/*

* Secret Labs' Regular Expression Engine

*

* regular expression matching engine

*

* partial history:

* 1999-10-24 fl created (based on existing template matcher code)

* 2000-03-06 fl first alpha, sort of

* 2000-08-01 fl fixes for 1.6b1

* 2000-08-07 fl use PyOS_CheckStack() if available

* 2000-09-20 fl added expand method

* 2001-03-20 fl lots of fixes for 2.1b2

* 2001-04-15 fl export copyright as Python attribute, not global

* 2001-04-28 fl added __copy__ methods (work in progress)

* 2001-05-14 fl fixes for 1.5.2 compatibility

* 2001-07-01 fl added BIGCHARSET support (from Martin von Loewis)

* 2001-10-18 fl fixed group reset issue (from Matthew Mueller)

* 2001-10-20 fl added split primitive; reenable unicode for 1.6/2.0/2.1

* 2001-10-21 fl added sub/subn primitive

* 2001-10-24 fl added finditer primitive (for 2.2 only)

* 2001-12-07 fl fixed memory leak in sub/subn (Guido van Rossum)

* 2002-11-09 fl fixed empty sub/subn return type

* 2003-04-18 mvl fully support 4-byte codes

* 2003-10-17 gn implemented non recursive scheme

* 2013-02-04 mrab added fullmatch primitive

*

* Copyright (c) 1997-2001 by Secret Labs AB. All rights reserved.

*

* This version of the SRE library can be redistributed under CNRI's

* Python 1.6 license. For any other use, please contact Secret Labs

* AB ([email protected]).

*

* Portions of this engine have been developed in cooperation with

* CNRI. Hewlett-Packard provided funding for 1.6 integration and

* other compatibility work.

*/

static char copyright[] =

" SRE 2.2.2 Copyright (c) 1997-2002 by Secret Labs AB ";

#define PY_SSIZE_T_CLEAN

#include "Python.h"

#include "structmember.h" /* offsetof */

#include "sre.h"

#define SRE_CODE_BITS (8 * sizeof(SRE_CODE))

#include <ctype.h>

/* name of this module, minus the leading underscore */

#if !defined(SRE_MODULE)

#define SRE_MODULE "sre"

#endif

#define SRE_PY_MODULE "re"

/* defining this one enables tracing */

#undef VERBOSE

/* -------------------------------------------------------------------- */

/* optional features */

/* enables fast searching */

#define USE_FAST_SEARCH

/* enables copy/deepcopy handling (work in progress) */

#undef USE_BUILTIN_COPY

/* -------------------------------------------------------------------- */

#if defined(_MSC_VER)

#pragma optimize("agtw", on) /* doesn't seem to make much difference... */

#pragma warning(disable: 4710) /* who cares if functions are not inlined ;-) */

/* fastest possible local call under MSVC */

#define LOCAL(type) static __inline type __fastcall

#elif defined(USE_INLINE)

#define LOCAL(type) static inline type

#else

#define LOCAL(type) static type

#endif

/* error codes */

#define SRE_ERROR_ILLEGAL -1 /* illegal opcode */

#define SRE_ERROR_STATE -2 /* illegal state */

#define SRE_ERROR_RECURSION_LIMIT -3 /* runaway recursion */

#define SRE_ERROR_MEMORY -9 /* out of memory */

#define SRE_ERROR_INTERRUPTED -10 /* signal handler raised exception */

#if defined(VERBOSE)

#define TRACE(v) printf v

#else

#define TRACE(v)

#endif

/* -------------------------------------------------------------------- */

/* search engine state */

/* default character predicates (run sre_chars.py to regenerate tables) */

#define SRE_DIGIT_MASK 1

#define SRE_SPACE_MASK 2

#define SRE_LINEBREAK_MASK 4

#define SRE_ALNUM_MASK 8

#define SRE_WORD_MASK 16

/* FIXME: this assumes ASCII. create tables in init_sre() instead */

static char sre_char_info[128] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 6, 2,

2, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0,

0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 25, 25, 25, 25, 25, 25, 25, 25,

25, 25, 0, 0, 0, 0, 0, 0, 0, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,

24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 0, 0,

0, 0, 16, 0, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,

24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 0, 0, 0, 0, 0 };

static char sre_char_lower[128] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,

10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,

27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43,

44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60,

61, 62, 63, 64, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107,

108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121,

122, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105,

106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119,

120, 121, 122, 123, 124, 125, 126, 127 };

#define SRE_IS_DIGIT(ch)\

((ch) < 128 ? (sre_char_info[(ch)] & SRE_DIGIT_MASK) : 0)

#define SRE_IS_SPACE(ch)\

((ch) < 128 ? (sre_char_info[(ch)] & SRE_SPACE_MASK) : 0)

#define SRE_IS_LINEBREAK(ch)\

((ch) < 128 ? (sre_char_info[(ch)] & SRE_LINEBREAK_MASK) : 0)

#define SRE_IS_ALNUM(ch)\

((ch) < 128 ? (sre_char_info[(ch)] & SRE_ALNUM_MASK) : 0)

#define SRE_IS_WORD(ch)\

((ch) < 128 ? (sre_char_info[(ch)] & SRE_WORD_MASK) : 0)

static unsigned int sre_lower(unsigned int ch)

{

return ((ch) < 128 ? (unsigned int)sre_char_lower[ch] : ch);

}

/* locale-specific character predicates */

/* !(c & ~N) == (c < N+1) for any unsigned c, this avoids

* warnings when c's type supports only numbers < N+1 */

#define SRE_LOC_IS_ALNUM(ch) (!((ch) & ~255) ? isalnum((ch)) : 0)

#define SRE_LOC_IS_WORD(ch) (SRE_LOC_IS_ALNUM((ch)) || (ch) == '_')

static unsigned int sre_lower_locale(unsigned int ch)

{

return ((ch) < 256 ? (unsigned int)tolower((ch)) : ch);

}

/* unicode-specific character predicates */

#define SRE_UNI_IS_DIGIT(ch) Py_UNICODE_ISDECIMAL(ch)

#define SRE_UNI_IS_SPACE(ch) Py_UNICODE_ISSPACE(ch)

#define SRE_UNI_IS_LINEBREAK(ch) Py_UNICODE_ISLINEBREAK(ch)

#define SRE_UNI_IS_ALNUM(ch) Py_UNICODE_ISALNUM(ch)

#define SRE_UNI_IS_WORD(ch) (SRE_UNI_IS_ALNUM(ch) || (ch) == '_')

static unsigned int sre_lower_unicode(unsigned int ch)

{

return (unsigned int) Py_UNICODE_TOLOWER(ch);

}

LOCAL(int)

sre_category(SRE_CODE category, unsigned int ch)

{

switch (category) {

case SRE_CATEGORY_DIGIT:

return SRE_IS_DIGIT(ch);

case SRE_CATEGORY_NOT_DIGIT:

return !SRE_IS_DIGIT(ch);

case SRE_CATEGORY_SPACE:

return SRE_IS_SPACE(ch);

case SRE_CATEGORY_NOT_SPACE:

return !SRE_IS_SPACE(ch);

case SRE_CATEGORY_WORD:

return SRE_IS_WORD(ch);

case SRE_CATEGORY_NOT_WORD:

return !SRE_IS_WORD(ch);

case SRE_CATEGORY_LINEBREAK:

return SRE_IS_LINEBREAK(ch);

case SRE_CATEGORY_NOT_LINEBREAK:

return !SRE_IS_LINEBREAK(ch);

case SRE_CATEGORY_LOC_WORD:

return SRE_LOC_IS_WORD(ch);

case SRE_CATEGORY_LOC_NOT_WORD:

return !SRE_LOC_IS_WORD(ch);

case SRE_CATEGORY_UNI_DIGIT:

return SRE_UNI_IS_DIGIT(ch);

case SRE_CATEGORY_UNI_NOT_DIGIT:

return !SRE_UNI_IS_DIGIT(ch);

case SRE_CATEGORY_UNI_SPACE:

return SRE_UNI_IS_SPACE(ch);

case SRE_CATEGORY_UNI_NOT_SPACE:

return !SRE_UNI_IS_SPACE(ch);

case SRE_CATEGORY_UNI_WORD:

return SRE_UNI_IS_WORD(ch);

case SRE_CATEGORY_UNI_NOT_WORD:

return !SRE_UNI_IS_WORD(ch);

case SRE_CATEGORY_UNI_LINEBREAK:

return SRE_UNI_IS_LINEBREAK(ch);

case SRE_CATEGORY_UNI_NOT_LINEBREAK:

return !SRE_UNI_IS_LINEBREAK(ch);

}

return 0;

}

/* helpers */

static void

data_stack_dealloc(SRE_STATE* state)

{

if (state->data_stack) {

PyMem_FREE(state->data_stack);

state->data_stack = NULL;

}

state->data_stack_size = state->data_stack_base = 0;

}

static int

data_stack_grow(SRE_STATE* state, Py_ssize_t size)

{

Py_ssize_t minsize, cursize;

minsize = state->data_stack_base+size;

cursize = state->data_stack_size;

if (cursize < minsize) {

void* stack;

cursize = minsize+minsize/4+1024;

TRACE(("allocate/grow stack %" PY_FORMAT_SIZE_T "d\n", cursize));

stack = PyMem_REALLOC(state->data_stack, cursize);

if (!stack) {

data_stack_dealloc(state);

return SRE_ERROR_MEMORY;

}

state->data_stack = (char *)stack;

state->data_stack_size = cursize;

}

return 0;

}

/* generate 8-bit version */

#define SRE_CHAR Py_UCS1

#define SIZEOF_SRE_CHAR 1

#define SRE(F) sre_ucs1_##F

#include "sre_lib.h"

/* generate 16-bit unicode version */

#define SRE_CHAR Py_UCS2

#define SIZEOF_SRE_CHAR 2

#define SRE(F) sre_ucs2_##F

#include "sre_lib.h"

/* generate 32-bit unicode version */

#define SRE_CHAR Py_UCS4

#define SIZEOF_SRE_CHAR 4

#define SRE(F) sre_ucs4_##F

#include "sre_lib.h"

/* -------------------------------------------------------------------- */

/* factories and destructors */

/* see sre.h for object declarations */

static PyObject*pattern_new_match(PatternObject*, SRE_STATE*, Py_ssize_t);

static PyObject*pattern_scanner(PatternObject*, PyObject*, PyObject* kw);

static PyObject *

sre_codesize(PyObject* self, PyObject *unused)

{

return PyLong_FromSize_t(sizeof(SRE_CODE));

}

static PyObject *

sre_getlower(PyObject* self, PyObject* args)

{

int character, flags;

if (!PyArg_ParseTuple(args, "ii", &character, &flags))

return NULL;

if (flags & SRE_FLAG_LOCALE)

return Py_BuildValue("i", sre_lower_locale(character));

if (flags & SRE_FLAG_UNICODE)

return Py_BuildValue("i", sre_lower_unicode(character));

return Py_BuildValue("i", sre_lower(character));

}

LOCAL(void)

state_reset(SRE_STATE* state)

{

/* FIXME: dynamic! */

/*memset(state->mark, 0, sizeof(*state->mark) * SRE_MARK_SIZE);*/

state->lastmark = -1;

state->lastindex = -1;

state->repeat = NULL;

data_stack_dealloc(state);

}

static void*

getstring(PyObject* string, Py_ssize_t* p_length,

int* p_isbytes, int* p_charsize,

Py_buffer *view)

{

/* given a python object, return a data pointer, a length (in

characters), and a character size. return NULL if the object

is not a string (or not compatible) */

/* Unicode objects do not support the buffer API. So, get the data

directly instead. */

if (PyUnicode_Check(string)) {

if (PyUnicode_READY(string) == -1)

return NULL;

*p_length = PyUnicode_GET_LENGTH(string);

*p_charsize = PyUnicode_KIND(string);

*p_isbytes = 0;

return PyUnicode_DATA(string);

}

/* get pointer to byte string buffer */

if (PyObject_GetBuffer(string, view, PyBUF_SIMPLE) != 0) {

PyErr_SetString(PyExc_TypeError, "expected string or buffer");

return NULL;

}

*p_length = view->len;

*p_charsize = 1;

*p_isbytes = 1;

if (view->buf == NULL) {

PyErr_SetString(PyExc_ValueError, "Buffer is NULL");

PyBuffer_Release(view);

view->buf = NULL;

return NULL;

}

return view->buf;

}

LOCAL(PyObject*)

state_init(SRE_STATE* state, PatternObject* pattern, PyObject* string,

Py_ssize_t start, Py_ssize_t end)

{

/* prepare state object */

Py_ssize_t length;

int isbytes, charsize;

void* ptr;

memset(state, 0, sizeof(SRE_STATE));

state->lastmark = -1;

state->lastindex = -1;

state->buffer.buf = NULL;

ptr = getstring(string, &length, &isbytes, &charsize, &state->buffer);

if (!ptr)

goto err;

if (isbytes && pattern->isbytes == 0) {

PyErr_SetString(PyExc_TypeError,

"can't use a string pattern on a bytes-like object");

goto err;

}

if (!isbytes && pattern->isbytes > 0) {

PyErr_SetString(PyExc_TypeError,

"can't use a bytes pattern on a string-like object");

goto err;

}

/* adjust boundaries */

if (start < 0)

start = 0;

else if (start > length)

start = length;

if (end < 0)

end = 0;

else if (end > length)

end = length;

state->isbytes = isbytes;

state->charsize = charsize;

state->beginning = ptr;

state->start = (void*) ((char*) ptr + start * state->charsize);

state->end = (void*) ((char*) ptr + end * state->charsize);

Py_INCREF(string);

state->string = string;

state->pos = start;

state->endpos = end;

if (pattern->flags & SRE_FLAG_LOCALE)

state->lower = sre_lower_locale;

else if (pattern->flags & SRE_FLAG_UNICODE)

state->lower = sre_lower_unicode;

else

state->lower = sre_lower;

return string;

err:

if (state->buffer.buf)

PyBuffer_Release(&state->buffer);

return NULL;

}

LOCAL(void)

state_fini(SRE_STATE* state)

{

if (state->buffer.buf)

PyBuffer_Release(&state->buffer);

Py_XDECREF(state->string);

data_stack_dealloc(state);

}

/* calculate offset from start of string */

#define STATE_OFFSET(state, member)\

(((char*)(member) - (char*)(state)->beginning) / (state)->charsize)

LOCAL(PyObject*)

getslice(int isbytes, const void *ptr,

PyObject* string, Py_ssize_t start, Py_ssize_t end)

{

if (isbytes) {

if (PyBytes_CheckExact(string) &&

start == 0 && end == PyBytes_GET_SIZE(string)) {

Py_INCREF(string);

return string;

}

return PyBytes_FromStringAndSize(

(const char *)ptr + start, end - start);

}

else {

return PyUnicode_Substring(string, start, end);

}

}

LOCAL(PyObject*)

state_getslice(SRE_STATE* state, Py_ssize_t index, PyObject* string, int empty)

{

Py_ssize_t i, j;

index = (index - 1) * 2;

if (string == Py_None || index >= state->lastmark || !state->mark[index] || !state->mark[index+1]) {

if (empty)

/* want empty string */

i = j = 0;

else {

Py_INCREF(Py_None);

return Py_None;

}

} else {

i = STATE_OFFSET(state, state->mark[index]);

j = STATE_OFFSET(state, state->mark[index+1]);

}

return getslice(state->isbytes, state->beginning, string, i, j);

}

static void

pattern_error(Py_ssize_t status)

{

switch (status) {

case SRE_ERROR_RECURSION_LIMIT:

PyErr_SetString(

PyExc_RuntimeError,

"maximum recursion limit exceeded"

);

break;

case SRE_ERROR_MEMORY:

PyErr_NoMemory();

break;

case SRE_ERROR_INTERRUPTED:

/* An exception has already been raised, so let it fly */

break;

default:

/* other error codes indicate compiler/engine bugs */

PyErr_SetString(

PyExc_RuntimeError,

"internal error in regular expression engine"

);

}

}

static void

pattern_dealloc(PatternObject* self)

{

if (self->weakreflist != NULL)

PyObject_ClearWeakRefs((PyObject *) self);

Py_XDECREF(self->pattern);

Py_XDECREF(self->groupindex);

Py_XDECREF(self->indexgroup);

PyObject_DEL(self);

}

LOCAL(Py_ssize_t)

sre_match(SRE_STATE* state, SRE_CODE* pattern, int match_all)

{

if (state->charsize == 1)

return sre_ucs1_match(state, pattern, match_all);

if (state->charsize == 2)

return sre_ucs2_match(state, pattern, match_all);

assert(state->charsize == 4);

return sre_ucs4_match(state, pattern, match_all);

}

LOCAL(Py_ssize_t)

sre_search(SRE_STATE* state, SRE_CODE* pattern)

{

if (state->charsize == 1)

return sre_ucs1_search(state, pattern);

if (state->charsize == 2)

return sre_ucs2_search(state, pattern);

assert(state->charsize == 4);

return sre_ucs4_search(state, pattern);

}

static PyObject *

fix_string_param(PyObject *string, PyObject *string2, const char *oldname)

{

if (string2 != NULL) {

if (string != NULL) {

PyErr_Format(PyExc_TypeError,

"Argument given by name ('%s') and position (1)",

oldname);

return NULL;

}

if (PyErr_WarnFormat(PyExc_DeprecationWarning, 1,

"The '%s' keyword parameter name is deprecated. "

"Use 'string' instead.", oldname) < 0)

return NULL;

return string2;

}

if (string == NULL) {

PyErr_SetString(PyExc_TypeError,

"Required argument 'string' (pos 1) not found");

return NULL;

}

return string;

}

static PyObject *

pattern_match(PatternObject *self, PyObject *args, PyObject *kwargs)

{

static char *_keywords[] = {"string", "pos", "endpos", "pattern", NULL};

PyObject *string = NULL;

Py_ssize_t pos = 0;

Py_ssize_t endpos = PY_SSIZE_T_MAX;

PyObject *pattern = NULL;

SRE_STATE state;

Py_ssize_t status;

if (!PyArg_ParseTupleAndKeywords(args, kwargs,

"|Onn$O:match", _keywords,

&string, &pos, &endpos, &pattern))

return NULL;

string = fix_string_param(string, pattern, "pattern");

if (!string)

return NULL;

string = state_init(&state, (PatternObject *)self, string, pos, endpos);

if (!string)

return NULL;

state.ptr = state.start;

TRACE(("|%p|%p|MATCH\n", PatternObject_GetCode(self), state.ptr));

status = sre_match(&state, PatternObject_GetCode(self), 0);

TRACE(("|%p|%p|END\n", PatternObject_GetCode(self), state.ptr));

if (PyErr_Occurred())

return NULL;

state_fini(&state);

return (PyObject *)pattern_new_match(self, &state, status);

}

static PyObject*

pattern_fullmatch(PatternObject* self, PyObject* args, PyObject* kw)

{

SRE_STATE state;

Py_ssize_t status;

PyObject *string = NULL, *string2 = NULL;

Py_ssize_t start = 0;

Py_ssize_t end = PY_SSIZE_T_MAX;

static char* kwlist[] = { "string", "pos", "endpos", "pattern", NULL };

if (!PyArg_ParseTupleAndKeywords(args, kw, "|Onn$O:fullmatch", kwlist,

&string, &start, &end, &string2))

return NULL;

string = fix_string_param(string, string2, "pattern");

if (!string)

return NULL;

string = state_init(&state, self, string, start, end);

if (!string)

return NULL;

state.ptr = state.start;

TRACE(("|%p|%p|FULLMATCH\n", PatternObject_GetCode(self), state.ptr));

status = sre_match(&state, PatternObject_GetCode(self), 1);

TRACE(("|%p|%p|END\n", PatternObject_GetCode(self), state.ptr));

if (PyErr_Occurred())

return NULL;

state_fini(&state);

return pattern_new_match(self, &state, status);

}

static PyObject*

pattern_search(PatternObject* self, PyObject* args, PyObject* kw)

{

SRE_STATE state;

Py_ssize_t status;

PyObject *string = NULL, *string2 = NULL;

Py_ssize_t start = 0;

Py_ssize_t end = PY_SSIZE_T_MAX;

static char* kwlist[] = { "string", "pos", "endpos", "pattern", NULL };

if (!PyArg_ParseTupleAndKeywords(args, kw, "|Onn$O:search", kwlist,

&string, &start, &end, &string2))

return NULL;

string = fix_string_param(string, string2, "pattern");

if (!string)

return NULL;

string = state_init(&state, self, string, start, end);

if (!string)

return NULL;

TRACE(("|%p|%p|SEARCH\n", PatternObject_GetCode(self), state.ptr));

status = sre_search(&state, PatternObject_GetCode(self));

TRACE(("|%p|%p|END\n", PatternObject_GetCode(self), state.ptr));

state_fini(&state);

if (PyErr_Occurred())

return NULL;

return pattern_new_match(self, &state, status);

}

static PyObject*

call(char* module, char* function, PyObject* args)

{

PyObject* name;

PyObject* mod;

PyObject* func;

PyObject* result;

if (!args)

return NULL;

name = PyUnicode_FromString(module);

if (!name)

return NULL;

mod = PyImport_Import(name);

Py_DECREF(name);

if (!mod)

return NULL;

func = PyObject_GetAttrString(mod, function);

Py_DECREF(mod);

if (!func)

return NULL;

result = PyObject_CallObject(func, args);

Py_DECREF(func);

Py_DECREF(args);

return result;

}

#ifdef USE_BUILTIN_COPY

static int

deepcopy(PyObject** object, PyObject* memo)

{

PyObject* copy;

copy = call(

"copy", "deepcopy",

PyTuple_Pack(2, *object, memo)

);

if (!copy)

return 0;

Py_DECREF(*object);

*object = copy;

return 1; /* success */

}

#endif

static PyObject*

pattern_findall(PatternObject* self, PyObject* args, PyObject* kw)

{

SRE_STATE state;

PyObject* list;

Py_ssize_t status;

Py_ssize_t i, b, e;

PyObject *string = NULL, *string2 = NULL;

Py_ssize_t start = 0;

Py_ssize_t end = PY_SSIZE_T_MAX;

static char* kwlist[] = { "string", "pos", "endpos", "source", NULL };

if (!PyArg_ParseTupleAndKeywords(args, kw, "|Onn$O:findall", kwlist,

&string, &start, &end, &string2))

return NULL;

string = fix_string_param(string, string2, "source");

if (!string)

return NULL;

string = state_init(&state, self, string, start, end);

if (!string)

return NULL;

list = PyList_New(0);

if (!list) {

state_fini(&state);

return NULL;

}

while (state.start <= state.end) {

PyObject* item;

state_reset(&state);

state.ptr = state.start;

status = sre_search(&state, PatternObject_GetCode(self));

if (PyErr_Occurred())

goto error;

if (status <= 0) {

if (status == 0)

break;

pattern_error(status);

goto error;

}

/* don't bother to build a match object */

switch (self->groups) {

case 0:

b = STATE_OFFSET(&state, state.start);

e = STATE_OFFSET(&state, state.ptr);

item = getslice(state.isbytes, state.beginning,

string, b, e);

if (!item)

goto error;

break;

case 1:

item = state_getslice(&state, 1, string, 1);

if (!item)

goto error;

break;

default:

item = PyTuple_New(self->groups);

if (!item)

goto error;

for (i = 0; i < self->groups; i++) {

PyObject* o = state_getslice(&state, i+1, string, 1);

if (!o) {

Py_DECREF(item);

goto error;

}

PyTuple_SET_ITEM(item, i, o);

}

break;

}

status = PyList_Append(list, item);

Py_DECREF(item);

if (status < 0)

goto error;

if (state.ptr == state.start)

state.start = (void*) ((char*) state.ptr + state.charsize);

else

state.start = state.ptr;

}

state_fini(&state);

return list;

error:

Py_DECREF(list);

state_fini(&state);

return NULL;

}

static PyObject*

pattern_finditer(PatternObject* pattern, PyObject* args, PyObject* kw)

{

PyObject* scanner;

PyObject* search;

PyObject* iterator;

scanner = pattern_scanner(pattern, args, kw);

if (!scanner)

return NULL;

search = PyObject_GetAttrString(scanner, "search");

Py_DECREF(scanner);

if (!search)

return NULL;

iterator = PyCallIter_New(search, Py_None);

Py_DECREF(search);

return iterator;

}

static PyObject*

pattern_split(PatternObject* self, PyObject* args, PyObject* kw)

{

SRE_STATE state;

PyObject* list;

PyObject* item;

Py_ssize_t status;

Py_ssize_t n;

Py_ssize_t i;

void* last;

PyObject *string = NULL, *string2 = NULL;

Py_ssize_t maxsplit = 0;

static char* kwlist[] = { "string", "maxsplit", "source", NULL };

if (!PyArg_ParseTupleAndKeywords(args, kw, "|On$O:split", kwlist,

&string, &maxsplit, &string2))

return NULL;

string = fix_string_param(string, string2, "source");

if (!string)

return NULL;

string = state_init(&state, self, string, 0, PY_SSIZE_T_MAX);

if (!string)

return NULL;

list = PyList_New(0);

if (!list) {

state_fini(&state);

return NULL;

}

n = 0;

last = state.start;

while (!maxsplit || n < maxsplit) {

state_reset(&state);

state.ptr = state.start;

status = sre_search(&state, PatternObject_GetCode(self));

if (PyErr_Occurred())

goto error;

if (status <= 0) {

if (status == 0)

break;

pattern_error(status);

goto error;

}

if (state.start == state.ptr) {

if (last == state.end)

break;

/* skip one character */

state.start = (void*) ((char*) state.ptr + state.charsize);

continue;

}

/* get segment before this match */

item = getslice(state.isbytes, state.beginning,

string, STATE_OFFSET(&state, last),

STATE_OFFSET(&state, state.start)

);

if (!item)

goto error;

status = PyList_Append(list, item);

Py_DECREF(item);

if (status < 0)

goto error;

/* add groups (if any) */

for (i = 0; i < self->groups; i++) {

item = state_getslice(&state, i+1, string, 0);

if (!item)

goto error;

status = PyList_Append(list, item);

Py_DECREF(item);

if (status < 0)

goto error;

}

n = n + 1;

last = state.start = state.ptr;

}

/* get segment following last match (even if empty) */

item = getslice(state.isbytes, state.beginning,

string, STATE_OFFSET(&state, last), state.endpos

);

if (!item)

goto error;

status = PyList_Append(list, item);

Py_DECREF(item);

if (status < 0)

goto error;

state_fini(&state);

return list;

error:

Py_DECREF(list);

state_fini(&state);

return NULL;

}

static PyObject*

pattern_subx(PatternObject* self, PyObject* ptemplate, PyObject* string,

Py_ssize_t count, Py_ssize_t subn)

{

SRE_STATE state;

PyObject* list;

PyObject* joiner;

PyObject* item;

PyObject* filter;

PyObject* args;

PyObject* match;

void* ptr;

Py_ssize_t status;

Py_ssize_t n;

Py_ssize_t i, b, e;

int isbytes, charsize;

int filter_is_callable;

Py_buffer view;

if (PyCallable_Check(ptemplate)) {

/* sub/subn takes either a function or a template */

filter = ptemplate;

Py_INCREF(filter);

filter_is_callable = 1;

} else {

/* if not callable, check if it's a literal string */

int literal;

view.buf = NULL;

ptr = getstring(ptemplate, &n, &isbytes, &charsize, &view);

b = charsize;

if (ptr) {

if (charsize == 1)

literal = memchr(ptr, '\\', n) == NULL;

else

literal = PyUnicode_FindChar(ptemplate, '\\', 0, n, 1) == -1;

} else {

PyErr_Clear();

literal = 0;

}

if (view.buf)

PyBuffer_Release(&view);

if (literal) {

filter = ptemplate;

Py_INCREF(filter);

filter_is_callable = 0;

} else {

/* not a literal; hand it over to the template compiler */

filter = call(

SRE_PY_MODULE, "_subx",

PyTuple_Pack(2, self, ptemplate)

);

if (!filter)

return NULL;

filter_is_callable = PyCallable_Check(filter);

}

}

string = state_init(&state, self, string, 0, PY_SSIZE_T_MAX);