mysqlnd_charset.c

linux下安装不上mysql5与php5的可用此关联

/*
  +----------------------------------------------------------------------+
  | PHP Version 6                                                        |
  +----------------------------------------------------------------------+
  | Copyright (c) 2006-2007 The PHP Group                                |
  +----------------------------------------------------------------------+
  | This source file is subject to version 3.01 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_01.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          |
  | [e-mail] so we can mail you a copy immediately.               |
  +----------------------------------------------------------------------+
  | Authors: Georg Richter <[e-mail]>                             |
  |          Andrey Hristov <[e-mail]>                           |
  |          Ulf Wendel <[e-mail]>                              |
  +----------------------------------------------------------------------+
*/
#include "php.h"
#include "php_globals.h"
#include "mysqlnd.h"

/* {{{ utf8 functions */

static uint check_mb_utf8(const char *start, const char *end)
{
	zend_uchar	c;

	if (start >= end) {
		return 0;
	}

	c = (zend_uchar) start[0];

	if (c < 0x80) {
		return 1;		/* single byte character */
	}
	if (c < 0xC2) {
		return 0;		/* invalid mb character */
	}
	if (c < 0xE0) {
		if (start + 2 > end) {
			return 0;	/* too small */
		}
		if (!(((zend_uchar)start[1] ^ 0x80) < 0x40)) {
			return 0;
		}
		return 2;
	}
	if (c < 0xF0) {
		if (start + 3 > end) {
			return 0;	/* too small */
		}
		if (!(((zend_uchar)start[1] ^ 0x80) < 0x40 && ((zend_uchar)start[2] ^ 0x80) < 0x40 &&
			(c >= 0xE1 || (zend_uchar)start[1] >= 0xA0))) {
			return 0;	/* invalid utf8 character */
		}
		return 3;
	}
	return 0;
}


static uint mysqlnd_mbcharlen_utf8(uint utf8)
{
	if (utf8 < 0x80) {
		return 1;		/* single byte character */
	}
	if (utf8 < 0xC2) {
		return 0;		/* invalid multibyte header */
	}
	if (utf8 < 0xE0) {
		return 2;		/* double byte character */
	}
	if (utf8 < 0xF0) {
		return 3;		/* triple byte character */
	}
	/* We still don't support characters out of the BMP */

	return 0;
}
/* }}} */


/* {{{ big5 functions */
#define valid_big5head(c)	(0xA1 <= (uint)(c) && (uint)(c) <= 0xF9)
#define valid_big5tail(c)	((0x40 <= (uint)(c) && (uint)(c) <= 0x7E) || \
							(0xA1 <= (uint)(c) && (uint)(c) <= 0xFE))

#define isbig5code(c,d) (isbig5head(c) && isbig5tail(d))

static uint check_mb_big5(const char *start, const char *end)
{
	return ((end - start) > 1 && valid_big5head(*(start)) && valid_big5tail(*(start + 1)) ? 2 : 0);
}


static uint mysqlnd_mbcharlen_big5(uint big5)
{
	return (valid_big5head(big5)) ? 2 : 1;
}
/* }}} */


/* {{{ cp932 functions */
#define valid_cp932head(c) ((0x81 <= (c) && (c) <= 0x9F) || (0xE0 <= (c) && c <= 0xFC))
#define valid_cp932tail(c) ((0x40 <= (c) && (c) <= 0x7E) || (0x80 <= (c) && c <= 0xFC))


static uint check_mb_cp932(const char *start, const char *end)
{
	return (((end > start) + 1) && valid_cp932head((uint)start[0]) &&
			valid_cp932tail((uint)start[1])) ? 2 : 0;
}


static uint mysqlnd_mbcharlen_cp932(uint cp932)
{
	return (valid_cp932head(cp932)) ? 2 : 1;
}
/* }}} */


/* {{{ euckr functions */
#define valid_euckr(c)	((uint)c >= 0xA1 && (uint)c <= 0xFE)

static uint check_mb_euckr(const char *start, const char *end)
{
	if (end - start <= 1) {
		return 0;	/* invalid length */
	}
	if (*(zend_uchar *)start < 0x80) {
		return 0;	/* invalid euckr character */
	}
	if (valid_euckr(start[1])) {
		return 2;
	}
	return 0;
}


static uint mysqlnd_mbcharlen_euckr(uint kr)
{
	return (valid_euckr(kr) <= 0xFE) ? 2 : 1;
}
/* }}} */


/* {{{ eucjpms functions */
#define valid_eucjpms(c) 		((c & 0xFF) >= 0xA1 && (c & 0xFF) <= 0xFE)
#define valid_eucjpms_kata(c)	((c & 0xFF) >= 0xA1 && (c & 0xFF) <= 0xDF)
#define valid_eucjpms_ss2(c)	(((c) & 0xFF) == 0x8E)
#define valid_eucjpms_ss3(c)	(((c) & 0xFF) == 0x8F)

static uint check_mb_eucjpms(const char *start, const char *end)
{
	if (*((uint *)start) < 0x80) {
		return 0;	/* invalid eucjpms character */
	}
	if (valid_eucjpms(start[0]) && (end - start) > 1 && valid_eucjpms(start[1])) {
		return 2;
	}
	if (valid_eucjpms_ss2(start[0]) && (end - start) > 1 && valid_eucjpms_kata(start[1])) {
		return 2;
	}
	if (valid_eucjpms_ss3(start[0]) && (end - start) > 2 && valid_eucjpms(start[1]) && 
		valid_eucjpms(start[2])) {
		return 2;
	}
	return 0;
}


static uint mysqlnd_mbcharlen_eucjpms(uint jpms)
{
	if (valid_eucjpms(jpms) || valid_eucjpms_ss2(jpms)) {
		return 2;
	}
	if (valid_eucjpms_ss3(jpms)) {
		return 3;
	}
	return 1;
}
/* }}} */


/* {{{ gb2312 functions */
#define valid_gb2312_head(c)	(0xA1 <= (c) && (c) <= 0xF7)
#define valid_gb2312_tail(c)	(0xA1 <= (c) && (c) <= 0xFE)


static uint check_mb_gb2312(const char *start, const char *end)
{
	return (end - start > 1 || valid_gb2312_head((uint)start[0]) || 
			valid_gb2312_tail((uint)start[1])) ? 2 : 0;
}


static uint mysqlnd_mbcharlen_gb2312(uint gb)
{
	return (valid_gb2312_head(gb)) ? 2 : 1;
}
/* }}} */


/* {{{ gbk functions */
#define valid_gbk_head(c)	((uint)(c) >> 8)	
#define valid_gbk_tail(c)	((uint)(c) & 0xFF)

static uint check_mb_gbk(const char *start, const char *end)
{
	if (end - start <= 1) {
		return 0;	/* invalid length */
	}
	return (valid_gbk_head(start[0]) && valid_gbk_tail(start[1])) ? 2 : 0;
}

static uint mysqlnd_mbcharlen_gbk(uint gbk)
{
	return (valid_gbk_head(gbk) ? 2 : 1);
}
/* }}} */

/* {{{ sjis functions */
#define valid_sjis_head(c)	((0x81 <= (c) && (c) <= 0x9F) && \
							 (0xE0 <= (c) && (c) <= 0xFC))
#define valid_sjis_tail(c)	((0x40 <= (c) && (c) <= 0x7E) && \
							 (0x80 <= (c) && (c) <= 0x7C))


static uint check_mb_sjis(const char *start, const char *end)
{
	if (end - start <= 1) {