (surrogate && DECODE_DIRECT(ch)) {
if (_PyUnicodeWriter_WriteCharInline(&writer, surrogate) < 0)
goto onError;
}
surrogate = 0;
if (ch == '-') {
/* '-' is absorbed; other terminating
characters are preserved */
s++;
}
}
}
else if ( ch == '+' ) {
startinpos = s-starts;
s++; /* consume '+' */
if (s < e && *s == '-') { /* '+-' encodes '+' */
s++;
if (_PyUnicodeWriter_WriteCharInline(&writer, '+') < 0)
goto onError;
}
else if (s < e && !IS_BASE64(*s)) {
s++;
errmsg = "ill-formed sequence";
goto utf7Error;
}
else { /* begin base64-encoded section */
inShift = 1;
surrogate = 0;
shiftOutStart = writer.pos;
base64bits = 0;
base64buffer = 0;
}
}
else if (DECODE_DIRECT(ch)) { /* character decodes as itself */
s++;
if (_PyUnicodeWriter_WriteCharInline(&writer, ch) < 0)
goto onError;
}
else {
startinpos = s-starts;
s++;
errmsg = "unexpected special character";
goto utf7Error;
}
continue;
utf7Error:
endinpos = s-starts;
if (unicode_decode_call_errorhandler_writer(
errors, &errorHandler,
"utf7", errmsg,
&starts, &e, &startinpos, &endinpos, &exc, &s,
&writer))
goto onError;
}
/* end of string */
if (inShift && !consumed) { /* in shift sequence, no more to follow */
/* if we're in an inconsistent state, that's an error */
inShift = 0;
if (surrogate ||
(base64bits >= 6) ||
(base64bits > 0 && base64buffer != 0)) {
endinpos = size;
if (unicode_decode_call_errorhandler_writer(
errors, &errorHandler,
"utf7", "unterminated shift sequence",
&starts, &e, &startinpos, &endinpos, &exc, &s,
&writer))
goto onError;
if (s < e)
goto restart;
}
}
/* return state */
if (consumed) {
if (inShift) {
*consumed = startinpos;
if (writer.pos != shiftOutStart && writer.maxchar > 127) {
PyObject *result = PyUnicode_FromKindAndData(
writer.kind, writer.data, shiftOutStart);
Py_XDECREF(errorHandler);
Py_XDECREF(exc);
_PyUnicodeWriter_Dealloc(&writer);
return result;
}
writer.pos = shiftOutStart; /* back off output */
}
else {
*consumed = s-starts;
}
}
Py_XDECREF(errorHandler);
Py_XDECREF(exc);
return _PyUnicodeWriter_Finish(&writer);
onError:
Py_XDECREF(errorHandler);
Py_XDECREF(exc);
_PyUnicodeWriter_Dealloc(&writer);
return NULL;
}
PyObject *
_PyUnicode_EncodeUTF7(PyObject *str,
const char *errors)
{
Py_ssize_t len = PyUnicode_GET_LENGTH(str);
if (len == 0) {
return Py_GetConstant(Py_CONSTANT_EMPTY_BYTES);
}
int kind = PyUnicode_KIND(str);
const void *data = PyUnicode_DATA(str);
/* It might be possible to tighten this worst case */
if (len > PY_SSIZE_T_MAX / 8) {
return PyErr_NoMemory();
}
PyBytesWriter *writer = PyBytesWriter_Create(len * 8);
if (writer == NULL) {
return NULL;
}
int inShift = 0;
unsigned int base64bits = 0;
unsigned long base64buffer = 0;
char *out = PyBytesWriter_GetData(writer);
for (Py_ssize_t i = 0; i < len; ++i) {
Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (inShift) {
if (ENCODE_DIRECT(ch)) {
/* shifting out */
if (base64bits) { /* output remaining bits */
*out++ = TO_BASE64(base64buffer = 0x10000) {
assert(ch > (base64bits-6));
base64bits -= 6;
}
/* prepare second surrogate */
ch = Py_UNICODE_LOW_SURROGATE(ch);
}
base64bits += 16;
base64buffer = (base64buffer = 6) {
*out++ = TO_BASE64(base64buffer >> (base64bits-6));
base64bits -= 6;
}
}
if (base64bits)
*out++= TO_BASE64(base64buffer data, writer->pos,
ch + 0xdc00);
writer->pos++;
}
s += (endinpos - startinpos);
break;
}
default:
if (unicode_decode_call_errorhandler_writer(
errors, &error_handler_obj,
"utf-8", errmsg,
&starts, &end, &startinpos, &endinpos, &exc, &s,
writer)) {
goto onError;
}
if (_PyUnicodeWriter_Prepare(writer, end - s, 127) < 0) {
goto onError;
}
}
}
End:
if (consumed)
*consumed = s - starts;
Py_XDECREF(error_handler_obj);
Py_XDECREF(exc);
return 0;
onError:
Py_XDECREF(error_handler_obj);
Py_XDECREF(exc);
return -1;
}
static PyObject *
unicode_decode_utf8(const char *s, Py_ssize_t size,
_Py_error_handler error_handler, const char *errors,
Py_ssize_t *consumed)
{
if (size == 0) {
if (consumed) {
*consumed = 0;
}
_Py_RETURN_UNICODE_EMPTY();
}
/* ASCII is equivalent to the first 128 ordinals in Unicode. */
if (size == 1 && (unsigned char)s[0] < 128) {
if (consumed) {
*consumed = 1;
}
return get_latin1_char((unsigned char)s[0]);
}
// I don't know this check is necessary or not. But there is a test
// case that requires size=PY_SSIZE_T_MAX cause MemoryError.
if (PY_SSIZE_T_MAX - sizeof(PyCompactUnicodeObject) < (size_t)size) {
PyErr_NoMemory();
return NULL;
}
const char *starts = s;
const char *end = s + size;
Py_ssize_t pos = find_first_nonascii((const unsigned char*)starts, (const unsigned char*)end);
if (pos == size) { // fast path: ASCII string.
PyObject *u = PyUnicode_New(size, 127);
if (u == NULL) {
return NULL;
}
memcpy(PyUnicode_1BYTE_DATA(u), s, size);
if (consumed) {
*consumed = size;
}
return u;
}
int maxchr = 127;
Py_ssize_t maxsize = size;
unsigned char ch = (unsigned char)(s[pos]);
// error handler other than strict may remove/replace the invalid byte.
// consumed != NULL allows 1~3 bytes remainings.
// 0x80 = 0xc2) {
// we only calculate the number of codepoints and don't determine the exact maxchr.
// This is because writing fast and portable SIMD code to find maxchr is difficult.
// If reallocation occurs for a larger maxchar, knowing the exact number of codepoints
// means that it is no longer necessary to allocate several times the required amount
// of memory.
maxsize = utf8_count_codepoints((const unsigned char *)s, (const unsigned char *)end);
if (ch < 0xc4) { // latin1
maxchr = 0xff;
}
else if (ch < 0xf0) { // ucs2
maxchr = 0xffff;
}
else { // ucs4
maxchr = 0x10ffff;
}
}
PyObject *u = PyUnicode_New(maxsize, maxchr);
if (!u) {
return NULL;
}
// Use _PyUnicodeWriter after fast path is failed.
_PyUnicodeWriter writer;
_PyUnicodeWriter_InitWithBuffer(&writer, u);
if (maxchr data + writer->pos * writer->kind;
if (writer->kind == PyUnicode_1BYTE_KIND) {
decoded = ascii_decode(s, end, dest);
writer->pos += decoded;
if (decoded == size) {
if (consumed) {
*consumed = size;
}
return 0;
}
s += decoded;
}
return unicode_decode_utf8_impl(writer, starts, s, end,
error_handler, errors, consumed);
}
PyObject *
PyUnicode_DecodeUTF8Stateful(const char *s,
Py_ssize_t size,
const char *errors,
Py_ssize_t *consumed)
{
return unicode_decode_utf8(s, size,
errors ? _Py_ERROR_UNKNOWN : _Py_ERROR_STRICT,
errors, consumed);
}
/* UTF-8 decoder: use surrogateescape error handler if 'surrogateescape' is
non-zero, use strict error handler otherwise.
On success, write a pointer to a newly allocated wide character string into
*wstr (use PyMem_RawFree() to free the memory) and write the output length
(in number of wchar_t units) into *wlen (if wlen is set).
On memory allocation failure, return -1.
On decoding error (if surrogateescape is zero), return -2. If wlen is
non-NULL, write the start of the illegal byte sequence into *wlen. If reason
is not NULL, write the decoding error message into *reason. */
int
_Py_DecodeUTF8Ex(const char *s, Py_ssize_t size, wchar_t **wstr, size_t *wlen,
const char **reason, _Py_error_handler errors)
{
const char *orig_s = s;
const char *e;
wchar_t *unicode;
Py_ssize_t outpos;
int surrogateescape = 0;
int surrogatepass = 0;
switch (errors)
{
case _Py_ERROR_STRICT:
break;
case _Py_ERROR_SURROGATEESCAPE:
surrogateescape = 1;
break;
case _Py_ERROR_SURROGATEPASS:
surrogatepass = 1;
break;
default:
return -3;
}
/* Note: size will always be longer than the resulting Unicode
character count */
if (PY_SSIZE_T_MAX / (Py_ssize_t)sizeof(wchar_t) - 1 < size) {
return -1;
}
unicode = PyMem_RawMalloc((size + 1) * sizeof(wchar_t));
if (!unicode) {
return -1;
}
/* Unpack UTF-8 encoded data */
e = s + size;
outpos = 0;
while (s < e) {
Py_UCS4 ch;
#if SIZEOF_WCHAR_T == 4
ch = ucs4lib_utf8_decode(&s, e, (Py_UCS4 *)unicode, &outpos);
#else
ch = ucs2lib_utf8_decode(&s, e, (Py_UCS2 *)unicode, &outpos);
#endif
if (ch > 0xFF) {
#if SIZEOF_WCHAR_T == 4
Py_UNREACHABLE();
#else
assert(ch > 0xFFFF && ch = 3
&& (s[0] & 0xf0) == 0xe0
&& (s[1] & 0xc0) == 0x80
&& (s[2] & 0xc0) == 0x80)
{
ch = ((s[0] & 0x0f) PY_SSIZE_T_MAX / max_char_size - 1) {
return -1;
}
char *bytes;
if (raw_malloc) {
bytes = PyMem_RawMalloc((len + 1) * max_char_size);
}
else {
bytes = PyMem_Malloc((len + 1) * max_char_size);
}
if (bytes == NULL) {
return -1;
}
char *p = bytes;
Py_ssize_t i;
for (i = 0; i < len; ) {
Py_ssize_t ch_pos = i;
Py_UCS4 ch = text[i];
i++;
if (sizeof(wchar_t) == 2
&& Py_UNICODE_IS_HIGH_SURROGATE(ch)
&& i < len
&& Py_UNICODE_IS_LOW_SURROGATE(text[i]))
{
ch = Py_UNICODE_JOIN_SURROGATES(ch, text[i]);
i++;
}
if (ch < 0x80) {
/* Encode ASCII */
*p++ = (char) ch;
}
else if (ch < 0x0800) {
/* Encode Latin-1 */
*p++ = (char)(0xc0 | (ch >> 6));
*p++ = (char)(0x80 | (ch & 0x3f));
}
else if (Py_UNICODE_IS_SURROGATE(ch) && !surrogatepass) {
/* surrogateescape error handler */
if (!surrogateescape || !(0xDC80 12));
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*p++ = (char)(0x80 | (ch & 0x3f));
}
else { /* ch >= 0x10000 */
assert(ch > 18));
*p++ = (char)(0x80 | ((ch >> 12) & 0x3f));
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*p++ = (char)(0x80 | (ch & 0x3f));
}
}
*p++ = '\0';
size_t final_size = (p - bytes);
char *bytes2;
if (raw_malloc) {
bytes2 = PyMem_RawRealloc(bytes, final_size);
}
else {
bytes2 = PyMem_Realloc(bytes, final_size);
}
if (bytes2 == NULL) {
if (error_pos != NULL) {
*error_pos = (size_t)-1;
}
if (raw_malloc) {
PyMem_RawFree(bytes);
}
else {
PyMem_Free(bytes);
}
return -1;
}
*str = bytes2;
return 0;
}
/* Primary internal function which creates utf8 encoded bytes objects.
Allocation strategy: if the string is short, convert into a stack buffer
and allocate exactly as much space needed at the end. Else allocate the
maximum possible needed (4 result bytes per Unicode character), and return
the excess memory at the end.
*/
static PyObject *
unicode_encode_utf8(PyObject *unicode, _Py_error_handler error_handler,
const char *errors)
{
if (!PyUnicode_Check(unicode)) {
PyErr_BadArgument();
return NULL;
}
if (PyUnicode_UTF8(unicode))
return PyBytes_FromStringAndSize(PyUnicode_UTF8(unicode),
PyUnicode_UTF8_LENGTH(unicode));
int kind = PyUnicode_KIND(unicode);
const void *data = PyUnicode_DATA(unicode);
Py_ssize_t size = PyUnicode_GET_LENGTH(unicode);
PyBytesWriter *writer;
char *end;
switch (kind) {
default:
Py_UNREACHABLE();
case PyUnicode_1BYTE_KIND:
/* the string cannot be ASCII, or PyUnicode_UTF8() would be set */
assert(!PyUnicode_IS_ASCII(unicode));
writer = ucs1lib_utf8_encoder(unicode, data, size,
error_handler, errors, &end);
break;
case PyUnicode_2BYTE_KIND:
writer = ucs2lib_utf8_encoder(unicode, data, size,
error_handler, errors, &end);
break;
case PyUnicode_4BYTE_KIND:
writer = ucs4lib_utf8_encoder(unicode, data, size,
error_handler, errors, &end);
break;
}
if (writer == NULL) {
PyBytesWriter_Discard(writer);
return NULL;
}
return PyBytesWriter_FinishWithPointer(writer, end);
}
static int
unicode_fill_utf8(PyObject *unicode)
{
_Py_CRITICAL_SECTION_ASSERT_OBJECT_LOCKED(unicode);
/* the string cannot be ASCII, or PyUnicode_UTF8() would be set */
assert(!PyUnicode_IS_ASCII(unicode));
int kind = PyUnicode_KIND(unicode);
const void *data = PyUnicode_DATA(unicode);
Py_ssize_t size = PyUnicode_GET_LENGTH(unicode);
PyBytesWriter *writer;
char *end;
switch (kind) {
default:
Py_UNREACHABLE();
case PyUnicode_1BYTE_KIND:
writer = ucs1lib_utf8_encoder(unicode, data, size,
_Py_ERROR_STRICT, NULL, &end);
break;
case PyUnicode_2BYTE_KIND:
writer = ucs2lib_utf8_encoder(unicode, data, size,
_Py_ERROR_STRICT, NULL, &end);
break;
case PyUnicode_4BYTE_KIND:
writer = ucs4lib_utf8_encoder(unicode, data, size,
_Py_ERROR_STRICT, NULL, &end);
break;
}
if (writer == NULL) {
return -1;
}
const char *start = PyBytesWriter_GetData(writer);
Py_ssize_t len = end - start;
char *cache = PyMem_Malloc(len + 1);
if (cache == NULL) {
PyBytesWriter_Discard(writer);
PyErr_NoMemory();
return -1;
}
memcpy(cache, start, len);
cache[len] = '\0';
PyUnicode_SET_UTF8_LENGTH(unicode, len);
PyUnicode_SET_UTF8(unicode, cache);
PyBytesWriter_Discard(writer);
return 0;
}
PyObject *
_PyUnicode_AsUTF8String(PyObject *unicode, const char *errors)
{
return unicode_encode_utf8(unicode, _Py_ERROR_UNKNOWN, errors);
}
PyObject *
PyUnicode_AsUTF8String(PyObject *unicode)
{
return _PyUnicode_AsUTF8String(unicode, NULL);
}
/* --- UTF-32 Codec ------------------------------------------------------- */
PyObject *
PyUnicode_DecodeUTF32(const char *s,
Py_ssize_t size,
const char *errors,
int *byteorder)
{
return PyUnicode_DecodeUTF32Stateful(s, size, errors, byteorder, NULL);
}
PyObject *
PyUnicode_DecodeUTF32Stateful(const char *s,
Py_ssize_t size,
const char *errors,
int *byteorder,
Py_ssize_t *consumed)
{
const char *starts = s;
Py_ssize_t startinpos;
Py_ssize_t endinpos;
_PyUnicodeWriter writer;
const unsigned char *q, *e;
int le, bo = 0; /* assume native ordering by default */
const char *encoding;
const char *errmsg = "";
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
q = (const unsigned char *)s;
e = q + size;
if (byteorder)
bo = *byteorder;
/* Check for BOM marks (U+FEFF) in the input and adjust current
byte order setting accordingly. In native mode, the leading BOM
mark is skipped, in all other modes, it is copied to the output
stream as-is (giving a ZWNBSP character). */
if (bo == 0 && size >= 4) {
Py_UCS4 bom = ((unsigned int)q[3] 0) {
ucs1lib_utf16_encode((const Py_UCS1 *)data, len, &out, native_ordering);
}
return v;
}
PyBytesWriter *writer = PyBytesWriter_Create(nsize * 2);
if (writer == NULL) {
return NULL;
}
/* output buffer is 2-bytes aligned */
assert(_Py_IS_ALIGNED(PyBytesWriter_GetData(writer), 2));
unsigned short *out = PyBytesWriter_GetData(writer);
if (byteorder == 0) {
*out++ = 0xFEFF;
}
if (len == 0) {
return PyBytesWriter_Finish(writer);
}
const char *encoding;
if (byteorder < 0) {
encoding = "utf-16-le";
}
else if (byteorder > 0) {
encoding = "utf-16-be";
}
else {
encoding = "utf-16";
}
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
PyObject *rep = NULL;
for (Py_ssize_t pos = 0; pos < len; ) {
if (kind == PyUnicode_2BYTE_KIND) {
pos += ucs2lib_utf16_encode((const Py_UCS2 *)data + pos, len - pos,
&out, native_ordering);
}
else {
assert(kind == PyUnicode_4BYTE_KIND);
pos += ucs4lib_utf16_encode((const Py_UCS4 *)data + pos, len - pos,
&out, native_ordering);
}
if (pos == len)
break;
Py_ssize_t newpos;
rep = unicode_encode_call_errorhandler(
errors, &errorHandler,
encoding, "surrogates not allowed",
str, &exc, pos, pos + 1, &newpos);
if (!rep)
goto error;
Py_ssize_t repsize, moreunits;
if (PyBytes_Check(rep)) {
repsize = PyBytes_GET_SIZE(rep);
if (repsize & 1) {
raise_encode_exception(&exc, encoding,
str, pos, pos + 1,
"surrogates not allowed");
goto error;
}
moreunits = repsize / 2;
}
else {
assert(PyUnicode_Check(rep));
moreunits = repsize = PyUnicode_GET_LENGTH(rep);
if (!PyUnicode_IS_ASCII(rep)) {
raise_encode_exception(&exc, encoding,
str, pos, pos + 1,
"surrogates not allowed");
goto error;
}
}
moreunits += pos - newpos;
pos = newpos;
/* two bytes are reserved for each surrogate */
if (moreunits > 0) {
out = PyBytesWriter_GrowAndUpdatePointer(writer, 2 * moreunits, out);
if (out == NULL) {
goto error;
}
}
if (PyBytes_Check(rep)) {
memcpy(out, PyBytes_AS_STRING(rep), repsize);
out += repsize / 2;
} else {
/* rep is unicode */
assert(PyUnicode_KIND(rep) == PyUnicode_1BYTE_KIND);
ucs1lib_utf16_encode(PyUnicode_1BYTE_DATA(rep), repsize,
&out, native_ordering);
}
Py_CLEAR(rep);
}
Py_XDECREF(errorHandler);
Py_XDECREF(exc);
/* Cut back to size actually needed. This is necessary for, for example,
encoding of a string containing isolated surrogates and the 'ignore' handler
is used. */
return PyBytesWriter_FinishWithPointer(writer, out);
error:
Py_XDECREF(rep);
Py_XDECREF(errorHandler);
Py_XDECREF(exc);
PyBytesWriter_Discard(writer);
return NULL;
}
PyObject *
PyUnicode_AsUTF16String(PyObject *unicode)
{
return _PyUnicode_EncodeUTF16(unicode, NULL, 0);
}
_PyUnicode_Name_CAPI *
_PyUnicode_GetNameCAPI(void)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
_PyUnicode_Name_CAPI *ucnhash_capi;
ucnhash_capi = _Py_atomic_load_ptr(&interp->unicode.ucnhash_capi);
if (ucnhash_capi == NULL) {
ucnhash_capi = (_PyUnicode_Name_CAPI *)PyCapsule_Import(
PyUnicodeData_CAPSULE_NAME, 1);
// It's fine if we overwrite the value here. It's always the same value.
_Py_atomic_store_ptr(&interp->unicode.ucnhash_capi, ucnhash_capi);
}
return ucnhash_capi;
}
/* --- Unicode Escape Codec ----------------------------------------------- */
PyObject *
_PyUnicode_DecodeUnicodeEscapeInternal2(const char *s,
Py_ssize_t size,
const char *errors,
Py_ssize_t *consumed,
int *first_invalid_escape_char,
const char **first_invalid_escape_ptr)
{
const char *starts = s;
const char *initial_starts = starts;
_PyUnicodeWriter writer;
const char *end;
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
_PyUnicode_Name_CAPI *ucnhash_capi;
// so we can remember if we've seen an invalid escape char or not
*first_invalid_escape_char = -1;
*first_invalid_escape_ptr = NULL;
if (size == 0) {
if (consumed) {
*consumed = 0;
}
_Py_RETURN_UNICODE_EMPTY();
}
/* Escaped strings will always be longer than the resulting
Unicode string, so we start with size here and then reduce the
length after conversion to the true value.
(but if the error callback returns a long replacement string
we'll have to allocate more space) */
_PyUnicodeWriter_Init(&writer);
writer.min_length = size;
if (_PyUnicodeWriter_Prepare(&writer, size, 127) < 0) {
goto onError;
}
end = s + size;
while (s < end) {
unsigned char c = (unsigned char) *s++;
Py_UCS4 ch;
int count;
const char *message;
#define WRITE_ASCII_CHAR(ch) \
do { \
assert(ch Py_ssize_t
self as str: self
sub as substr: unicode
start: slice_index(accept={int, NoneType}, c_default='0') = None
end: slice_index(accept={int, NoneType}, c_default='PY_SSIZE_T_MAX') = None
/
Return the number of non-overlapping occurrences of substring sub in string S[start:end].
Optional arguments start and end are interpreted as in slice
notation.
[clinic start generated code]*/
static Py_ssize_t
unicode_count_impl(PyObject *str, PyObject *substr, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=8fcc3aef0b18edbf input=c9209e05438cc352]*/
{
assert(PyUnicode_Check(str));
assert(PyUnicode_Check(substr));
Py_ssize_t result;
int kind1, kind2;
const void *buf1 = NULL, *buf2 = NULL;
Py_ssize_t len1, len2;
kind1 = PyUnicode_KIND(str);
kind2 = PyUnicode_KIND(substr);
if (kind1 < kind2)
return 0;
len1 = PyUnicode_GET_LENGTH(str);
len2 = PyUnicode_GET_LENGTH(substr);
ADJUST_INDICES(start, end, len1);
if (end - start < len2)
return 0;
buf1 = PyUnicode_DATA(str);
buf2 = PyUnicode_DATA(substr);
if (kind2 != kind1) {
buf2 = unicode_askind(kind2, buf2, len2, kind1);
if (!buf2)
goto onError;
}
// We don't reuse `anylib_count` here because of the explicit casts.
switch (kind1) {
case PyUnicode_1BYTE_KIND:
result = ucs1lib_count(
((const Py_UCS1*)buf1) + start, end - start,
buf2, len2, PY_SSIZE_T_MAX
);
break;
case PyUnicode_2BYTE_KIND:
result = ucs2lib_count(
((const Py_UCS2*)buf1) + start, end - start,
buf2, len2, PY_SSIZE_T_MAX
);
break;
case PyUnicode_4BYTE_KIND:
result = ucs4lib_count(
((const Py_UCS4*)buf1) + start, end - start,
buf2, len2, PY_SSIZE_T_MAX
);
break;
default:
Py_UNREACHABLE();
}
assert((kind2 != kind1) == (buf2 != PyUnicode_DATA(substr)));
if (kind2 != kind1)
PyMem_Free((void *)buf2);
return result;
onError:
assert((kind2 != kind1) == (buf2 != PyUnicode_DATA(substr)));
if (kind2 != kind1)
PyMem_Free((void *)buf2);
return -1;
}
/*[clinic input]
str.encode as unicode_encode
encoding: str(c_default="NULL") = 'utf-8'
The encoding in which to encode the string.
errors: str(c_default="NULL") = 'strict'
The error handling scheme to use for encoding errors.
The default is 'strict' meaning that encoding errors raise a
UnicodeEncodeError. Other possible values are 'ignore', 'replace'
and 'xmlcharrefreplace' as well as any other name registered with
codecs.register_error that can handle UnicodeEncodeErrors.
Encode the string using the codec registered for encoding.
[clinic start generated code]*/
static PyObject *
unicode_encode_impl(PyObject *self, const char *encoding, const char *errors)
/*[clinic end generated code: output=bf78b6e2a9470e3c input=b85a9645cb33b729]*/
{
return PyUnicode_AsEncodedString(self, encoding, errors);
}
/*[clinic input]
str.expandtabs as unicode_expandtabs
tabsize: int = 8
Return a copy where all tab characters are expanded using spaces.
If tabsize is not given, a tab size of 8 characters is assumed.
[clinic start generated code]*/
static PyObject *
unicode_expandtabs_impl(PyObject *self, int tabsize)
/*[clinic end generated code: output=3457c5dcee26928f input=8a01914034af4c85]*/
{
Py_ssize_t i, j, line_pos, src_len, incr;
Py_UCS4 ch;
PyObject *u;
const void *src_data;
void *dest_data;
int kind;
int found;
/* First pass: determine size of output string */
src_len = PyUnicode_GET_LENGTH(self);
i = j = line_pos = 0;
kind = PyUnicode_KIND(self);
src_data = PyUnicode_DATA(self);
found = 0;
for (; i < src_len; i++) {
ch = PyUnicode_READ(kind, src_data, i);
if (ch == '\t') {
found = 1;
if (tabsize > 0) {
incr = tabsize - (line_pos % tabsize); /* cannot overflow */
if (j > PY_SSIZE_T_MAX - incr)
goto overflow;
line_pos += incr;
j += incr;
}
}
else {
if (j > PY_SSIZE_T_MAX - 1)
goto overflow;
line_pos++;
j++;
if (ch == '\n' || ch == '\r')
line_pos = 0;
}
}
if (!found)
return unicode_result_unchanged(self);
/* Second pass: create output string and fill it */
u = PyUnicode_New(j, PyUnicode_MAX_CHAR_VALUE(self));
if (!u)
return NULL;
dest_data = PyUnicode_DATA(u);
i = j = line_pos = 0;
for (; i < src_len; i++) {
ch = PyUnicode_READ(kind, src_data, i);
if (ch == '\t') {
if (tabsize > 0) {
incr = tabsize - (line_pos % tabsize);
line_pos += incr;
_PyUnicode_Fill(kind, dest_data, ' ', j, incr);
j += incr;
}
}
else {
line_pos++;
PyUnicode_WRITE(kind, dest_data, j, ch);
j++;
if (ch == '\n' || ch == '\r')
line_pos = 0;
}
}
assert (j == PyUnicode_GET_LENGTH(u));
return unicode_result(u);
overflow:
PyErr_SetString(PyExc_OverflowError, "new string is too long");
return NULL;
}
/*[clinic input]
@permit_long_summary
str.find as unicode_find = str.count
Return the lowest index in S where substring sub is found, such that sub is contained within S[start:end].
Optional arguments start and end are interpreted as in slice
notation. Return -1 on failure.
[clinic start generated code]*/
static Py_ssize_t
unicode_find_impl(PyObject *str, PyObject *substr, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=51dbe6255712e278 input=f57e93c59d1ee927]*/
{
Py_ssize_t result = any_find_slice(str, substr, start, end, 1);
if (result < 0) {
return -1;
}
return result;
}
static PyObject *
unicode_getitem(PyObject *self, Py_ssize_t index)
{
const void *data;
int kind;
Py_UCS4 ch;
if (!PyUnicode_Check(self)) {
PyErr_BadArgument();
return NULL;
}
if (index < 0 || index >= PyUnicode_GET_LENGTH(self)) {
PyErr_SetString(PyExc_IndexError, "string index out of range");
return NULL;
}
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
ch = PyUnicode_READ(kind, data, index);
return unicode_char(ch);
}
/* Believe it or not, this produces the same value for ASCII strings
as bytes_hash(). */
static Py_hash_t
unicode_hash(PyObject *self)
{
Py_uhash_t x; /* Unsigned for defined overflow behavior. */
#ifdef Py_DEBUG
assert(_Py_HashSecret_Initialized);
#endif
Py_hash_t hash = PyUnicode_HASH(self);
if (hash != -1) {
return hash;
}
x = Py_HashBuffer(PyUnicode_DATA(self),
PyUnicode_GET_LENGTH(self) * PyUnicode_KIND(self));
PyUnicode_SET_HASH(self, x);
return x;
}
/*[clinic input]
@permit_long_summary
str.index as unicode_index = str.count
Return the lowest index in S where substring sub is found, such that sub is contained within S[start:end].
Optional arguments start and end are interpreted as in slice
notation. Raises ValueError when the substring is not found.
[clinic start generated code]*/
static Py_ssize_t
unicode_index_impl(PyObject *str, PyObject *substr, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=77558288837cdf40 input=5900ab84de55e628]*/
{
Py_ssize_t result = any_find_slice(str, substr, start, end, 1);
if (result == -1) {
PyErr_SetString(PyExc_ValueError, "substring not found");
}
else if (result < 0) {
return -1;
}
return result;
}
/*[clinic input]
@permit_long_summary
str.isascii as unicode_isascii
Return True if all characters in the string are ASCII, False otherwise.
ASCII characters have code points in the range U+0000-U+007F.
Empty string is ASCII too.
[clinic start generated code]*/
static PyObject *
unicode_isascii_impl(PyObject *self)
/*[clinic end generated code: output=c5910d64b5a8003f input=dc74e1ced821159f]*/
{
return PyBool_FromLong(PyUnicode_IS_ASCII(self));
}
/*[clinic input]
str.islower as unicode_islower
Return True if the string is a lowercase string, False otherwise.
A string is lowercase if all cased characters in the string are
lowercase and there is at least one cased character in the string.
[clinic start generated code]*/
static PyObject *
unicode_islower_impl(PyObject *self)
/*[clinic end generated code: output=dbd41995bd005b81 input=1879b48dfc628366]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
int cased;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISLOWER(PyUnicode_READ(kind, data, 0)));
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
cased = 0;
for (i = 0; i < length; i++) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (Py_UNICODE_ISUPPER(ch) || Py_UNICODE_ISTITLE(ch))
Py_RETURN_FALSE;
else if (!cased && Py_UNICODE_ISLOWER(ch))
cased = 1;
}
return PyBool_FromLong(cased);
}
/*[clinic input]
str.isupper as unicode_isupper
Return True if the string is an uppercase string, False otherwise.
A string is uppercase if all cased characters in the string are
uppercase and there is at least one cased character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isupper_impl(PyObject *self)
/*[clinic end generated code: output=049209c8e7f15f59 input=77d29904aef0e3a0]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
int cased;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISUPPER(PyUnicode_READ(kind, data, 0)) != 0);
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
cased = 0;
for (i = 0; i < length; i++) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (Py_UNICODE_ISLOWER(ch) || Py_UNICODE_ISTITLE(ch))
Py_RETURN_FALSE;
else if (!cased && Py_UNICODE_ISUPPER(ch))
cased = 1;
}
return PyBool_FromLong(cased);
}
/*[clinic input]
str.istitle as unicode_istitle
Return True if the string is a title-cased string, False otherwise.
In a title-cased string, upper- and title-case characters may only
follow uncased characters and lowercase characters only cased ones.
[clinic start generated code]*/
static PyObject *
unicode_istitle_impl(PyObject *self)
/*[clinic end generated code: output=e9bf6eb91f5d3f0e input=98d32bd2e1f06f8c]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
int cased, previous_is_cased;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1) {
Py_UCS4 ch = PyUnicode_READ(kind, data, 0);
return PyBool_FromLong((Py_UNICODE_ISTITLE(ch) != 0) ||
(Py_UNICODE_ISUPPER(ch) != 0));
}
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
cased = 0;
previous_is_cased = 0;
for (i = 0; i < length; i++) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (Py_UNICODE_ISUPPER(ch) || Py_UNICODE_ISTITLE(ch)) {
if (previous_is_cased)
Py_RETURN_FALSE;
previous_is_cased = 1;
cased = 1;
}
else if (Py_UNICODE_ISLOWER(ch)) {
if (!previous_is_cased)
Py_RETURN_FALSE;
previous_is_cased = 1;
cased = 1;
}
else
previous_is_cased = 0;
}
return PyBool_FromLong(cased);
}
/*[clinic input]
str.isspace as unicode_isspace
Return True if the string is a whitespace string, False otherwise.
A string is whitespace if all characters in the string are
whitespace and there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isspace_impl(PyObject *self)
/*[clinic end generated code: output=163a63bfa08ac2b9 input=29e09560fc23fbeb]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISSPACE(PyUnicode_READ(kind, data, 0)));
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
for (i = 0; i < length; i++) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (!Py_UNICODE_ISSPACE(ch))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
/*[clinic input]
str.isalpha as unicode_isalpha
Return True if the string is an alphabetic string, False otherwise.
A string is alphabetic if all characters in the string are
alphabetic and there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isalpha_impl(PyObject *self)
/*[clinic end generated code: output=cc81b9ac3883ec4f input=9906a07f3e04892e]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISALPHA(PyUnicode_READ(kind, data, 0)));
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
for (i = 0; i < length; i++) {
if (!Py_UNICODE_ISALPHA(PyUnicode_READ(kind, data, i)))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
/*[clinic input]
@permit_long_summary
str.isalnum as unicode_isalnum
Return True if the string is an alpha-numeric string, False otherwise.
A string is alpha-numeric if all characters in the string are
alpha-numeric and there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isalnum_impl(PyObject *self)
/*[clinic end generated code: output=a5a23490ffc3660c input=892f64ebc171fd4f]*/
{
int kind;
const void *data;
Py_ssize_t len, i;
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
len = PyUnicode_GET_LENGTH(self);
/* Shortcut for single character strings */
if (len == 1) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, 0);
return PyBool_FromLong(Py_UNICODE_ISALNUM(ch));
}
/* Special case for empty strings */
if (len == 0)
Py_RETURN_FALSE;
for (i = 0; i < len; i++) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (!Py_UNICODE_ISALNUM(ch))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
/*[clinic input]
str.isdecimal as unicode_isdecimal
Return True if the string is a decimal string, False otherwise.
A string is a decimal string if all characters in the string are
decimal and there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isdecimal_impl(PyObject *self)
/*[clinic end generated code: output=fb2dcdb62d3fc548 input=63b0453c48cad0af]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISDECIMAL(PyUnicode_READ(kind, data, 0)));
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
for (i = 0; i < length; i++) {
if (!Py_UNICODE_ISDECIMAL(PyUnicode_READ(kind, data, i)))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
/*[clinic input]
str.isdigit as unicode_isdigit
Return True if the string is a digit string, False otherwise.
A string is a digit string if all characters in the string are
digits and there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isdigit_impl(PyObject *self)
/*[clinic end generated code: output=10a6985311da6858 input=353b03747b062e4b]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1) {
const Py_UCS4 ch = PyUnicode_READ(kind, data, 0);
return PyBool_FromLong(Py_UNICODE_ISDIGIT(ch));
}
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
for (i = 0; i < length; i++) {
if (!Py_UNICODE_ISDIGIT(PyUnicode_READ(kind, data, i)))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
/*[clinic input]
str.isnumeric as unicode_isnumeric
Return True if the string is a numeric string, False otherwise.
A string is numeric if all characters in the string are numeric and
there is at least one character in the string.
[clinic start generated code]*/
static PyObject *
unicode_isnumeric_impl(PyObject *self)
/*[clinic end generated code: output=9172a32d9013051a input=83b2a072ed7aff48]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISNUMERIC(PyUnicode_READ(kind, data, 0)));
/* Special case for empty strings */
if (length == 0)
Py_RETURN_FALSE;
for (i = 0; i < length; i++) {
if (!Py_UNICODE_ISNUMERIC(PyUnicode_READ(kind, data, i)))
Py_RETURN_FALSE;
}
Py_RETURN_TRUE;
}
Py_ssize_t
_PyUnicode_ScanIdentifier(PyObject *self)
{
Py_ssize_t i;
Py_ssize_t len = PyUnicode_GET_LENGTH(self);
if (len == 0) {
/* an empty string is not a valid identifier */
return 0;
}
int kind = PyUnicode_KIND(self);
const void *data = PyUnicode_DATA(self);
Py_UCS4 ch = PyUnicode_READ(kind, data, 0);
/* PEP 3131 says that the first character must be in
XID_Start and subsequent characters in XID_Continue,
and for the ASCII range, the 2.x rules apply (i.e
start with letters and underscore, continue with
letters, digits, underscore). However, given the current
definition of XID_Start and XID_Continue, it is sufficient
to check just for these, except that _ must be allowed
as starting an identifier. */
if (!_PyUnicode_IsXidStart(ch) && ch != 0x5F /* LOW LINE */) {
return 0;
}
for (i = 1; i < len; i++) {
ch = PyUnicode_READ(kind, data, i);
if (!_PyUnicode_IsXidContinue(ch)) {
return i;
}
}
return i;
}
int
PyUnicode_IsIdentifier(PyObject *self)
{
Py_ssize_t i = _PyUnicode_ScanIdentifier(self);
Py_ssize_t len = PyUnicode_GET_LENGTH(self);
/* an empty string is not a valid identifier */
return len && i == len;
}
/*[clinic input]
@permit_long_summary
str.isidentifier as unicode_isidentifier
Return True if the string is a valid Python identifier, False otherwise.
Call keyword.iskeyword(s) to test whether string s is a reserved
identifier, such as "def" or "class".
[clinic start generated code]*/
static PyObject *
unicode_isidentifier_impl(PyObject *self)
/*[clinic end generated code: output=fe585a9666572905 input=cabde62c20a3be6b]*/
{
return PyBool_FromLong(PyUnicode_IsIdentifier(self));
}
/*[clinic input]
@permit_long_summary
str.isprintable as unicode_isprintable
Return True if all characters in the string are printable, False otherwise.
A character is printable if repr() may use it in its output.
[clinic start generated code]*/
static PyObject *
unicode_isprintable_impl(PyObject *self)
/*[clinic end generated code: output=3ab9626cd32dd1a0 input=18345ba847084ec5]*/
{
Py_ssize_t i, length;
int kind;
const void *data;
length = PyUnicode_GET_LENGTH(self);
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
/* Shortcut for single character strings */
if (length == 1)
return PyBool_FromLong(
Py_UNICODE_ISPRINTABLE(PyUnicode_READ(kind, data, 0)));
for (i = 0; i < length; i++) {
if (!Py_UNICODE_ISPRINTABLE(PyUnicode_READ(kind, data, i))) {
Py_RETURN_FALSE;
}
}
Py_RETURN_TRUE;
}
/*[clinic input]
str.join as unicode_join
iterable: object
/
Concatenate any number of strings.
The string whose method is called is inserted in between each given
string. The result is returned as a new string.
Example: '.'.join(['ab', 'pq', 'rs']) -> 'ab.pq.rs'
[clinic start generated code]*/
static PyObject *
unicode_join(PyObject *self, PyObject *iterable)
/*[clinic end generated code: output=6857e7cecfe7bf98 input=fd330a11ee845fb2]*/
{
return PyUnicode_Join(self, iterable);
}
static Py_ssize_t
unicode_length(PyObject *self)
{
return PyUnicode_GET_LENGTH(self);
}
/*[clinic input]
str.ljust as unicode_ljust
width: Py_ssize_t
fillchar: Py_UCS4 = ' '
/
Return a left-justified string of length width.
Padding is done using the specified fill character (default is
a space).
[clinic start generated code]*/
static PyObject *
unicode_ljust_impl(PyObject *self, Py_ssize_t width, Py_UCS4 fillchar)
/*[clinic end generated code: output=1cce0e0e0a0b84b3 input=8a55f06694c20ed6]*/
{
if (PyUnicode_GET_LENGTH(self) >= width)
return unicode_result_unchanged(self);
return pad(self, 0, width - PyUnicode_GET_LENGTH(self), fillchar);
}
/*[clinic input]
str.lower as unicode_lower
Return a copy of the string converted to lowercase.
[clinic start generated code]*/
static PyObject *
unicode_lower_impl(PyObject *self)
/*[clinic end generated code: output=84ef9ed42efad663 input=60a2984b8beff23a]*/
{
if (PyUnicode_IS_ASCII(self))
return ascii_upper_or_lower(self, 1);
return case_operation(self, do_lower);
}
#define LEFTSTRIP 0
#define RIGHTSTRIP 1
#define BOTHSTRIP 2
/* Arrays indexed by above */
static const char *stripfuncnames[] = {"lstrip", "rstrip", "strip"};
#define STRIPNAME(i) (stripfuncnames[i])
/* externally visible for str.strip(unicode) */
PyObject *
_PyUnicode_XStrip(PyObject *self, int striptype, PyObject *sepobj)
{
const void *data;
int kind;
Py_ssize_t i, j, len;
BLOOM_MASK sepmask;
Py_ssize_t seplen;
kind = PyUnicode_KIND(self);
data = PyUnicode_DATA(self);
len = PyUnicode_GET_LENGTH(self);
seplen = PyUnicode_GET_LENGTH(sepobj);
sepmask = make_bloom_mask(PyUnicode_KIND(sepobj),
PyUnicode_DATA(sepobj),
seplen);
i = 0;
if (striptype != RIGHTSTRIP) {
while (i < len) {
Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (!BLOOM(sepmask, ch))
break;
if (PyUnicode_FindChar(sepobj, ch, 0, seplen, 1) < 0)
break;
i++;
}
}
j = len;
if (striptype != LEFTSTRIP) {
j--;
while (j >= i) {
Py_UCS4 ch = PyUnicode_READ(kind, data, j);
if (!BLOOM(sepmask, ch))
break;
if (PyUnicode_FindChar(sepobj, ch, 0, seplen, 1) < 0)
break;
j--;
}
j++;
}
return PyUnicode_Substring(self, i, j);
}
PyObject*
_PyUnicode_BinarySlice(PyObject *container, PyObject *start_o, PyObject *stop_o)
{
assert(PyUnicode_CheckExact(container));
Py_ssize_t len = PyUnicode_GET_LENGTH(container);
Py_ssize_t istart, istop;
if (!_PyEval_UnpackIndices(start_o, stop_o, len, &istart, &istop)) {
return NULL;
}
return PyUnicode_Substring(container, istart, istop);
}
PyObject*
PyUnicode_Substring(PyObject *self, Py_ssize_t start, Py_ssize_t end)
{
const unsigned char *data;
int kind;
Py_ssize_t length;
length = PyUnicode_GET_LENGTH(self);
end = Py_MIN(end, length);
if (start == 0 && end == length)
return unicode_result_unchanged(self);
if (start < 0 || end < 0) {
PyErr_SetString(PyExc_IndexError, "string index out of range");
return NULL;
}
if (start >= length || end < start)
_Py_RETURN_UNICODE_EMPTY();
length = end - start;
if (PyUnicode_IS_ASCII(self)) {
data = PyUnicode_1BYTE_DATA(self);
return _PyUnicode_FromASCII((const char*)(data + start), length);
}
else {
kind = PyUnicode_KIND(self);
data = PyUnicode_1BYTE_DATA(self);
return PyUnicode_FromKindAndData(kind,
data + kind * start,
length);
}
}
static PyObject *
do_strip(PyObject *self, int striptype)
{
Py_ssize_t len, i, j;
len = PyUnicode_GET_LENGTH(self);
if (PyUnicode_IS_ASCII(self)) {
const Py_UCS1 *data = PyUnicode_1BYTE_DATA(self);
i = 0;
if (striptype != RIGHTSTRIP) {
while (i < len) {
Py_UCS1 ch = data[i];
if (!_Py_ascii_whitespace[ch])
break;
i++;
}
}
j = len;
if (striptype != LEFTSTRIP) {
j--;
while (j >= i) {
Py_UCS1 ch = data[j];
if (!_Py_ascii_whitespace[ch])
break;
j--;
}
j++;
}
}
else {
int kind = PyUnicode_KIND(self);
const void *data = PyUnicode_DATA(self);
i = 0;
if (striptype != RIGHTSTRIP) {
while (i < len) {
Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (!Py_UNICODE_ISSPACE(ch))
break;
i++;
}
}
j = len;
if (striptype != LEFTSTRIP) {
j--;
while (j >= i) {
Py_UCS4 ch = PyUnicode_READ(kind, data, j);
if (!Py_UNICODE_ISSPACE(ch))
break;
j--;
}
j++;
}
}
return PyUnicode_Substring(self, i, j);
}
static PyObject *
do_argstrip(PyObject *self, int striptype, PyObject *sep)
{
if (sep != Py_None) {
if (PyUnicode_Check(sep))
return _PyUnicode_XStrip(self, striptype, sep);
else {
PyErr_Format(PyExc_TypeError,
"%s arg must be None or str",
STRIPNAME(striptype));
return NULL;
}
}
return do_strip(self, striptype);
}
/*[clinic input]
@permit_long_summary
str.strip as unicode_strip
chars: object = None
/
Return a copy of the string with leading and trailing whitespace removed.
If chars is given and not None, remove characters in chars instead.
[clinic start generated code]*/
static PyObject *
unicode_strip_impl(PyObject *self, PyObject *chars)
/*[clinic end generated code: output=ca19018454345d57 input=8bc6353450345fbd]*/
{
return do_argstrip(self, BOTHSTRIP, chars);
}
/*[clinic input]
str.lstrip as unicode_lstrip
chars: object = None
/
Return a copy of the string with leading whitespace removed.
If chars is given and not None, remove characters in chars instead.
[clinic start generated code]*/
static PyObject *
unicode_lstrip_impl(PyObject *self, PyObject *chars)
/*[clinic end generated code: output=3b43683251f79ca7 input=529f9f3834448671]*/
{
return do_argstrip(self, LEFTSTRIP, chars);
}
/*[clinic input]
str.rstrip as unicode_rstrip
chars: object = None
/
Return a copy of the string with trailing whitespace removed.
If chars is given and not None, remove characters in chars instead.
[clinic start generated code]*/
static PyObject *
unicode_rstrip_impl(PyObject *self, PyObject *chars)
/*[clinic end generated code: output=4a59230017cc3b7a input=62566c627916557f]*/
{
return do_argstrip(self, RIGHTSTRIP, chars);
}
PyObject *
_PyUnicode_Repeat(PyObject *str, Py_ssize_t len)
{
PyObject *u;
Py_ssize_t nchars, n;
if (len < 1)
_Py_RETURN_UNICODE_EMPTY();
/* no repeat, return original string */
if (len == 1)
return unicode_result_unchanged(str);
if (PyUnicode_GET_LENGTH(str) > PY_SSIZE_T_MAX / len) {
PyErr_SetString(PyExc_OverflowError,
"repeated string is too long");
return NULL;
}
nchars = len * PyUnicode_GET_LENGTH(str);
u = PyUnicode_New(nchars, PyUnicode_MAX_CHAR_VALUE(str));
if (!u)
return NULL;
assert(PyUnicode_KIND(u) == PyUnicode_KIND(str));
if (PyUnicode_GET_LENGTH(str) == 1) {
int kind = PyUnicode_KIND(str);
Py_UCS4 fill_char = PyUnicode_READ(kind, PyUnicode_DATA(str), 0);
if (kind == PyUnicode_1BYTE_KIND) {
void *to = PyUnicode_DATA(u);
memset(to, (unsigned char)fill_char, len);
}
else if (kind == PyUnicode_2BYTE_KIND) {
Py_UCS2 *ucs2 = PyUnicode_2BYTE_DATA(u);
for (n = 0; n < len; ++n)
ucs2[n] = fill_char;
} else {
Py_UCS4 *ucs4 = PyUnicode_4BYTE_DATA(u);
assert(kind == PyUnicode_4BYTE_KIND);
for (n = 0; n < len; ++n)
ucs4[n] = fill_char;
}
}
else {
Py_ssize_t char_size = PyUnicode_KIND(str);
char *to = (char *) PyUnicode_DATA(u);
_PyBytes_RepeatBuffer(to, nchars * char_size, PyUnicode_DATA(str),
PyUnicode_GET_LENGTH(str) * char_size);
}
assert(_PyUnicode_CheckConsistency(u, 1));
return u;
}
PyObject *
PyUnicode_Replace(PyObject *str,
PyObject *substr,
PyObject *replstr,
Py_ssize_t maxcount)
{
if (ensure_unicode(str) < 0 || ensure_unicode(substr) < 0 ||
ensure_unicode(replstr) < 0)
return NULL;
return replace(str, substr, replstr, maxcount);
}
/*[clinic input]
str.replace as unicode_replace
old: unicode
new: unicode
/
count: Py_ssize_t = -1
Maximum number of occurrences to replace.
-1 (the default value) means replace all occurrences.
Return a copy with all occurrences of substring old replaced by new.
If count is given, only the first count occurrences are replaced.
If count is not specified or -1, then all occurrences are replaced.
[clinic start generated code]*/
static PyObject *
unicode_replace_impl(PyObject *self, PyObject *old, PyObject *new,
Py_ssize_t count)
/*[clinic end generated code: output=b63f1a8b5eebf448 input=d15a6886b05e2edc]*/
{
return replace(self, old, new, count);
}
/*[clinic input]
str.removeprefix as unicode_removeprefix
prefix: unicode
/
Return a str with the given prefix string removed if present.
If the string starts with the prefix string, return
string[len(prefix):]. Otherwise, return a copy of the original
string.
[clinic start generated code]*/
static PyObject *
unicode_removeprefix_impl(PyObject *self, PyObject *prefix)
/*[clinic end generated code: output=f1e5945e9763bcb9 input=90d162724944bfa7]*/
{
int match = tailmatch(self, prefix, 0, PY_SSIZE_T_MAX, -1);
if (match == -1) {
return NULL;
}
if (match) {
return PyUnicode_Substring(self, PyUnicode_GET_LENGTH(prefix),
PyUnicode_GET_LENGTH(self));
}
return unicode_result_unchanged(self);
}
/*[clinic input]
str.removesuffix as unicode_removesuffix
suffix: unicode
/
Return a str with the given suffix string removed if present.
If the string ends with the suffix string and that suffix is not
empty, return string[:-len(suffix)]. Otherwise, return a copy of
the original string.
[clinic start generated code]*/
static PyObject *
unicode_removesuffix_impl(PyObject *self, PyObject *suffix)
/*[clinic end generated code: output=d36629e227636822 input=6efc96152d4bfcd5]*/
{
int match = tailmatch(self, suffix, 0, PY_SSIZE_T_MAX, +1);
if (match == -1) {
return NULL;
}
if (match) {
return PyUnicode_Substring(self, 0, PyUnicode_GET_LENGTH(self)
- PyUnicode_GET_LENGTH(suffix));
}
return unicode_result_unchanged(self);
}
static PyObject *
unicode_repr(PyObject *unicode)
{
Py_ssize_t isize = PyUnicode_GET_LENGTH(unicode);
const void *idata = PyUnicode_DATA(unicode);
/* Compute length of output, quote characters, and
maximum character */
Py_ssize_t osize = 0;
Py_UCS4 maxch = 127;
Py_ssize_t squote = 0;
Py_ssize_t dquote = 0;
int ikind = PyUnicode_KIND(unicode);
for (Py_ssize_t i = 0; i < isize; i++) {
Py_UCS4 ch = PyUnicode_READ(ikind, idata, i);
Py_ssize_t incr = 1;
switch (ch) {
case '\'': squote++; break;
case '"': dquote++; break;
case '\\': case '\t': case '\r': case '\n':
incr = 2;
break;
default:
/* Fast-path ASCII */
if (ch < ' ' || ch == 0x7f)
incr = 4; /* \xHH */
else if (ch < 0x7f)
;
else if (Py_UNICODE_ISPRINTABLE(ch))
maxch = (ch > maxch) ? ch : maxch;
else if (ch < 0x100)
incr = 4; /* \xHH */
else if (ch < 0x10000)
incr = 6; /* \uHHHH */
else
incr = 10; /* \uHHHHHHHH */
}
if (osize > PY_SSIZE_T_MAX - incr) {
PyErr_SetString(PyExc_OverflowError,
"string is too long to generate repr");
return NULL;
}
osize += incr;
}
Py_UCS4 quote = '\'';
int changed = (osize != isize);
if (squote) {
changed = 1;
if (dquote)
/* Both squote and dquote present. Use squote,
and escape them */
osize += squote;
else
quote = '"';
}
osize += 2; /* quotes */
PyObject *repr = PyUnicode_New(osize, maxch);
if (repr == NULL)
return NULL;
int okind = PyUnicode_KIND(repr);
void *odata = PyUnicode_DATA(repr);
if (!changed) {
PyUnicode_WRITE(okind, odata, 0, quote);
_PyUnicode_FastCopyCharacters(repr, 1,
unicode, 0,
isize);
PyUnicode_WRITE(okind, odata, osize-1, quote);
}
else {
switch (okind) {
case PyUnicode_1BYTE_KIND:
ucs1lib_repr(unicode, quote, odata);
break;
case PyUnicode_2BYTE_KIND:
ucs2lib_repr(unicode, quote, odata);
break;
default:
assert(okind == PyUnicode_4BYTE_KIND);
ucs4lib_repr(unicode, quote, odata);
}
}
assert(_PyUnicode_CheckConsistency(repr, 1));
return repr;
}
/*[clinic input]
@permit_long_summary
str.rfind as unicode_rfind = str.count
Return the highest index in S where substring sub is found, such that sub is contained within S[start:end].
Optional arguments start and end are interpreted as in slice
notation. Return -1 on failure.
[clinic start generated code]*/
static Py_ssize_t
unicode_rfind_impl(PyObject *str, PyObject *substr, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=880b29f01dd014c8 input=2e67789533baf2f5]*/
{
Py_ssize_t result = any_find_slice(str, substr, start, end, -1);
if (result < 0) {
return -1;
}
return result;
}
/*[clinic input]
@permit_long_summary
str.rindex as unicode_rindex = str.count
Return the highest index in S where substring sub is found, such that sub is contained within S[start:end].
Optional arguments start and end are interpreted as in slice
notation. Raises ValueError when the substring is not found.
[clinic start generated code]*/
static Py_ssize_t
unicode_rindex_impl(PyObject *str, PyObject *substr, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=5f3aef124c867fe1 input=e29d446c8234c9d9]*/
{
Py_ssize_t result = any_find_slice(str, substr, start, end, -1);
if (result == -1) {
PyErr_SetString(PyExc_ValueError, "substring not found");
}
else if (result < 0) {
return -1;
}
return result;
}
/*[clinic input]
str.rjust as unicode_rjust
width: Py_ssize_t
fillchar: Py_UCS4 = ' '
/
Return a right-justified string of length width.
Padding is done using the specified fill character (default is
a space).
[clinic start generated code]*/
static PyObject *
unicode_rjust_impl(PyObject *self, Py_ssize_t width, Py_UCS4 fillchar)
/*[clinic end generated code: output=804a1a57fbe8d5cf input=1256a8d659589907]*/
{
if (PyUnicode_GET_LENGTH(self) >= width)
return unicode_result_unchanged(self);
return pad(self, width - PyUnicode_GET_LENGTH(self), 0, fillchar);
}
PyObject *
PyUnicode_Split(PyObject *s, PyObject *sep, Py_ssize_t maxsplit)
{
if (ensure_unicode(s) < 0 || (sep != NULL && ensure_unicode(sep) < 0))
return NULL;
return split(s, sep, maxsplit);
}
/*[clinic input]
@permit_long_summary
str.split as unicode_split
sep: object = None
The separator used to split the string.
When set to None (the default value), will split on any
whitespace character (including \n \r \t \f and spaces) and
will discard empty strings from the result.
maxsplit: Py_ssize_t = -1
Maximum number of splits.
-1 (the default value) means no limit.
Return a list of the substrings in the string, using sep as the separator string.
Splitting starts at the front of the string and works to the end.
Note, str.split() is mainly useful for data that has been
intentionally delimited. With natural text that includes
punctuation, consider using the regular expression module.
[clinic start generated code]*/
static PyObject *
unicode_split_impl(PyObject *self, PyObject *sep, Py_ssize_t maxsplit)
/*[clinic end generated code: output=3a65b1db356948dc input=288cfd6bc8828f5a]*/
{
if (sep == Py_None)
return split(self, NULL, maxsplit);
if (PyUnicode_Check(sep))
return split(self, sep, maxsplit);
PyErr_Format(PyExc_TypeError,
"must be str or None, not %.100s",
Py_TYPE(sep)->tp_name);
return NULL;
}
PyObject *
PyUnicode_Partition(PyObject *str_obj, PyObject *sep_obj)
{
PyObject* out;
int kind1, kind2;
const void *buf1, *buf2;
Py_ssize_t len1, len2;
if (ensure_unicode(str_obj) < 0 || ensure_unicode(sep_obj) < 0)
return NULL;
kind1 = PyUnicode_KIND(str_obj);
kind2 = PyUnicode_KIND(sep_obj);
len1 = PyUnicode_GET_LENGTH(str_obj);
len2 = PyUnicode_GET_LENGTH(sep_obj);
if (kind1 < kind2 || len1 < len2) {
PyObject *empty = _PyUnicode_GetEmpty(); // Borrowed reference
return PyTuple_Pack(3, str_obj, empty, empty);
}
buf1 = PyUnicode_DATA(str_obj);
buf2 = PyUnicode_DATA(sep_obj);
if (kind2 != kind1) {
buf2 = unicode_askind(kind2, buf2, len2, kind1);
if (!buf2)
return NULL;
}
switch (kind1) {
case PyUnicode_1BYTE_KIND:
if (PyUnicode_IS_ASCII(str_obj) && PyUnicode_IS_ASCII(sep_obj))
out = asciilib_partition(str_obj, buf1, len1, sep_obj, buf2, len2);
else
out = ucs1lib_partition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
case PyUnicode_2BYTE_KIND:
out = ucs2lib_partition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
case PyUnicode_4BYTE_KIND:
out = ucs4lib_partition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
default:
Py_UNREACHABLE();
}
assert((kind2 == kind1) == (buf2 == PyUnicode_DATA(sep_obj)));
if (kind2 != kind1)
PyMem_Free((void *)buf2);
return out;
}
PyObject *
PyUnicode_RPartition(PyObject *str_obj, PyObject *sep_obj)
{
PyObject* out;
int kind1, kind2;
const void *buf1, *buf2;
Py_ssize_t len1, len2;
if (ensure_unicode(str_obj) < 0 || ensure_unicode(sep_obj) < 0)
return NULL;
kind1 = PyUnicode_KIND(str_obj);
kind2 = PyUnicode_KIND(sep_obj);
len1 = PyUnicode_GET_LENGTH(str_obj);
len2 = PyUnicode_GET_LENGTH(sep_obj);
if (kind1 < kind2 || len1 < len2) {
PyObject *empty = _PyUnicode_GetEmpty(); // Borrowed reference
return PyTuple_Pack(3, empty, empty, str_obj);
}
buf1 = PyUnicode_DATA(str_obj);
buf2 = PyUnicode_DATA(sep_obj);
if (kind2 != kind1) {
buf2 = unicode_askind(kind2, buf2, len2, kind1);
if (!buf2)
return NULL;
}
switch (kind1) {
case PyUnicode_1BYTE_KIND:
if (PyUnicode_IS_ASCII(str_obj) && PyUnicode_IS_ASCII(sep_obj))
out = asciilib_rpartition(str_obj, buf1, len1, sep_obj, buf2, len2);
else
out = ucs1lib_rpartition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
case PyUnicode_2BYTE_KIND:
out = ucs2lib_rpartition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
case PyUnicode_4BYTE_KIND:
out = ucs4lib_rpartition(str_obj, buf1, len1, sep_obj, buf2, len2);
break;
default:
Py_UNREACHABLE();
}
assert((kind2 == kind1) == (buf2 == PyUnicode_DATA(sep_obj)));
if (kind2 != kind1)
PyMem_Free((void *)buf2);
return out;
}
/*[clinic input]
str.partition as unicode_partition
sep: object
/
Partition the string into three parts using the given separator.
This will search for the separator in the string. If the separator
is found, returns a 3-tuple containing the part before the
separator, the separator itself, and the part after it.
If the separator is not found, returns a 3-tuple containing
the original string and two empty strings.
[clinic start generated code]*/
static PyObject *
unicode_partition(PyObject *self, PyObject *sep)
/*[clinic end generated code: output=e4ced7bd253ca3c4 input=e45faa8c26270cb1]*/
{
return PyUnicode_Partition(self, sep);
}
/*[clinic input]
str.rpartition as unicode_rpartition = str.partition
Partition the string into three parts using the given separator.
This will search for the separator in the string, starting at the
end. If the separator is found, returns a 3-tuple containing the
part before the separator, the separator itself, and the part after
it.
If the separator is not found, returns a 3-tuple containing two
empty strings and the original string.
[clinic start generated code]*/
static PyObject *
unicode_rpartition(PyObject *self, PyObject *sep)
/*[clinic end generated code: output=1aa13cf1156572aa input=53a7f8cb19975b7c]*/
{
return PyUnicode_RPartition(self, sep);
}
PyObject *
PyUnicode_RSplit(PyObject *s, PyObject *sep, Py_ssize_t maxsplit)
{
if (ensure_unicode(s) < 0 || (sep != NULL && ensure_unicode(sep) < 0))
return NULL;
return rsplit(s, sep, maxsplit);
}
/*[clinic input]
@permit_long_summary
str.rsplit as unicode_rsplit = str.split
Return a list of the substrings in the string, using sep as the separator string.
Splitting starts at the end of the string and works to the front.
[clinic start generated code]*/
static PyObject *
unicode_rsplit_impl(PyObject *self, PyObject *sep, Py_ssize_t maxsplit)
/*[clinic end generated code: output=c2b815c63bcabffc input=0f762e30d267fa83]*/
{
if (sep == Py_None)
return rsplit(self, NULL, maxsplit);
if (PyUnicode_Check(sep))
return rsplit(self, sep, maxsplit);
PyErr_Format(PyExc_TypeError,
"must be str or None, not %.100s",
Py_TYPE(sep)->tp_name);
return NULL;
}
/*[clinic input]
@permit_long_summary
str.splitlines as unicode_splitlines
keepends: bool = False
Return a list of the lines in the string, breaking at line boundaries.
Line breaks are not included in the resulting list unless keepends
is given and true.
[clinic start generated code]*/
static PyObject *
unicode_splitlines_impl(PyObject *self, int keepends)
/*[clinic end generated code: output=f664dcdad153ec40 input=b45ea0f87645a06d]*/
{
return PyUnicode_Splitlines(self, keepends);
}
static
PyObject *unicode_str(PyObject *self)
{
return unicode_result_unchanged(self);
}
/*[clinic input]
@permit_long_summary
str.swapcase as unicode_swapcase
Convert uppercase characters to lowercase and lowercase characters to uppercase.
[clinic start generated code]*/
static PyObject *
unicode_swapcase_impl(PyObject *self)
/*[clinic end generated code: output=5d28966bf6d7b2af input=85bc39a9b4e8ee91]*/
{
return case_operation(self, do_swapcase);
}
static int
unicode_maketrans_from_dict(PyObject *x, PyObject *newdict)
{
PyObject *key, *value;
Py_ssize_t i = 0;
int res;
while (PyDict_Next(x, &i, &key, &value)) {
if (PyUnicode_Check(key)) {
PyObject *newkey;
int kind;
const void *data;
if (PyUnicode_GET_LENGTH(key) != 1) {
PyErr_SetString(PyExc_ValueError, "string keys in translate"
"table must be of length 1");
return -1;
}
kind = PyUnicode_KIND(key);
data = PyUnicode_DATA(key);
newkey = PyLong_FromLong(PyUnicode_READ(kind, data, 0));
if (!newkey)
return -1;
res = PyDict_SetItem(newdict, newkey, value);
Py_DECREF(newkey);
if (res < 0)
return -1;
}
else if (PyLong_Check(key)) {
if (PyDict_SetItem(newdict, key, value) < 0)
return -1;
}
else {
PyErr_SetString(PyExc_TypeError, "keys in translate table must"
"be strings or integers");
return -1;
}
}
return 0;
}
/*[clinic input]
@staticmethod
str.maketrans as unicode_maketrans
x: object
y: unicode=NULL
z: unicode=NULL
/
Return a translation table usable for str.translate().
If there is only one argument, it must be a dictionary mapping
Unicode ordinals (integers) or characters to Unicode ordinals,
strings or None. Character keys will be then converted to ordinals.
If there are two arguments, they must be strings of equal length,
and in the resulting dictionary, each character in x will be mapped
to the character at the same position in y. If there is a third
argument, it must be a string, whose characters will be mapped to
None in the result.
[clinic start generated code]*/
static PyObject *
unicode_maketrans_impl(PyObject *x, PyObject *y, PyObject *z)
/*[clinic end generated code: output=a925c89452bd5881 input=66bc00a1b4258a6e]*/
{
PyObject *new = NULL, *key, *value;
Py_ssize_t i = 0;
int res;
new = PyDict_New();
if (!new)
return NULL;
if (y != NULL) {
int x_kind, y_kind, z_kind;
const void *x_data, *y_data, *z_data;
/* x must be a string too, of equal length */
if (!PyUnicode_Check(x)) {
PyErr_SetString(PyExc_TypeError, "first maketrans argument must "
"be a string if there is a second argument");
goto err;
}
if (PyUnicode_GET_LENGTH(x) != PyUnicode_GET_LENGTH(y)) {
PyErr_SetString(PyExc_ValueError, "the first two maketrans "
"arguments must have equal length");
goto err;
}
/* create entries for translating chars in x to those in y */
x_kind = PyUnicode_KIND(x);
y_kind = PyUnicode_KIND(y);
x_data = PyUnicode_DATA(x);
y_data = PyUnicode_DATA(y);
for (i = 0; i < PyUnicode_GET_LENGTH(x); i++) {
key = PyLong_FromLong(PyUnicode_READ(x_kind, x_data, i));
if (!key)
goto err;
value = PyLong_FromLong(PyUnicode_READ(y_kind, y_data, i));
if (!value) {
Py_DECREF(key);
goto err;
}
res = PyDict_SetItem(new, key, value);
Py_DECREF(key);
Py_DECREF(value);
if (res < 0)
goto err;
}
/* create entries for deleting chars in z */
if (z != NULL) {
z_kind = PyUnicode_KIND(z);
z_data = PyUnicode_DATA(z);
for (i = 0; i < PyUnicode_GET_LENGTH(z); i++) {
key = PyLong_FromLong(PyUnicode_READ(z_kind, z_data, i));
if (!key)
goto err;
res = PyDict_SetItem(new, key, Py_None);
Py_DECREF(key);
if (res < 0)
goto err;
}
}
} else {
/* x must be a dict */
if (!PyAnyDict_CheckExact(x)) {
PyErr_SetString(PyExc_TypeError, "if you give only one argument "
"to maketrans it must be a dict");
goto err;
}
/* copy entries into the new dict, converting string keys to int keys */
int errcode;
Py_BEGIN_CRITICAL_SECTION(x);
errcode = unicode_maketrans_from_dict(x, new);
Py_END_CRITICAL_SECTION();
if (errcode < 0)
goto err;
}
return new;
err:
Py_DECREF(new);
return NULL;
}
/*[clinic input]
@permit_long_summary
str.translate as unicode_translate
table: object
Translation table, which must be a mapping of Unicode ordinals
to Unicode ordinals, strings, or None.
/
Replace each character in the string using the given translation table.
The table must implement lookup/indexing via __getitem__, for
instance a dictionary or list. If this operation raises
LookupError, the character is left untouched. Characters mapped to
None are deleted.
[clinic start generated code]*/
static PyObject *
unicode_translate(PyObject *self, PyObject *table)
/*[clinic end generated code: output=3cb448ff2fd96bf3 input=48cf0efe06bc1b75]*/
{
return _PyUnicode_TranslateCharmap(self, table, "ignore");
}
/*[clinic input]
str.upper as unicode_upper
Return a copy of the string converted to uppercase.
[clinic start generated code]*/
static PyObject *
unicode_upper_impl(PyObject *self)
/*[clinic end generated code: output=1b7ddd16bbcdc092 input=db3d55682dfe2e6c]*/
{
if (PyUnicode_IS_ASCII(self))
return ascii_upper_or_lower(self, 0);
return case_operation(self, do_upper);
}
/*[clinic input]
@permit_long_summary
str.zfill as unicode_zfill
width: Py_ssize_t
/
Pad a numeric string with zeros on the left, to fill a field of the given width.
The string is never truncated.
[clinic start generated code]*/
static PyObject *
unicode_zfill_impl(PyObject *self, Py_ssize_t width)
/*[clinic end generated code: output=e13fb6bdf8e3b9df input=25a4ee0ea3e58ce0]*/
{
Py_ssize_t fill;
PyObject *u;
int kind;
const void *data;
Py_UCS4 chr;
if (PyUnicode_GET_LENGTH(self) >= width)
return unicode_result_unchanged(self);
fill = width - PyUnicode_GET_LENGTH(self);
u = pad(self, fill, 0, '0');
if (u == NULL)
return NULL;
kind = PyUnicode_KIND(u);
data = PyUnicode_DATA(u);
chr = PyUnicode_READ(kind, data, fill);
if (chr == '+' || chr == '-') {
/* move sign to beginning of string */
PyUnicode_WRITE(kind, data, 0, chr);
PyUnicode_WRITE(kind, data, fill, '0');
}
assert(_PyUnicode_CheckConsistency(u, 1));
return u;
}
/*[clinic input]
@permit_long_summary
@text_signature "($self, prefix[, start[, end]], /)"
str.startswith as unicode_startswith
prefix as subobj: object
A string or a tuple of strings to try.
start: slice_index(accept={int, NoneType}, c_default='0') = None
Optional start position. Default: start of the string.
end: slice_index(accept={int, NoneType}, c_default='PY_SSIZE_T_MAX') = None
Optional stop position. Default: end of the string.
/
Return True if the string starts with the specified prefix, False otherwise.
[clinic start generated code]*/
static PyObject *
unicode_startswith_impl(PyObject *self, PyObject *subobj, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=4bd7cfd0803051d4 input=766bdbd33df251dc]*/
{
if (PyTuple_Check(subobj)) {
Py_ssize_t i;
for (i = 0; i < PyTuple_GET_SIZE(subobj); i++) {
PyObject *substring = PyTuple_GET_ITEM(subobj, i);
if (!PyUnicode_Check(substring)) {
PyErr_Format(PyExc_TypeError,
"tuple for startswith must only contain str, "
"not %.100s",
Py_TYPE(substring)->tp_name);
return NULL;
}
int result = tailmatch(self, substring, start, end, -1);
if (result < 0) {
return NULL;
}
if (result) {
Py_RETURN_TRUE;
}
}
/* nothing matched */
Py_RETURN_FALSE;
}
if (!PyUnicode_Check(subobj)) {
PyErr_Format(PyExc_TypeError,
"startswith first arg must be str or "
"a tuple of str, not %.100s", Py_TYPE(subobj)->tp_name);
return NULL;
}
int result = tailmatch(self, subobj, start, end, -1);
if (result < 0) {
return NULL;
}
return PyBool_FromLong(result);
}
/*[clinic input]
@permit_long_summary
@text_signature "($self, suffix[, start[, end]], /)"
str.endswith as unicode_endswith
suffix as subobj: object
A string or a tuple of strings to try.
start: slice_index(accept={int, NoneType}, c_default='0') = None
Optional start position. Default: start of the string.
end: slice_index(accept={int, NoneType}, c_default='PY_SSIZE_T_MAX') = None
Optional stop position. Default: end of the string.
/
Return True if the string ends with the specified suffix, False otherwise.
[clinic start generated code]*/
static PyObject *
unicode_endswith_impl(PyObject *self, PyObject *subobj, Py_ssize_t start,
Py_ssize_t end)
/*[clinic end generated code: output=cce6f8ceb0102ca9 input=b66bf6d5547ba1aa]*/
{
if (PyTuple_Check(subobj)) {
Py_ssize_t i;
for (i = 0; i < PyTuple_GET_SIZE(subobj); i++) {
PyObject *substring = PyTuple_GET_ITEM(subobj, i);
if (!PyUnicode_Check(substring)) {
PyErr_Format(PyExc_TypeError,
"tuple for endswith must only contain str, "
"not %.100s",
Py_TYPE(substring)->tp_name);
return NULL;
}
int result = tailmatch(self, substring, start, end, +1);
if (result < 0) {
return NULL;
}
if (result) {
Py_RETURN_TRUE;
}
}
Py_RETURN_FALSE;
}
if (!PyUnicode_Check(subobj)) {
PyErr_Format(PyExc_TypeError,
"endswith first arg must be str or "
"a tuple of str, not %.100s", Py_TYPE(subobj)->tp_name);
return NULL;
}
int result = tailmatch(self, subobj, start, end, +1);
if (result < 0) {
return NULL;
}
return PyBool_FromLong(result);
}
#include "stringlib/unicode_format.h"
PyDoc_STRVAR(format__doc__,
"format($self, /, *args, **kwargs)\n\
--\n\
\n\
Return a formatted version of the string, using substitutions from args and kwargs.\n\
The substitutions are identified by braces ('{' and '}').");
PyDoc_STRVAR(format_map__doc__,
"format_map($self, mapping, /)\n\
--\n\
\n\
Return a formatted version of the string, using substitutions from mapping.\n\
The substitutions are identified by braces ('{' and '}').");
/*[clinic input]
@permit_long_summary
str.__format__ as unicode___format__
format_spec: unicode
/
Return a formatted version of the string as described by format_spec.
[clinic start generated code]*/
static PyObject *
unicode___format___impl(PyObject *self, PyObject *format_spec)
/*[clinic end generated code: output=45fceaca6d2ba4c8 input=77a2a19f3f7969f2]*/
{
_PyUnicodeWriter writer;
int ret;
_PyUnicodeWriter_Init(&writer);
ret = _PyUnicode_FormatAdvancedWriter(&writer,
self, format_spec, 0,
PyUnicode_GET_LENGTH(format_spec));
if (ret == -1) {
_PyUnicodeWriter_Dealloc(&writer);
return NULL;
}
return _PyUnicodeWriter_Finish(&writer);
}
/*[clinic input]
str.__sizeof__ as unicode_sizeof
Return the size of the string in memory, in bytes.
[clinic start generated code]*/
static PyObject *
unicode_sizeof_impl(PyObject *self)
/*[clinic end generated code: output=6dbc2f5a408b6d4f input=6dd011c108e33fb0]*/
{
Py_ssize_t size;
/* If it's a compact object, account for base structure +
character data. */
if (PyUnicode_IS_COMPACT_ASCII(self)) {
size = sizeof(PyASCIIObject) + PyUnicode_GET_LENGTH(self) + 1;
}
else if (PyUnicode_IS_COMPACT(self)) {
size = sizeof(PyCompactUnicodeObject) +
(PyUnicode_GET_LENGTH(self) + 1) * PyUnicode_KIND(self);
}
else {
/* If it is a two-block object, account for base object, and
for character block if present. */
size = sizeof(PyUnicodeObject);
if (_PyUnicode_DATA_ANY(self))
size += (PyUnicode_GET_LENGTH(self) + 1) *
PyUnicode_KIND(self);
}
if (_PyUnicode_HAS_UTF8_MEMORY(self))
size += PyUnicode_UTF8_LENGTH(self) + 1;
return PyLong_FromSsize_t(size);
}
static PyObject *
unicode_getnewargs(PyObject *v, PyObject *Py_UNUSED(ignored))
{
PyObject *copy = _PyUnicode_Copy(v);
if (!copy)
return NULL;
return Py_BuildValue("(N)", copy);
}
/*
This function searchs the longest common leading whitespace
of all lines in the [src, end).
It returns the length of the common leading whitespace and sets `output` to
point to the beginning of the common leading whitespace if length > 0.
*/
static Py_ssize_t
search_longest_common_leading_whitespace(
const char *const src,
const char *const end,
const char **output)
{
// [_start, _start + _len)
// describes the current longest common leading whitespace
const char *_start = NULL;
Py_ssize_t _len = 0;
for (const char *iter = src; iter < end; ++iter) {
const char *line_start = iter;
const char *leading_whitespace_end = NULL;
// scan the whole line
while (iter < end && *iter != '\n') {
if (!leading_whitespace_end && *iter != ' ' && *iter != '\t') {
/* `iter` points to the first non-whitespace character
in this line */
if (iter == line_start) {
// some line has no indent, fast exit!
return 0;
}
leading_whitespace_end = iter;
}
++iter;
}
// if this line has all white space, skip it
if (!leading_whitespace_end) {
continue;
}
if (!_start) {
// update the first leading whitespace
_start = line_start;
_len = leading_whitespace_end - line_start;
assert(_len > 0);
}
else {
/* We then compare with the current longest leading whitespace.
[line_start, leading_whitespace_end) is the leading
whitespace of this line,
[_start, _start + _len) is the leading whitespace of the
current longest leading whitespace. */
Py_ssize_t new_len = 0;
const char *_iter = _start, *line_iter = line_start;
while (_iter < _start + _len && line_iter < leading_whitespace_end
&& *_iter == *line_iter)
{
++_iter;
++line_iter;
++new_len;
}
_len = new_len;
if (_len == 0) {
// No common things now, fast exit!
return 0;
}
}
}
assert(_len >= 0);
if (_len > 0) {
*output = _start;
}
return _len;
}
/* Dedent a string.
Intended to dedent Python source. Unlike `textwrap.dedent`, this
only supports spaces and tabs and doesn't normalize empty lines.
Return a new reference on success, NULL with exception set on error.
*/
PyObject *
_PyUnicode_Dedent(PyObject *unicode)
{
Py_ssize_t src_len = 0;
const char *src = PyUnicode_AsUTF8AndSize(unicode, &src_len);
if (!src) {
return NULL;
}
assert(src_len >= 0);
if (src_len == 0) {
return Py_NewRef(unicode);
}
const char *const end = src + src_len;
// [whitespace_start, whitespace_start + whitespace_len)
// describes the current longest common leading whitespace
const char *whitespace_start = NULL;
Py_ssize_t whitespace_len = search_longest_common_leading_whitespace(
src, end, &whitespace_start);
if (whitespace_len == 0) {
return Py_NewRef(unicode);
}
// now we should trigger a dedent
char *dest = PyMem_Malloc(src_len);
if (!dest) {
PyErr_NoMemory();
return NULL;
}
char *dest_iter = dest;
for (const char *iter = src; iter < end; ++iter) {
const char *line_start = iter;
bool in_leading_space = true;
// iterate over a line to find the end of a line
while (iter < end && *iter != '\n') {
if (in_leading_space && *iter != ' ' && *iter != '\t') {
in_leading_space = false;
}
++iter;
}
// invariant: *iter == '\n' or iter == end
bool append_newline = iter < end;
// if this line has all white space, write '\n' and continue
if (in_leading_space && append_newline) {
*dest_iter++ = '\n';
continue;
}
/* copy [new_line_start + whitespace_len, iter) to buffer, then
conditionally append '\n' */
Py_ssize_t new_line_len = iter - line_start - whitespace_len;
assert(new_line_len >= 0);
memcpy(dest_iter, line_start + whitespace_len, new_line_len);
dest_iter += new_line_len;
if (append_newline) {
*dest_iter++ = '\n';
}
}
PyObject *res = PyUnicode_FromStringAndSize(dest, dest_iter - dest);
PyMem_Free(dest);
return res;
}
static PyMethodDef unicode_methods[] = {
UNICODE_ENCODE_METHODDEF
UNICODE_REPLACE_METHODDEF
UNICODE_SPLIT_METHODDEF
UNICODE_RSPLIT_METHODDEF
UNICODE_JOIN_METHODDEF
UNICODE_CAPITALIZE_METHODDEF
UNICODE_CASEFOLD_METHODDEF
UNICODE_TITLE_METHODDEF
UNICODE_CENTER_METHODDEF
UNICODE_COUNT_METHODDEF
UNICODE_EXPANDTABS_METHODDEF
UNICODE_FIND_METHODDEF
UNICODE_PARTITION_METHODDEF
UNICODE_INDEX_METHODDEF
UNICODE_LJUST_METHODDEF
UNICODE_LOWER_METHODDEF
UNICODE_LSTRIP_METHODDEF
UNICODE_RFIND_METHODDEF
UNICODE_RINDEX_METHODDEF
UNICODE_RJUST_METHODDEF
UNICODE_RSTRIP_METHODDEF
UNICODE_RPARTITION_METHODDEF
UNICODE_SPLITLINES_METHODDEF
UNICODE_STRIP_METHODDEF
UNICODE_SWAPCASE_METHODDEF
UNICODE_TRANSLATE_METHODDEF
UNICODE_UPPER_METHODDEF
UNICODE_STARTSWITH_METHODDEF
UNICODE_ENDSWITH_METHODDEF
UNICODE_REMOVEPREFIX_METHODDEF
UNICODE_REMOVESUFFIX_METHODDEF
UNICODE_ISASCII_METHODDEF
UNICODE_ISLOWER_METHODDEF
UNICODE_ISUPPER_METHODDEF
UNICODE_ISTITLE_METHODDEF
UNICODE_ISSPACE_METHODDEF
UNICODE_ISDECIMAL_METHODDEF
UNICODE_ISDIGIT_METHODDEF
UNICODE_ISNUMERIC_METHODDEF
UNICODE_ISALPHA_METHODDEF
UNICODE_ISALNUM_METHODDEF
UNICODE_ISIDENTIFIER_METHODDEF
UNICODE_ISPRINTABLE_METHODDEF
UNICODE_ZFILL_METHODDEF
{"format", _PyCFunction_CAST(do_string_format), METH_VARARGS | METH_KEYWORDS, format__doc__},
{"format_map", do_string_format_map, METH_O, format_map__doc__},
UNICODE___FORMAT___METHODDEF
UNICODE_MAKETRANS_METHODDEF
UNICODE_SIZEOF_METHODDEF
{"__getnewargs__", unicode_getnewargs, METH_NOARGS},
{NULL, NULL}
};
static PyObject *
unicode_mod(PyObject *v, PyObject *w)
{
if (!PyUnicode_Check(v))
Py_RETURN_NOTIMPLEMENTED;
return PyUnicode_Format(v, w);
}
static PyNumberMethods unicode_as_number = {
0, /*nb_add*/
0, /*nb_subtract*/
0, /*nb_multiply*/
unicode_mod, /*nb_remainder*/
};
static PySequenceMethods unicode_as_sequence = {
unicode_length, /* sq_length */
PyUnicode_Concat, /* sq_concat */
_PyUnicode_Repeat, /* sq_repeat */
unicode_getitem, /* sq_item */
0, /* sq_slice */
0, /* sq_ass_item */
0, /* sq_ass_slice */
PyUnicode_Contains, /* sq_contains */
};
static PyObject*
unicode_subscript(PyObject* self, PyObject* item)
{
if (_PyIndex_Check(item)) {
Py_ssize_t i = PyNumber_AsSsize_t(item, PyExc_IndexError);
if (i == -1 && PyErr_Occurred())
return NULL;
if (i < 0)
i += PyUnicode_GET_LENGTH(self);
return unicode_getitem(self, i);
} else if (PySlice_Check(item)) {
Py_ssize_t start, stop, step, slicelength, i;
size_t cur;
PyObject *result;
const void *src_data;
void *dest_data;
int src_kind, dest_kind;
Py_UCS4 ch, max_char, kind_limit;
if (PySlice_Unpack(item, &start, &stop, &step) < 0) {
return NULL;
}
slicelength = PySlice_AdjustIndices(PyUnicode_GET_LENGTH(self),
&start, &stop, step);
if (slicelength max_char) {
max_char = ch;
if (max_char >= kind_limit)
break;
}
}
}
else
max_char = 127;
result = PyUnicode_New(slicelength, max_char);
if (result == NULL)
return NULL;
dest_kind = PyUnicode_KIND(result);
dest_data = PyUnicode_DATA(result);
for (cur = start, i = 0; i < slicelength; cur += step, i++) {
Py_UCS4 ch = PyUnicode_READ(src_kind, src_data, cur);
PyUnicode_WRITE(dest_kind, dest_data, i, ch);
}
assert(_PyUnicode_CheckConsistency(result, 1));
return result;
} else {
PyErr_Format(PyExc_TypeError, "string indices must be integers, not '%.200s'",
Py_TYPE(item)->tp_name);
return NULL;
}
}
static PyMappingMethods unicode_as_mapping = {
unicode_length, /* mp_length */
unicode_subscript, /* mp_subscript */
0, /* mp_ass_subscript */
};
static PyObject *
unicode_subtype_new(PyTypeObject *type, PyObject *unicode);
/*[clinic input]
@classmethod
str.__new__ as unicode_new
object as x: object = NULL
encoding: str = NULL
errors: str = NULL
[clinic start generated code]*/
static PyObject *
unicode_new_impl(PyTypeObject *type, PyObject *x, const char *encoding,
const char *errors)
/*[clinic end generated code: output=fc72d4878b0b57e9 input=e81255e5676d174e]*/
{
PyObject *unicode;
if (x == NULL) {
unicode = _PyUnicode_GetEmpty();
}
else if (encoding == NULL && errors == NULL) {
unicode = PyObject_Str(x);
}
else {
unicode = PyUnicode_FromEncodedObject(x, encoding, errors);
}
if (unicode != NULL && type != &PyUnicode_Type) {
Py_SETREF(unicode, unicode_subtype_new(type, unicode));
}
return unicode;
}
static const char *
arg_as_utf8(PyObject *obj, const char *name)
{
if (!PyUnicode_Check(obj)) {
PyErr_Format(PyExc_TypeError,
"str() argument '%s' must be str, not %T",
name, obj);
return NULL;
}
return _PyUnicode_AsUTF8NoNUL(obj);
}
static PyObject *
unicode_vectorcall(PyObject *type, PyObject *const *args,
size_t nargsf, PyObject *kwnames)
{
assert(Py_Is(_PyType_CAST(type), &PyUnicode_Type));
Py_ssize_t nargs = PyVectorcall_NARGS(nargsf);
if (kwnames != NULL && PyTuple_GET_SIZE(kwnames) != 0) {
// Fallback to unicode_new()
PyObject *tuple = PyTuple_FromArray(args, nargs);
if (tuple == NULL) {
return NULL;
}
PyObject *dict = _PyStack_AsDict(args + nargs, kwnames);
if (dict == NULL) {
Py_DECREF(tuple);
return NULL;
}
PyObject *ret = unicode_new(_PyType_CAST(type), tuple, dict);
Py_DECREF(tuple);
Py_DECREF(dict);
return ret;
}
if (!_PyArg_CheckPositional("str", nargs, 0, 3)) {
return NULL;
}
if (nargs == 0) {
return _PyUnicode_GetEmpty();
}
PyObject *object = args[0];
if (nargs == 1) {
return PyObject_Str(object);
}
const char *encoding = arg_as_utf8(args[1], "encoding");
if (encoding == NULL) {
return NULL;
}
const char *errors = NULL;
if (nargs == 3) {
errors = arg_as_utf8(args[2], "errors");
if (errors == NULL) {
return NULL;
}
}
return PyUnicode_FromEncodedObject(object, encoding, errors);
}
static PyObject *
unicode_subtype_new(PyTypeObject *type, PyObject *unicode)
{
PyObject *self;
Py_ssize_t length, char_size;
int share_utf8;
int kind;
void *data;
assert(PyType_IsSubtype(type, &PyUnicode_Type));
assert(_PyUnicode_CHECK(unicode));
self = type->tp_alloc(type, 0);
if (self == NULL) {
return NULL;
}
kind = PyUnicode_KIND(unicode);
length = PyUnicode_GET_LENGTH(unicode);
_PyUnicode_LENGTH(self) = length;
#ifdef Py_DEBUG
_PyUnicode_HASH(self) = -1;
#else
_PyUnicode_HASH(self) = _PyUnicode_HASH(unicode);
#endif
_PyUnicode_STATE(self).interned = 0;
_PyUnicode_STATE(self).kind = kind;
_PyUnicode_STATE(self).compact = 0;
_PyUnicode_STATE(self).ascii = _PyUnicode_STATE(unicode).ascii;
_PyUnicode_STATE(self).statically_allocated = 0;
PyUnicode_SET_UTF8_LENGTH(self, 0);
PyUnicode_SET_UTF8(self, NULL);
_PyUnicode_DATA_ANY(self) = NULL;
share_utf8 = 0;
if (kind == PyUnicode_1BYTE_KIND) {
char_size = 1;
if (PyUnicode_MAX_CHAR_VALUE(unicode) < 128)
share_utf8 = 1;
}
else if (kind == PyUnicode_2BYTE_KIND) {
char_size = 2;
}
else {
assert(kind == PyUnicode_4BYTE_KIND);
char_size = 4;
}
/* Ensure we won't overflow the length. */
if (length > (PY_SSIZE_T_MAX / char_size - 1)) {
PyErr_NoMemory();
goto onError;
}
data = PyMem_Malloc((length + 1) * char_size);
if (data == NULL) {
PyErr_NoMemory();
goto onError;
}
_PyUnicode_DATA_ANY(self) = data;
if (share_utf8) {
PyUnicode_SET_UTF8_LENGTH(self, length);
PyUnicode_SET_UTF8(self, data);
}
memcpy(data, PyUnicode_DATA(unicode), kind * (length + 1));
assert(_PyUnicode_CheckConsistency(self, 1));
#ifdef Py_DEBUG
_PyUnicode_HASH(self) = _PyUnicode_HASH(unicode);
#endif
return self;
onError:
Py_DECREF(self);
return NULL;
}
static _PyObjectIndexPair
unicode_iteritem(PyObject *obj, Py_ssize_t index)
{
if (index >= PyUnicode_GET_LENGTH(obj)) {
return (_PyObjectIndexPair) { .object = NULL, .index = index };
}
const void *data = PyUnicode_DATA(obj);
int kind = PyUnicode_KIND(obj);
Py_UCS4 ch = PyUnicode_READ(kind, data, index);
PyObject *result = unicode_char(ch);
index = (result == NULL) ? -1 : index + 1;
return (_PyObjectIndexPair) { .object = result, .index = index };
}
void
_PyUnicode_ExactDealloc(PyObject *op)
{
assert(PyUnicode_CheckExact(op));
unicode_dealloc(op);
}
PyDoc_STRVAR(unicode_doc,
"str(object='') -> str\n\
str(bytes_or_buffer[, encoding[, errors]]) -> str\n\
\n\
Create a new string object from the given object. If encoding or\n\
errors is specified, then the object must expose a data buffer\n\
that will be decoded using the given encoding and error handler.\n\
Otherwise, returns the result of object.__str__() (if defined)\n\
or repr(object).\n\
encoding defaults to 'utf-8'.\n\
errors defaults to 'strict'.");
static PyObject *unicode_iter(PyObject *seq);
PyTypeObject PyUnicode_Type = {
PyVarObject_HEAD_INIT(&PyType_Type, 0)
"str", /* tp_name */
sizeof(PyUnicodeObject), /* tp_basicsize */
0, /* tp_itemsize */
/* Slots */
unicode_dealloc, /* tp_dealloc */
0, /* tp_vectorcall_offset */
0, /* tp_getattr */
0, /* tp_setattr */
0, /* tp_as_async */
unicode_repr, /* tp_repr */
&unicode_as_number, /* tp_as_number */
&unicode_as_sequence, /* tp_as_sequence */
&unicode_as_mapping, /* tp_as_mapping */
unicode_hash, /* tp_hash*/
0, /* tp_call*/
unicode_str, /* tp_str */
PyObject_GenericGetAttr, /* tp_getattro */
0, /* tp_setattro */
0, /* tp_as_buffer */
Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE |
Py_TPFLAGS_UNICODE_SUBCLASS |
_Py_TPFLAGS_MATCH_SELF, /* tp_flags */
unicode_doc, /* tp_doc */
0, /* tp_traverse */
0, /* tp_clear */
PyUnicode_RichCompare, /* tp_richcompare */
0, /* tp_weaklistoffset */
unicode_iter, /* tp_iter */
0, /* tp_iternext */
unicode_methods, /* tp_methods */
0, /* tp_members */
0, /* tp_getset */
0, /* tp_base */
0, /* tp_dict */
0, /* tp_descr_get */
0, /* tp_descr_set */
0, /* tp_dictoffset */
0, /* tp_init */
0, /* tp_alloc */
unicode_new, /* tp_new */
PyObject_Free, /* tp_free */
.tp_vectorcall = unicode_vectorcall,
._tp_iteritem = unicode_iteritem,
};
/* Initialize the Unicode implementation */
static void
_init_global_state(void)
{
static int initialized = 0;
if (initialized) {
return;
}
initialized = 1;
/* initialize the linebreak bloom filter */
const Py_UCS2 linebreak[] = {
0x000A, /* LINE FEED */
0x000D, /* CARRIAGE RETURN */
0x001C, /* FILE SEPARATOR */
0x001D, /* GROUP SEPARATOR */
0x001E, /* RECORD SEPARATOR */
0x0085, /* NEXT LINE */
0x2028, /* LINE SEPARATOR */
0x2029, /* PARAGRAPH SEPARATOR */
};
bloom_linebreak = make_bloom_mask(
PyUnicode_2BYTE_KIND, linebreak,
Py_ARRAY_LENGTH(linebreak));
}
void
_PyUnicode_InitState(PyInterpreterState *interp)
{
if (!_Py_IsMainInterpreter(interp)) {
return;
}
_init_global_state();
}
PyStatus
_PyUnicode_InitGlobalObjects(PyInterpreterState *interp)
{
if (_Py_IsMainInterpreter(interp)) {
PyStatus status = init_global_interned_strings(interp);
if (_PyStatus_EXCEPTION(status)) {
return status;
}
}
assert(INTERNED_STRINGS);
if (init_interned_dict(interp)) {
PyErr_Clear();
return _PyStatus_ERR("failed to create interned dict");
}
return _PyStatus_OK();
}
PyStatus
_PyUnicode_InitTypes(PyInterpreterState *interp)
{
if (_PyStaticType_InitBuiltin(interp, &EncodingMapType) < 0) {
goto error;
}
if (_PyStaticType_InitBuiltin(interp, &PyFieldNameIter_Type) < 0) {
goto error;
}
if (_PyStaticType_InitBuiltin(interp, &PyFormatterIter_Type) < 0) {
goto error;
}
return _PyStatus_OK();
error:
return _PyStatus_ERR("Can't initialize unicode types");
}
static /* non-null */ PyObject*
intern_static(PyInterpreterState *interp, PyObject *s /* stolen */)
{
// Note that this steals a reference to `s`, but in many cases that
// stolen ref is returned, requiring no decref/incref.
assert(s != NULL);
assert(_PyUnicode_CHECK(s));
assert(_PyUnicode_STATE(s).statically_allocated);
assert(!PyUnicode_CHECK_INTERNED(s));
#ifdef Py_DEBUG
/* We must not add process-global interned string if there's already a
* per-interpreter interned_dict, which might contain duplicates.
*/
PyObject *interned = get_interned_dict(interp);
assert(interned == NULL);
#endif
/* Look in the global cache first. */
PyObject *r = (PyObject *)_Py_hashtable_get(INTERNED_STRINGS, s);
/* We should only init each string once */
assert(r == NULL);
/* but just in case (for the non-debug build), handle this */
if (r != NULL && r != s) {
assert(_PyUnicode_STATE(r).interned == SSTATE_INTERNED_IMMORTAL_STATIC);
assert(_PyUnicode_CHECK(r));
Py_DECREF(s);
return Py_NewRef(r);
}
if (_Py_hashtable_set(INTERNED_STRINGS, s, s) < -1) {
Py_FatalError("failed to intern static string");
}
_PyUnicode_STATE(s).interned = SSTATE_INTERNED_IMMORTAL_STATIC;
return s;
}
void
_PyUnicode_InternStatic(PyInterpreterState *interp, PyObject **p)
{
// This should only be called as part of runtime initialization
assert(!Py_IsInitialized());
*p = intern_static(interp, *p);
assert(*p);
}
static void
immortalize_interned(PyObject *s)
{
assert(PyUnicode_CHECK_INTERNED(s) == SSTATE_INTERNED_MORTAL);
assert(!_Py_IsImmortal(s));
#ifdef Py_REF_DEBUG
/* The reference count value should be excluded from the RefTotal.
The decrements to these objects will not be registered so they
need to be accounted for in here. */
for (Py_ssize_t i = 0; i < Py_REFCNT(s); i++) {
_Py_DecRefTotal(_PyThreadState_GET());
}
#endif
_Py_SetImmortal(s);
// The switch to SSTATE_INTERNED_IMMORTAL must be the last thing done here
// to synchronize with the check in intern_common() that avoids locking if
// the string is already immortal.
FT_ATOMIC_STORE_UINT8(_PyUnicode_STATE(s).interned, SSTATE_INTERNED_IMMORTAL);
}
#ifdef Py_GIL_DISABLED
static bool
can_immortalize_safely(PyObject *s)
{
if (_Py_IsOwnedByCurrentThread(s) || _Py_IsImmortal(s)) {
return true;
}
Py_ssize_t shared = _Py_atomic_load_ssize(&s->ob_ref_shared);
return _Py_REF_IS_MERGED(shared);
}
#endif
static /* non-null */ PyObject*
intern_common(PyInterpreterState *interp, PyObject *s /* stolen */,
bool immortalize)
{
// Note that this steals a reference to `s`, but in many cases that
// stolen ref is returned, requiring no decref/incref.
#ifdef Py_DEBUG
assert(s != NULL);
assert(_PyUnicode_CHECK(s));
#else
if (s == NULL || !PyUnicode_Check(s)) {
return s;
}
#endif
/* If it's a subclass, we don't really know what putting
it in the interned dict might do. */
if (!PyUnicode_CheckExact(s)) {
return s;
}
/* Is it already interned? */
switch (PyUnicode_CHECK_INTERNED(s)) {
case SSTATE_NOT_INTERNED:
// no, go on
break;
case SSTATE_INTERNED_MORTAL:
#ifndef Py_GIL_DISABLED
// yes but we might need to make it immortal
if (immortalize) {
immortalize_interned(s);
}
return s;
#else
// not fully interned yet; fall through to the locking path
break;
#endif
default:
// all done
return s;
}
/* Statically allocated strings must be already interned. */
assert(!_PyUnicode_STATE(s).statically_allocated);
#if Py_GIL_DISABLED
/* In the free-threaded build, all interned strings are immortal */
immortalize = 1;
#endif
/* If it's already immortal, intern it as such */
if (_Py_IsImmortal(s)) {
immortalize = 1;
}
/* if it's a short string, get the singleton */
if (PyUnicode_GET_LENGTH(s) == 1 &&
PyUnicode_KIND(s) == PyUnicode_1BYTE_KIND) {
PyObject *r = LATIN1(*(unsigned char*)PyUnicode_DATA(s));
assert(PyUnicode_CHECK_INTERNED(r));
Py_DECREF(s);
return r;
}
#ifdef Py_DEBUG
assert(!unicode_is_singleton(s));
#endif
/* Look in the global cache now. */
{
PyObject *r = (PyObject *)_Py_hashtable_get(INTERNED_STRINGS, s);
if (r != NULL) {
assert(_PyUnicode_STATE(r).statically_allocated);
assert(r != s); // r must be statically_allocated; s is not
Py_DECREF(s);
return Py_NewRef(r);
}
}
/* Do a setdefault on the per-interpreter cache. */
PyObject *interned = get_interned_dict(interp);
assert(interned != NULL);
#ifdef Py_GIL_DISABLED
# define INTERN_MUTEX &_Py_INTERP_CACHED_OBJECT(interp, interned_mutex)
// Lock-free fast path: check if there's already an interned copy that
// is in its final immortal state.
PyObject *r;
int res = PyDict_GetItemRef(interned, s, &r);
if (res < 0) {
PyErr_Clear();
return s;
}
if (res > 0) {
unsigned int state = _Py_atomic_load_uint8(&_PyUnicode_STATE(r).interned);
if (state == SSTATE_INTERNED_IMMORTAL) {
Py_DECREF(s);
return r;
}
// Not yet fully interned; fall through to the locking path.
Py_DECREF(r);
}
#endif
#ifdef Py_GIL_DISABLED
// Immortalization writes to the refcount fields non-atomically. That
// races with Py_INCREF / Py_DECREF on the thread that owns `s`. If we
// don't own it (and its refcount hasn't been merged), intern a copy
// we own instead.
if (!can_immortalize_safely(s)) {
PyObject *copy = _PyUnicode_Copy(s);
if (copy == NULL) {
PyErr_Clear();
return s;
}
Py_DECREF(s);
s = copy;
}
#endif
// Why _Py_LOCK_DONT_DETACH is used here: waiting for the interned mutex
// must not detach the thread state. Extension code is expected to
// detach before blocking on opaque external synchronization. However,
// the lock used for C++ static initialization is hidden, making
// that difficult, and it is common for C++ extensions to call
// PyUnicode_InternFromString() from static initializers. Detaching here
// can therefore deadlock: a stop-the-world pause may prevent the lock
// owner from reattaching while the pause waits for another attached
// thread blocked on the hidden lock.
FT_MUTEX_LOCK_FLAGS(INTERN_MUTEX, _Py_LOCK_DONT_DETACH);
PyObject *t;
{
int res = PyDict_SetDefaultRef(interned, s, s, &t);
if (res < 0) {
PyErr_Clear();
FT_MUTEX_UNLOCK(INTERN_MUTEX);
return s;
}
else if (res == 1) {
// value was already present (not inserted)
Py_DECREF(s);
if (immortalize &&
PyUnicode_CHECK_INTERNED(t) == SSTATE_INTERNED_MORTAL) {
immortalize_interned(t);
}
FT_MUTEX_UNLOCK(INTERN_MUTEX);
return t;
}
else {
// value was newly inserted
assert (s == t);
Py_DECREF(t);
}
}
/* NOT_INTERNED -> INTERNED_MORTAL */
assert(_PyUnicode_STATE(s).interned == SSTATE_NOT_INTERNED);
if (!_Py_IsImmortal(s)) {
/* The two references in interned dict (key and value) are not counted.
unicode_dealloc() and _PyUnicode_ClearInterned() take care of this. */
Py_DECREF(s);
Py_DECREF(s);
}
FT_ATOMIC_STORE_UINT8(_PyUnicode_STATE(s).interned, SSTATE_INTERNED_MORTAL);
/* INTERNED_MORTAL -> INTERNED_IMMORTAL (if needed) */
#ifdef Py_DEBUG
if (_Py_IsImmortal(s)) {
assert(immortalize);
}
#endif
if (immortalize) {
immortalize_interned(s);
}
FT_MUTEX_UNLOCK(INTERN_MUTEX);
return s;
}
void
_PyUnicode_InternImmortal(PyInterpreterState *interp, PyObject **p)
{
*p = intern_common(interp, *p, 1);
assert(*p);
}
void
_PyUnicode_InternMortal(PyInterpreterState *interp, PyObject **p)
{
*p = intern_common(interp, *p, 0);
assert(*p);
}
void
_PyUnicode_InternInPlace(PyInterpreterState *interp, PyObject **p)
{
_PyUnicode_InternImmortal(interp, p);
return;
}
void
PyUnicode_InternInPlace(PyObject **p)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
_PyUnicode_InternMortal(interp, p);
}
// Public-looking name kept for the stable ABI; user should not call this:
PyAPI_FUNC(void) PyUnicode_InternImmortal(PyObject **);
void
PyUnicode_InternImmortal(PyObject **p)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
_PyUnicode_InternImmortal(interp, p);
}
PyObject *
PyUnicode_InternFromString(const char *cp)
{
PyObject *s = PyUnicode_FromString(cp);
if (s == NULL) {
return NULL;
}
PyInterpreterState *interp = _PyInterpreterState_GET();
_PyUnicode_InternMortal(interp, &s);
return s;
}
void
_PyUnicode_ClearInterned(PyInterpreterState *interp)
{
PyObject *interned = get_interned_dict(interp);
if (interned == NULL) {
return;
}
assert(PyDict_CheckExact(interned));
if (has_shared_intern_dict(interp)) {
// the dict doesn't belong to this interpreter, skip the debug
// checks on it and just clear the pointer to it
clear_interned_dict(interp);
return;
}
#ifdef INTERNED_STATS
fprintf(stderr, "releasing %zd interned strings\n",
PyDict_GET_SIZE(interned));
Py_ssize_t total_length = 0;
#endif
Py_ssize_t pos = 0;
PyObject *s, *ignored_value;
while (PyDict_Next(interned, &pos, &s, &ignored_value)) {
int shared = 0;
switch (PyUnicode_CHECK_INTERNED(s)) {
case SSTATE_INTERNED_IMMORTAL:
/* Make immortal interned strings mortal again. */
// Skip the Immortal Instance check and restore
// the two references (key and value) ignored
// by PyUnicode_InternInPlace().
_Py_SetMortal(s, 2);
#ifdef Py_REF_DEBUG
/* let's be pedantic with the ref total */
_Py_IncRefTotal(_PyThreadState_GET());
_Py_IncRefTotal(_PyThreadState_GET());
#endif
#ifdef INTERNED_STATS
total_length += PyUnicode_GET_LENGTH(s);
#endif
break;
case SSTATE_INTERNED_IMMORTAL_STATIC:
/* It is shared between interpreters, so we should unmark it
only when this is the last interpreter in which it's
interned. We immortalize all the statically initialized
strings during startup, so we can rely on the
main interpreter to be the last one. */
if (!_Py_IsMainInterpreter(interp)) {
shared = 1;
}
break;
case SSTATE_INTERNED_MORTAL:
// Restore 2 references held by the interned dict; these will
// be decref'd by clear_interned_dict's PyDict_Clear.
_Py_RefcntAdd(s, 2);
#ifdef Py_REF_DEBUG
/* let's be pedantic with the ref total */
_Py_IncRefTotal(_PyThreadState_GET());
_Py_IncRefTotal(_PyThreadState_GET());
#endif
break;
case SSTATE_NOT_INTERNED:
_Py_FALLTHROUGH;
default:
Py_UNREACHABLE();
}
if (!shared) {
FT_ATOMIC_STORE_UINT8_RELAXED(_PyUnicode_STATE(s).interned, SSTATE_NOT_INTERNED);
}
}
#ifdef INTERNED_STATS
fprintf(stderr,
"total length of all interned strings: %zd characters\n",
total_length);
#endif
struct _Py_unicode_state *state = &interp->unicode;
struct _Py_unicode_ids *ids = &state->ids;
for (Py_ssize_t i=0; i < ids->size; i++) {
Py_XINCREF(ids->array[i]);
}
clear_interned_dict(interp);
if (_Py_IsMainInterpreter(interp)) {
clear_global_interned_strings();
}
}
/********************* Unicode Iterator **************************/
typedef struct {
PyObject_HEAD
Py_ssize_t it_index;
PyObject *it_seq; /* Set to NULL when iterator is exhausted */
} unicodeiterobject;
static void
unicodeiter_dealloc(PyObject *op)
{
unicodeiterobject *it = (unicodeiterobject *)op;
_PyObject_GC_UNTRACK(it);
Py_XDECREF(it->it_seq);
PyObject_GC_Del(it);
}
static int
unicodeiter_traverse(PyObject *op, visitproc visit, void *arg)
{
unicodeiterobject *it = (unicodeiterobject *)op;
Py_VISIT(it->it_seq);
return 0;
}
static PyObject *
unicodeiter_next(PyObject *op)
{
unicodeiterobject *it = (unicodeiterobject *)op;
PyObject *seq;
assert(it != NULL);
seq = it->it_seq;
if (seq == NULL)
return NULL;
assert(_PyUnicode_CHECK(seq));
if (it->it_index < PyUnicode_GET_LENGTH(seq)) {
int kind = PyUnicode_KIND(seq);
const void *data = PyUnicode_DATA(seq);
Py_UCS4 chr = PyUnicode_READ(kind, data, it->it_index);
it->it_index++;
return unicode_char(chr);
}
it->it_seq = NULL;
Py_DECREF(seq);
return NULL;
}
static PyObject *
unicode_ascii_iter_next(PyObject *op)
{
unicodeiterobject *it = (unicodeiterobject *)op;
assert(it != NULL);
PyObject *seq = it->it_seq;
if (seq == NULL) {
return NULL;
}
assert(_PyUnicode_CHECK(seq));
assert(PyUnicode_IS_COMPACT_ASCII(seq));
if (it->it_index < PyUnicode_GET_LENGTH(seq)) {
const void *data = ((void*)(_PyASCIIObject_CAST(seq) + 1));
Py_UCS1 chr = (Py_UCS1)PyUnicode_READ(PyUnicode_1BYTE_KIND,
data, it->it_index);
it->it_index++;
return (PyObject*)&_Py_SINGLETON(strings).ascii[chr];
}
it->it_seq = NULL;
Py_DECREF(seq);
return NULL;
}
static PyObject *
unicodeiter_len(PyObject *op, PyObject *Py_UNUSED(ignored))
{
unicodeiterobject *it = (unicodeiterobject *)op;
Py_ssize_t len = 0;
if (it->it_seq)
len = PyUnicode_GET_LENGTH(it->it_seq) - it->it_index;
return PyLong_FromSsize_t(len);
}
PyDoc_STRVAR(length_hint_doc, "Private method returning an estimate of len(list(it)).");
static PyObject *
unicodeiter_reduce(PyObject *op, PyObject *Py_UNUSED(ignored))
{
unicodeiterobject *it = (unicodeiterobject *)op;
PyObject *iter = _PyEval_GetBuiltin(&_Py_ID(iter));
/* _PyEval_GetBuiltin can invoke arbitrary code,
* call must be before access of iterator pointers.
* see issue #101765 */
if (it->it_seq != NULL) {
return Py_BuildValue("N(O)n", iter, it->it_seq, it->it_index);
} else {
PyObject *u = _PyUnicode_GetEmpty();
if (u == NULL) {
Py_XDECREF(iter);
return NULL;
}
return Py_BuildValue("N(N)", iter, u);
}
}
PyDoc_STRVAR(reduce_doc, "Return state information for pickling.");
static PyObject *
unicodeiter_setstate(PyObject *op, PyObject *state)
{
unicodeiterobject *it = (unicodeiterobject *)op;
Py_ssize_t index = PyLong_AsSsize_t(state);
if (index == -1 && PyErr_Occurred())
return NULL;
if (it->it_seq != NULL) {
if (index < 0)
index = 0;
else if (index > PyUnicode_GET_LENGTH(it->it_seq))
index = PyUnicode_GET_LENGTH(it->it_seq); /* iterator truncated */
it->it_index = index;
}
Py_RETURN_NONE;
}
PyDoc_STRVAR(setstate_doc, "Set state information for unpickling.");
static PyMethodDef unicodeiter_methods[] = {
{"__length_hint__", unicodeiter_len, METH_NOARGS, length_hint_doc},
{"__reduce__", unicodeiter_reduce, METH_NOARGS, reduce_doc},
{"__setstate__", unicodeiter_setstate, METH_O, setstate_doc},
{NULL, NULL} /* sentinel */
};
PyTypeObject PyUnicodeIter_Type = {
PyVarObject_HEAD_INIT(&PyType_Type, 0)
"str_iterator", /* tp_name */
sizeof(unicodeiterobject), /* tp_basicsize */
0, /* tp_itemsize */
/* methods */
unicodeiter_dealloc,/* tp_dealloc */
0, /* tp_vectorcall_offset */
0, /* tp_getattr */
0, /* tp_setattr */
0, /* tp_as_async */
0, /* tp_repr */
0, /* tp_as_number */
0, /* tp_as_sequence */
0, /* tp_as_mapping */
0, /* tp_hash */
0, /* tp_call */
0, /* tp_str */
PyObject_GenericGetAttr, /* tp_getattro */
0, /* tp_setattro */
0, /* tp_as_buffer */
Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC,/* tp_flags */
0, /* tp_doc */
unicodeiter_traverse, /* tp_traverse */
0, /* tp_clear */
0, /* tp_richcompare */
0, /* tp_weaklistoffset */
PyObject_SelfIter, /* tp_iter */
unicodeiter_next, /* tp_iternext */
unicodeiter_methods, /* tp_methods */
0,
};
PyTypeObject _PyUnicodeASCIIIter_Type = {
PyVarObject_HEAD_INIT(&PyType_Type, 0)
.tp_name = "str_ascii_iterator",
.tp_basicsize = sizeof(unicodeiterobject),
.tp_dealloc = unicodeiter_dealloc,
.tp_getattro = PyObject_GenericGetAttr,
.tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC,
.tp_traverse = unicodeiter_traverse,
.tp_iter = PyObject_SelfIter,
.tp_iternext = unicode_ascii_iter_next,
.tp_methods = unicodeiter_methods,
};
static PyObject *
unicode_iter(PyObject *seq)
{
unicodeiterobject *it;
if (!PyUnicode_Check(seq)) {
PyErr_BadInternalCall();
return NULL;
}
if (PyUnicode_IS_COMPACT_ASCII(seq)) {
it = PyObject_GC_New(unicodeiterobject, &_PyUnicodeASCIIIter_Type);
}
else {
it = PyObject_GC_New(unicodeiterobject, &PyUnicodeIter_Type);
}
if (it == NULL)
return NULL;
it->it_index = 0;
it->it_seq = Py_NewRef(seq);
_PyObject_GC_TRACK(it);
return (PyObject *)it;
}
static int
encode_wstr_utf8(wchar_t *wstr, char **str, const char *name)
{
int res;
res = _Py_EncodeUTF8Ex(wstr, str, NULL, NULL, 1, _Py_ERROR_STRICT);
if (res == -2) {
PyErr_Format(PyExc_RuntimeError, "cannot encode %s", name);
return -1;
}
if (res < 0) {
PyErr_NoMemory();
return -1;
}
return 0;
}
static int
config_get_codec_name(wchar_t **config_encoding)
{
char *encoding;
if (encode_wstr_utf8(*config_encoding, &encoding, "stdio_encoding") < 0) {
return -1;
}
PyObject *name_obj = NULL;
PyObject *codec = _PyCodec_Lookup(encoding);
PyMem_RawFree(encoding);
if (!codec)
goto error;
name_obj = PyObject_GetAttrString(codec, "name");
Py_CLEAR(codec);
if (!name_obj) {
goto error;
}
wchar_t *wname = PyUnicode_AsWideCharString(name_obj, NULL);
Py_DECREF(name_obj);
if (wname == NULL) {
goto error;
}
wchar_t *raw_wname = _PyMem_RawWcsdup(wname);
if (raw_wname == NULL) {
PyMem_Free(wname);
PyErr_NoMemory();
goto error;
}
PyMem_RawFree(*config_encoding);
*config_encoding = raw_wname;
PyMem_Free(wname);
return 0;
error:
Py_XDECREF(codec);
Py_XDECREF(name_obj);
return -1;
}
static PyStatus
init_stdio_encoding(PyInterpreterState *interp)
{
/* Update the stdio encoding to the normalized Python codec name. */
PyConfig *config = (PyConfig*)_PyInterpreterState_GetConfig(interp);
if (config_get_codec_name(&config->stdio_encoding) < 0) {
return _PyStatus_ERR("failed to get the Python codec name "
"of the stdio encoding");
}
return _PyStatus_OK();
}
static int
init_fs_codec(PyInterpreterState *interp)
{
const PyConfig *config = _PyInterpreterState_GetConfig(interp);
_Py_error_handler error_handler;
error_handler = get_error_handler_wide(config->filesystem_errors);
if (error_handler == _Py_ERROR_UNKNOWN) {
PyErr_SetString(PyExc_RuntimeError, "unknown filesystem error handler");
return -1;
}
char *encoding, *errors;
if (encode_wstr_utf8(config->filesystem_encoding,
&encoding,
"filesystem_encoding") < 0) {
return -1;
}
if (encode_wstr_utf8(config->filesystem_errors,
&errors,
"filesystem_errors") < 0) {
PyMem_RawFree(encoding);
return -1;
}
struct _Py_unicode_fs_codec *fs_codec = &interp->unicode.fs_codec;
PyMem_RawFree(fs_codec->encoding);
fs_codec->encoding = encoding;
/* encoding has been normalized by init_fs_encoding() */
fs_codec->utf8 = (strcmp(encoding, "utf-8") == 0);
PyMem_RawFree(fs_codec->errors);
fs_codec->errors = errors;
fs_codec->error_handler = error_handler;
#ifdef _Py_FORCE_UTF8_FS_ENCODING
assert(fs_codec->utf8 == 1);
#endif
/* At this point, PyUnicode_EncodeFSDefault() and
PyUnicode_DecodeFSDefault() can now use the Python codec rather than
the C implementation of the filesystem encoding. */
/* Set Py_FileSystemDefaultEncoding and Py_FileSystemDefaultEncodeErrors
global configuration variables. */
if (_Py_IsMainInterpreter(interp)) {
if (_Py_SetFileSystemEncoding(fs_codec->encoding,
fs_codec->errors) < 0) {
PyErr_NoMemory();
return -1;
}
}
return 0;
}
static PyStatus
init_fs_encoding(PyThreadState *tstate)
{
PyInterpreterState *interp = tstate->interp;
/* Update the filesystem encoding to the normalized Python codec name.
For example, replace "ANSI_X3.4-1968" (locale encoding) with "ascii"
(Python codec name). */
PyConfig *config = (PyConfig*)_PyInterpreterState_GetConfig(interp);
if (config_get_codec_name(&config->filesystem_encoding) < 0) {
_Py_DumpPathConfig(tstate);
return _PyStatus_ERR("failed to get the Python codec "
"of the filesystem encoding");
}
if (init_fs_codec(interp) < 0) {
return _PyStatus_ERR("cannot initialize filesystem codec");
}
return _PyStatus_OK();
}
PyStatus
_PyUnicode_InitEncodings(PyThreadState *tstate)
{
PyStatus status = _PyCodec_InitRegistry(tstate->interp);
if (_PyStatus_EXCEPTION(status)) {
return status;
}
status = init_fs_encoding(tstate);
if (_PyStatus_EXCEPTION(status)) {
return status;
}
return init_stdio_encoding(tstate->interp);
}
static void
_PyUnicode_FiniEncodings(struct _Py_unicode_fs_codec *fs_codec)
{
PyMem_RawFree(fs_codec->encoding);
fs_codec->encoding = NULL;
fs_codec->utf8 = 0;
PyMem_RawFree(fs_codec->errors);
fs_codec->errors = NULL;
fs_codec->error_handler = _Py_ERROR_UNKNOWN;
}
#ifdef Py_DEBUG
static inline int
unicode_is_finalizing(void)
{
return (get_interned_dict(_PyInterpreterState_Main()) == NULL);
}
#endif
void
_PyUnicode_FiniTypes(PyInterpreterState *interp)
{
_PyStaticType_FiniBuiltin(interp, &EncodingMapType);
_PyStaticType_FiniBuiltin(interp, &PyFieldNameIter_Type);
_PyStaticType_FiniBuiltin(interp, &PyFormatterIter_Type);
}
void
_PyUnicode_Fini(PyInterpreterState *interp)
{
struct _Py_unicode_state *state = &interp->unicode;
if (!has_shared_intern_dict(interp)) {
// _PyUnicode_ClearInterned() must be called before _PyUnicode_Fini()
assert(get_interned_dict(interp) == NULL);
}
_PyUnicode_FiniEncodings(&state->fs_codec);
// bpo-47182: force a unicodedata CAPI capsule re-import on
// subsequent initialization of interpreter.
interp->unicode.ucnhash_capi = NULL;
unicode_clear_identifiers(state);
}
/* A _string module, to export formatter_parser and formatter_field_name_split
to the string.Formatter class implemented in Python. */
static PyMethodDef _string_methods[] = {
{"formatter_field_name_split", formatter_field_name_split,
METH_O, PyDoc_STR("split the argument as a field name")},
{"formatter_parser", formatter_parser,
METH_O, PyDoc_STR("parse the argument as a format string")},
{NULL, NULL}
};
static PyModuleDef_Slot module_slots[] = {
_Py_ABI_SLOT,
{Py_mod_multiple_interpreters, Py_MOD_PER_INTERPRETER_GIL_SUPPORTED},
{Py_mod_gil, Py_MOD_GIL_NOT_USED},
{0, NULL}
};
static struct PyModuleDef _string_module = {
PyModuleDef_HEAD_INIT,
.m_name = "_string",
.m_doc = PyDoc_STR("string helper module"),
.m_size = 0,
.m_methods = _string_methods,
.m_slots = module_slots,
};
PyMODINIT_FUNC
PyInit__string(void)
{
return PyModuleDef_Init(&_string_module);
}
#undef PyUnicode_KIND
int PyUnicode_KIND(PyObject *op)
{
if (!PyUnicode_Check(op)) {
PyErr_Format(PyExc_TypeError, "expect str, got %T", op);
return -1;
}
return _PyASCIIObject_CAST(op)->state.kind;
}
#undef PyUnicode_DATA
void* PyUnicode_DATA(PyObject *op)
{
if (!PyUnicode_Check(op)) {
PyErr_Format(PyExc_TypeError, "expect str, got %T", op);
return NULL;
}
return _PyUnicode_DATA(op);
}