php-src/ext/standard/var_unserializer.re

/*
  +----------------------------------------------------------------------+
  | PHP Version 5                                                        |
  +----------------------------------------------------------------------+
  | Copyright (c) 1997-2004 The PHP Group                                |
  +----------------------------------------------------------------------+
  | This source file is subject to version 3.0 of the PHP license,       |
  | that is bundled with this package in the file LICENSE, and is        |
  | available through the world-wide-web at the following url:           |
  | http://www.php.net/license/3_0.txt.                                  |
  | If you did not receive a copy of the PHP license and are unable to   |
  | obtain it through the world-wide-web, please send a note to          |
  | license@php.net so we can mail you a copy immediately.               |
  +----------------------------------------------------------------------+
  | Author: Sascha Schumann <sascha@schumann.cx>                         |
  +----------------------------------------------------------------------+
*/

/* $Id$ */

#include "php.h"
#include "ext/standard/php_var.h"
#include "php_incomplete_class.h"

/* {{{ reference-handling for unserializer: var_* */
#define VAR_ENTRIES_MAX 1024

typedef struct {
	zval *data[VAR_ENTRIES_MAX];
	int used_slots;
	void *next;
} var_entries;

static inline void var_push(php_unserialize_data_t *var_hashx, zval **rval)
{
	var_entries *var_hash = var_hashx->first, *prev = NULL;

	while (var_hash && var_hash->used_slots == VAR_ENTRIES_MAX) {
		prev = var_hash;
		var_hash = var_hash->next;
	}

	if (!var_hash) {
		var_hash = emalloc(sizeof(var_entries));
		var_hash->used_slots = 0;
		var_hash->next = 0;

		if (!var_hashx->first)
			var_hashx->first = var_hash;
		else
			prev->next = var_hash;
	}

	var_hash->data[var_hash->used_slots++] = *rval;
}

static inline void var_push_dtor(php_unserialize_data_t *var_hashx, zval **rval)
{
	var_entries *var_hash = var_hashx->first_dtor, *prev = NULL;

	while (var_hash && var_hash->used_slots == VAR_ENTRIES_MAX) {
		prev = var_hash;
		var_hash = var_hash->next;
	}

	if (!var_hash) {
		var_hash = emalloc(sizeof(var_entries));
		var_hash->used_slots = 0;
		var_hash->next = 0;

		if (!var_hashx->first_dtor)
			var_hashx->first_dtor = var_hash;
		else
			prev->next = var_hash;
	}

	(*rval)->refcount++;
	var_hash->data[var_hash->used_slots++] = *rval;
}

PHPAPI void var_replace(php_unserialize_data_t *var_hashx, zval *ozval, zval **nzval)
{
	int i;
	var_entries *var_hash = var_hashx->first;
	
	while (var_hash) {
		for (i = 0; i < var_hash->used_slots; i++) {
			if (var_hash->data[i] == ozval) {
				var_hash->data[i] = *nzval;
				/* do not break here */
			}
		}
		var_hash = var_hash->next;
	}
}

static int var_access(php_unserialize_data_t *var_hashx, int id, zval ***store)
{
	var_entries *var_hash = var_hashx->first;
	
	while (id >= VAR_ENTRIES_MAX && var_hash && var_hash->used_slots == VAR_ENTRIES_MAX) {
		var_hash = var_hash->next;
		id -= VAR_ENTRIES_MAX;
	}

	if (!var_hash) return !SUCCESS;

	if (id < 0 || id >= var_hash->used_slots) return !SUCCESS;

	*store = &var_hash->data[id];

	return SUCCESS;
}

PHPAPI void var_destroy(php_unserialize_data_t *var_hashx)
{
	void *next;
	int i;
	var_entries *var_hash = var_hashx->first;
	
	while (var_hash) {
		next = var_hash->next;
		efree(var_hash);
		var_hash = next;
	}

	var_hash = var_hashx->first_dtor;
	
	while (var_hash) {
		for (i = 0; i < var_hash->used_slots; i++) {
			zval_ptr_dtor(&var_hash->data[i]);
		}
		next = var_hash->next;
		efree(var_hash);
		var_hash = next;
	}
}

/* }}} */

#define YYFILL(n) do { } while (0)
#define YYCTYPE unsigned char
#define YYCURSOR cursor
#define YYLIMIT limit
#define YYMARKER marker


/*!re2c
uiv = [+]? [0-9]+;
iv = [+-]? [0-9]+;
nv = [+-]? ([0-9]* "." [0-9]+|[0-9]+ "." [0-9]*);
nvexp = (iv | nv) [eE] [+-]? iv;
any = [\000-\277];
object = [OC];
*/



static inline int parse_iv2(const unsigned char *p, const unsigned char **q)
{
	char cursor;
	int result = 0;
	int neg = 0;

	switch (*p) {
		case '-':
			neg++;
			/* fall-through */
		case '+':
			p++;
	}
	
	while (1) {
		cursor = (char)*p;
		if (cursor >= '0' && cursor <= '9') {
			result = result * 10 + cursor - '0';
		} else {
			break;
		}
		p++;
	}
	if (q) *q = p;
	if (neg) return -result;
	return result;
}

static inline int parse_iv(const unsigned char *p)
{
	return parse_iv2(p, NULL);
}

/* no need to check for length - re2c already did */
static inline size_t parse_uiv(const unsigned char *p)
{
	unsigned char cursor;
	size_t result = 0;

	if (*p == '+') {
		p++;
	}
	
	while (1) {
		cursor = *p;
		if (cursor >= '0' && cursor <= '9') {
			result = result * 10 + (size_t)(cursor - (unsigned char)'0');
		} else {
			break;
		}
		p++;
	}
	return result;
}

#define UNSERIALIZE_PARAMETER zval **rval, const unsigned char **p, const unsigned char *max, php_unserialize_data_t *var_hash TSRMLS_DC
#define UNSERIALIZE_PASSTHRU rval, p, max, var_hash TSRMLS_CC

static inline int process_nested_data(UNSERIALIZE_PARAMETER, HashTable *ht, int elements)
{
	while (elements-- > 0) {
		zval *key, *data, **old_data;

		ALLOC_INIT_ZVAL(key);

		if (!php_var_unserialize(&key, p, max, NULL TSRMLS_CC)) {
			zval_dtor(key);
			FREE_ZVAL(key);
			return 0;
		}

		if (Z_TYPE_P(key) != IS_LONG && Z_TYPE_P(key) != IS_STRING) {
			zval_dtor(key);
			FREE_ZVAL(key);
			return 0;
		}

		ALLOC_INIT_ZVAL(data);

		if (!php_var_unserialize(&data, p, max, var_hash TSRMLS_CC)) {
			zval_dtor(key);
			FREE_ZVAL(key);
			zval_dtor(data);
			FREE_ZVAL(data);
			return 0;
		}

		switch (Z_TYPE_P(key)) {
			case IS_LONG:
				if (zend_hash_index_find(ht, Z_LVAL_P(key), (void **)&old_data)==SUCCESS) {
					var_push_dtor(var_hash, old_data);
				}
				zend_hash_index_update(ht, Z_LVAL_P(key), &data, sizeof(data), NULL);
				break;
			case IS_STRING:
				if (zend_hash_find(ht, Z_STRVAL_P(key), Z_STRLEN_P(key) + 1, (void **)&old_data)==SUCCESS) {
					var_push_dtor(var_hash, old_data);
				}
				zend_hash_update(ht, Z_STRVAL_P(key), Z_STRLEN_P(key) + 1, &data, sizeof(data), NULL);
				break;
		}
		
		zval_dtor(key);
		FREE_ZVAL(key);

		if (elements && *(*p-1) != ';' &&  *(*p-1) != '}') {
			(*p)--;
			return 0;
		}
	}

	return 1;
}

static inline int finish_nested_data(UNSERIALIZE_PARAMETER)
{
	if (*((*p)++) == '}') 
		return 1;

#if SOMETHING_NEW_MIGHT_LEAD_TO_CRASH_ENABLE_IF_YOU_ARE_BRAVE
	zval_ptr_dtor(rval);
#endif
	return 0;
}

static inline int object_custom(UNSERIALIZE_PARAMETER, zend_class_entry *ce)
{
	int datalen;

	if(ce->unserialize == NULL) {
		zend_error(E_WARNING, "Class %s has no unserializer", ce->name);
		return 0;
	}

	datalen = parse_iv2((*p) + 2, p);

	(*p) += 2;

	if((*p) + datalen >= max) {
		zend_error(E_WARNING, "Unsifficient data for unserializing - %d required, %d present", datalen, max - (*p));
		return 0;
	}
	
	if(ce->unserialize(rval, *p, datalen, (zend_unserialize_data *)var_hash TSRMLS_CC) != SUCCESS) {
		return 0;
	}

	(*p) += datalen;

	return finish_nested_data(UNSERIALIZE_PASSTHRU);
}

static inline int object_common1(UNSERIALIZE_PARAMETER, zend_class_entry *ce)
{
	int elements;

	elements = parse_iv2((*p) + 2, p);

	(*p) += 2;
	
	object_init_ex(*rval, ce);
	return elements;
}

static inline int object_common2(UNSERIALIZE_PARAMETER, int elements)
{
	zval *retval_ptr = NULL;
	zval fname;

	if (!process_nested_data(UNSERIALIZE_PASSTHRU, Z_OBJPROP_PP(rval), elements)) {
		return 0;
	}

	if(Z_OBJCE_PP(rval) != PHP_IC_ENTRY) {
		INIT_PZVAL(&fname);
		ZVAL_STRINGL(&fname, "__wakeup", sizeof("__wakeup") - 1, 0);
		call_user_function_ex(CG(function_table), rval, &fname, &retval_ptr, 0, 0, 1, NULL TSRMLS_CC);
	}

	if (retval_ptr)
		zval_ptr_dtor(&retval_ptr);

	return finish_nested_data(UNSERIALIZE_PASSTHRU);

}

PHPAPI int php_var_unserialize(UNSERIALIZE_PARAMETER)
{
	const unsigned char *cursor, *limit, *marker, *start;
	zval **rval_ref;

	limit = cursor = *p;
	
	if (var_hash && cursor[0] != 'R') {
		var_push(var_hash, rval);
	}

	start = cursor;

	
	
/*!re2c

"R:" iv ";"		{
	int id;

 	*p = YYCURSOR;
	if (!var_hash) return 0;

	id = parse_iv(start + 2) - 1;
	if (id == -1 || var_access(var_hash, id, &rval_ref) != SUCCESS) {
		return 0;
	}

	if (*rval != NULL) {
		zval_ptr_dtor(rval);
	}
	*rval = *rval_ref;
	(*rval)->refcount++;
	(*rval)->is_ref = 1;
	
	return 1;
}

"r:" iv ";"		{
	int id;

 	*p = YYCURSOR;
	if (!var_hash) return 0;

	id = parse_iv(start + 2) - 1;
	if (id == -1 || var_access(var_hash, id, &rval_ref) != SUCCESS) {
		return 0;
	}

	if (*rval == *rval_ref) return 0;

	if (*rval != NULL) {
		zval_ptr_dtor(rval);
	}
	*rval = *rval_ref;
	(*rval)->refcount++;
	(*rval)->is_ref = 0;
	
	return 1;
}

"N;"	{
	*p = YYCURSOR;
	INIT_PZVAL(*rval);
	ZVAL_NULL(*rval);
	return 1;
}

"b:" [01] ";"	{
	*p = YYCURSOR;
	INIT_PZVAL(*rval);
	ZVAL_BOOL(*rval, parse_iv(start + 2));
	return 1;
}

"i:" iv ";"	{
	*p = YYCURSOR;
	INIT_PZVAL(*rval);
	ZVAL_LONG(*rval, parse_iv(start + 2));
	return 1;
}

"d:" ("NAN" | "-"? "INF") ";"	{
	*p = YYCURSOR;
	INIT_PZVAL(*rval);

	if (!strncmp(start + 2, "NAN", 3)) {
		ZVAL_DOUBLE(*rval, php_get_nan());
	} else if (!strncmp(start + 2, "INF", 3)) {
		ZVAL_DOUBLE(*rval, php_get_inf());
	} else if (!strncmp(start + 2, "-INF", 4)) {
		ZVAL_DOUBLE(*rval, -php_get_inf());
	}

	return 1;
}

"d:" (iv | nv | nvexp) ";"	{
	*p = YYCURSOR;
	INIT_PZVAL(*rval);
	ZVAL_DOUBLE(*rval, zend_strtod(start + 2, NULL));
	return 1;
}

"s:" uiv ":" ["] 	{
	size_t len, maxlen;
	char *str;

	len = parse_uiv(start + 2);
	maxlen = max - YYCURSOR;
	if (maxlen < len) {
		*p = start + 2;
		return 0;
	}

	str = (char*)YYCURSOR;

	YYCURSOR += len;

	if (*(YYCURSOR) != '"') {
		*p = YYCURSOR;
		return 0;
	}

	YYCURSOR += 2;
	*p = YYCURSOR;

	INIT_PZVAL(*rval);
	ZVAL_STRINGL(*rval, str, len, 1);
	return 1;
}

"a:" uiv ":" "{" {
	int elements = parse_iv(start + 2);

	*p = YYCURSOR;

	INIT_PZVAL(*rval);
	Z_TYPE_PP(rval) = IS_ARRAY;
	ALLOC_HASHTABLE(Z_ARRVAL_PP(rval));

	zend_hash_init(Z_ARRVAL_PP(rval), elements + 1, NULL, ZVAL_PTR_DTOR, 0);

	if (!process_nested_data(UNSERIALIZE_PASSTHRU, Z_ARRVAL_PP(rval), elements)) {
		return 0;
	}

	return finish_nested_data(UNSERIALIZE_PASSTHRU);
}

"o:" iv ":" ["] {

	INIT_PZVAL(*rval);
	
	return object_common2(UNSERIALIZE_PASSTHRU,
			object_common1(UNSERIALIZE_PASSTHRU, ZEND_STANDARD_CLASS_DEF_PTR));
}

object ":" uiv ":" ["]	{
	size_t len, len2, len3, maxlen;
	int elements;
	char *class_name;
	zend_class_entry *ce;
	zend_class_entry **pce;
	int incomplete_class = 0;

	int custom_object = 0;

	zval *user_func;
	zval *retval_ptr;
	zval **args[1];
	zval *arg_func_name;

	if(*start == 'C') {
		custom_object = 1;
	}
	
	INIT_PZVAL(*rval);
	len2 = len = parse_uiv(start + 2);
	maxlen = max - YYCURSOR;
	if (maxlen < len || len == 0) {
		*p = start + 2;
		return 0;
	}

	class_name = (char*)YYCURSOR;

	YYCURSOR += len;

	if (*(YYCURSOR) != '"') {
		*p = YYCURSOR;
		return 0;
	}
	if (*(YYCURSOR+1) != ':') {
		*p = YYCURSOR+1;
		return 0;
	}

	len3 = strspn(class_name, "0123456789_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ");
	if (len3 != len)
	{
		*p = YYCURSOR + len3 - len;
		return 0;
	}

	class_name = estrndup(class_name, len);

	do {
		/* Try to find class directly */
		if (zend_lookup_class(class_name, len2, &pce TSRMLS_CC) == SUCCESS) {
			ce = *pce;
			break;
		}
		
		/* Check for unserialize callback */
		if ((PG(unserialize_callback_func) == NULL) || (PG(unserialize_callback_func)[0] == '\0')) {
			incomplete_class = 1;
			ce = PHP_IC_ENTRY;
			break;
		}
		
		/* Call unserialize callback */
		MAKE_STD_ZVAL(user_func);
		ZVAL_STRING(user_func, PG(unserialize_callback_func), 1);
		args[0] = &arg_func_name;
		MAKE_STD_ZVAL(arg_func_name);
		ZVAL_STRING(arg_func_name, class_name, 1);
		if (call_user_function_ex(CG(function_table), NULL, user_func, &retval_ptr, 1, args, 0, NULL TSRMLS_CC) != SUCCESS) {
			php_error_docref(NULL TSRMLS_CC, E_WARNING, "defined (%s) but not found", user_func->value.str.val);
			incomplete_class = 1;
			ce = PHP_IC_ENTRY;
			zval_ptr_dtor(&user_func);
			zval_ptr_dtor(&arg_func_name);
			break;
		}
		if (retval_ptr) {
			zval_ptr_dtor(&retval_ptr);
		}
		
		/* The callback function may have defined the class */
		if (zend_lookup_class(class_name, len2, &pce TSRMLS_CC) == SUCCESS) {
			ce = *pce;
		} else {
			php_error_docref(NULL TSRMLS_CC, E_WARNING, "Function %s() hasn't defined the class it was called for", user_func->value.str.val);
			incomplete_class = 1;
			ce = PHP_IC_ENTRY;
		}

		zval_ptr_dtor(&user_func);
		zval_ptr_dtor(&arg_func_name);
		break;
	} while (1);

	*p = YYCURSOR;

	if(custom_object) {
		efree(class_name);
		return object_custom(UNSERIALIZE_PASSTHRU, ce);
	}
	
	elements = object_common1(UNSERIALIZE_PASSTHRU, ce);

	if (incomplete_class) {
		php_store_class_name(*rval, class_name, len2);
	}
	efree(class_name);

	return object_common2(UNSERIALIZE_PASSTHRU, elements);
}

"}" {
	/* this is the case where we have less data than planned */
	php_error_docref(NULL TSRMLS_CC, E_NOTICE, "Unexpected end of serialized data");
	return 0; /* not sure if it should be 0 or 1 here? */
}

any	{ return 0; }

*/

	return 0;
}