and extend coding system to EUC/SJIS/JIS/HEX at 1994.10.11
and add all jis codes sequence type at 1995.8.16
Notes: Hexadecimal code by <ohki@gssm.otuka.tsukuba.ac.jp>
+ Adding features about Machine dependent codes and User Defined Codes
+ by Hiroshi MIURA <miura@samba.gr.jp> 2000.3.19
*/
#define _KANJI_C_
* are loaded.
*/
-char *(*multibyte_strchr)(char *, int ) = (char *(*)(char *, int )) strchr;
-char *(*multibyte_strrchr)(char *, int ) = (char *(*)(char *, int )) strrchr;
-char *(*multibyte_strstr)(char *, char *) = (char *(*)(char *, char *)) strstr;
-char *(*multibyte_strtok)(char *, char *) = (char *(*)(char *, char *)) strtok;
+const char *(*multibyte_strchr)(const char *, int ) = (const char *(*)(const char *, int )) strchr;
+const char *(*multibyte_strrchr)(const char *, int ) = (const char *(*)(const char *, int )) strrchr;
+const char *(*multibyte_strstr)(const char *, const char *) = (const char *(*)(const char *, const char *)) strstr;
+char *(*multibyte_strtok)(char *, const char *) = (char *(*)(char *, const char *)) strtok;
/*
* Kanji is treated differently here due to historical accident of
* charcnv.c.
*/
-static int skip_non_multibyte_char(char);
+static size_t skip_non_multibyte_char(char);
static BOOL not_multibyte_char_1(char);
char *(*_dos_to_unix)(char *, BOOL) = dos2unix_format;
char *(*_unix_to_dos)(char *, BOOL) = unix2dos_format;
-int (*_skip_multibyte_char)(char) = skip_non_multibyte_char;
+size_t (*_skip_multibyte_char)(char) = skip_non_multibyte_char;
BOOL (*is_multibyte_char_1)(char) = not_multibyte_char_1;
#else /* KANJI */
*/
static char *sj_to_sj(char *from, BOOL overwrite);
-static int skip_kanji_multibyte_char(char);
+static size_t skip_kanji_multibyte_char(char);
static BOOL is_kanji_multibyte_char_1(char);
char *(*_dos_to_unix)(char *, BOOL) = sj_to_sj;
char *(*_unix_to_dos)(char *, BOOL) = sj_to_sj;
-int (*_skip_multibyte_char)(char) = skip_kanji_multibyte_char;
+size_t (*_skip_multibyte_char)(char) = skip_kanji_multibyte_char;
int (*is_multibyte_char_1)(char) = is_kanji_multibyte_char_1;
#endif /* KANJI */
+BOOL global_is_multibyte_codepage = False;
+
/* jis si/so sequence */
static char jis_kso = JIS_KSO;
static char jis_ksi = JIS_KSI;
/*******************************************************************
SHIFT JIS functions
********************************************************************/
+
/*******************************************************************
search token from S1 separated any char of S2
S1 contains SHIFT JIS chars.
********************************************************************/
-static char *sj_strtok(char *s1, char *s2)
+
+static char *sj_strtok(char *s1, const char *s2)
{
- static char *s = NULL;
- char *q;
- if (!s1) {
- if (!s) {
- return NULL;
- }
- s1 = s;
- }
- for (q = s1; *s1; ) {
- if (is_shift_jis (*s1)) {
- s1 += 2;
- } else if (is_kana (*s1)) {
- s1++;
- } else {
- char *p = strchr (s2, *s1);
- if (p) {
- if (s1 != q) {
- s = s1 + 1;
- *s1 = '\0';
- return q;
- }
- q = s1 + 1;
- }
- s1++;
- }
+ static char *s = NULL;
+ char *q;
+ if (!s1) {
+ if (!s) {
+ return NULL;
}
- s = NULL;
- if (*q) {
- return q;
+ s1 = s;
+ }
+ for (q = s1; *s1; ) {
+ if (is_shift_jis (*s1)) {
+ s1 += 2;
+ } else if (is_kana (*s1)) {
+ s1++;
+ } else {
+ char *p = strchr (s2, *s1);
+ if (p) {
+ if (s1 != q) {
+ s = s1 + 1;
+ *s1 = '\0';
+ return q;
+ }
+ q = s1 + 1;
+ }
+ s1++;
}
- return NULL;
+ }
+ s = NULL;
+ if (*q) {
+ return q;
+ }
+ return NULL;
}
/*******************************************************************
search string S2 from S1
S1 contains SHIFT JIS chars.
********************************************************************/
-static char *sj_strstr(char *s1, char *s2)
+
+static const char *sj_strstr(const char *s1, const char *s2)
{
- int len = strlen ((char *) s2);
- if (!*s2)
- return (char *) s1;
- for (;*s1;) {
- if (*s1 == *s2) {
- if (strncmp (s1, s2, len) == 0)
- return (char *) s1;
- }
- if (is_shift_jis (*s1)) {
- s1 += 2;
- } else {
- s1++;
- }
+ size_t len = strlen (s2);
+ if (!*s2)
+ return (const char *) s1;
+ for (;*s1;) {
+ if (*s1 == *s2) {
+ if (strncmp (s1, s2, len) == 0)
+ return (const char *) s1;
+ }
+ if (is_shift_jis (*s1)) {
+ s1 += 2;
+ } else {
+ s1++;
}
- return 0;
+ }
+ return NULL;
}
/*******************************************************************
Search char C from beginning of S.
S contains SHIFT JIS chars.
********************************************************************/
-static char *sj_strchr (char *s, int c)
+
+static const char *sj_strchr (const char *s, int c)
{
- for (; *s; ) {
- if (*s == c)
- return (char *) s;
- if (is_shift_jis (*s)) {
- s += 2;
- } else {
- s++;
- }
+ for (; *s; ) {
+ if (*s == c)
+ return (const char *) s;
+ if (is_shift_jis (*s)) {
+ s += 2;
+ } else {
+ s++;
}
- return 0;
+ }
+ return NULL;
}
/*******************************************************************
Search char C end of S.
S contains SHIFT JIS chars.
********************************************************************/
-static char *sj_strrchr(char *s, int c)
+
+static const char *sj_strrchr(const char *s, int c)
{
- char *q;
+ const char *q;
- for (q = 0; *s; ) {
- if (*s == c) {
- q = (char *) s;
- }
- if (is_shift_jis (*s)) {
- s += 2;
- } else {
- s++;
- }
+ for (q = 0; *s; ) {
+ if (*s == c) {
+ q = (const char *) s;
}
- return q;
+ if (is_shift_jis (*s)) {
+ s += 2;
+ } else {
+ s++;
+ }
+ }
+ return q;
}
/*******************************************************************
Kanji multibyte char skip function.
*******************************************************************/
-static int skip_kanji_multibyte_char(char c)
+static size_t skip_kanji_multibyte_char(char c)
{
if(is_shift_jis(c)) {
return 2;
S1 contains generic multibyte chars.
********************************************************************/
-static char *generic_multibyte_strtok(char *s1, char *s2)
+static char *generic_multibyte_strtok(char *s1, const char *s2)
{
- static char *s = NULL;
- char *q;
- if (!s1) {
- if (!s) {
- return NULL;
- }
- s1 = s;
+ static char *s = NULL;
+ char *q;
+ if (!s1) {
+ if (!s) {
+ return NULL;
}
- for (q = s1; *s1; ) {
- if ((*is_multibyte_char_1)(*s1)) {
- s1 += 2;
- } else {
- char *p = strchr (s2, *s1);
- if (p) {
- if (s1 != q) {
- s = s1 + 1;
- *s1 = '\0';
- return q;
- }
- q = s1 + 1;
- }
- s1++;
+ s1 = s;
+ }
+ for (q = s1; *s1; ) {
+ if ((*is_multibyte_char_1)(*s1)) {
+ s1 += 2;
+ } else {
+ char *p = strchr (s2, *s1);
+ if (p) {
+ if (s1 != q) {
+ s = s1 + 1;
+ *s1 = '\0';
+ return q;
}
+ q = s1 + 1;
+ }
+ s1++;
}
- s = NULL;
- if (*q) {
- return q;
- }
- return NULL;
+ }
+ s = NULL;
+ if (*q) {
+ return q;
+ }
+ return NULL;
}
/*******************************************************************
S1 contains generic multibyte chars.
********************************************************************/
-static char *generic_multibyte_strstr(char *s1, char *s2)
+static const char *generic_multibyte_strstr(const char *s1, const char *s2)
{
- int len = strlen ((char *) s2);
- if (!*s2)
- return (char *) s1;
- for (;*s1;) {
- if (*s1 == *s2) {
- if (strncmp (s1, s2, len) == 0)
- return (char *) s1;
- }
- if ((*is_multibyte_char_1)(*s1)) {
- s1 += 2;
- } else {
- s1++;
- }
+ size_t len = strlen (s2);
+ if (!*s2)
+ return (const char *) s1;
+ for (;*s1;) {
+ if (*s1 == *s2) {
+ if (strncmp (s1, s2, len) == 0)
+ return (const char *) s1;
+ }
+ if ((*is_multibyte_char_1)(*s1)) {
+ s1 += 2;
+ } else {
+ s1++;
}
- return 0;
+ }
+ return NULL;
}
/*******************************************************************
S contains generic multibyte chars.
********************************************************************/
-static char *generic_multibyte_strchr(char *s, int c)
+static const char *generic_multibyte_strchr(const char *s, int c)
{
- for (; *s; ) {
- if (*s == c)
- return (char *) s;
- if ((*is_multibyte_char_1)(*s)) {
- s += 2;
- } else {
- s++;
- }
+ for (; *s; ) {
+ if (*s == c)
+ return (const char *) s;
+ if ((*is_multibyte_char_1)(*s)) {
+ s += 2;
+ } else {
+ s++;
}
- return 0;
+ }
+ return NULL;
}
/*******************************************************************
S contains generic multibyte chars.
********************************************************************/
-static char *generic_multibyte_strrchr(char *s, int c)
+static const char *generic_multibyte_strrchr(const char *s, int c)
{
- char *q;
+ const char *q;
- for (q = 0; *s; ) {
- if (*s == c) {
- q = (char *) s;
- }
- if ((*is_multibyte_char_1)(*s)) {
- s += 2;
- } else {
- s++;
- }
+ for (q = 0; *s; ) {
+ if (*s == c) {
+ q = (const char *) s;
}
- return q;
+ if ((*is_multibyte_char_1)(*s)) {
+ s += 2;
+ } else {
+ s++;
+ }
+ }
+ return q;
}
/*******************************************************************
Generic multibyte char skip function.
*******************************************************************/
-static int skip_generic_multibyte_char(char c)
+static size_t skip_generic_multibyte_char(char c)
{
if( (*is_multibyte_char_1)(c)) {
return 2;
/*******************************************************************
Code conversion
********************************************************************/
+
/* convesion buffer */
-static char cvtbuf[1024];
+static char cvtbuf[2*sizeof(pstring)];
/*******************************************************************
EUC <-> SJIS
********************************************************************/
+
static int euc2sjis (int hi, int lo)
{
- if (hi & 1)
- return ((hi / 2 + (hi < 0xdf ? 0x31 : 0x71)) << 8) |
- (lo - (lo >= 0xe0 ? 0x60 : 0x61));
- else
- return ((hi / 2 + (hi < 0xdf ? 0x30 : 0x70)) << 8) | (lo - 2);
+ int w;
+ int maxidx = SJISREVTBLSIZ;
+ int minidx = 0;
+ int i = 2;
+
+ if (hi & 1) {
+ hi = hi / 2 + (hi < 0xdf ? 0x31 : 0x71);
+ w = (hi << 8) | (lo - (lo >= 0xe0 ? 0x60 : 0x61));
+ } else {
+ hi = hi / 2 + (hi < 0xdf ? 0x30 : 0x70);
+ w = (hi << 8) | (lo - 2);
+ }
+ if ( (0x87 < hi ) && (hi < 0xed ) ) {
+ return w;
+ }
+ while ( maxidx >= minidx ) {
+ if ( sjisrev[i].start > w ) {
+ maxidx = i-1;
+ } else if ( w > sjisrev[i].end ) {
+ minidx = i+1;
+ } else {
+ w -= sjisrev[i].start;
+ w += sjisrev[i].rstart;
+ break;
+ }
+ i = (int)( minidx + (maxidx - minidx) % 2 );
+ }
+ return w;
}
static int sjis2euc (int hi, int lo)
{
- if (lo >= 0x9f)
- return ((hi * 2 - (hi >= 0xe0 ? 0xe0 : 0x60)) << 8) | (lo + 2);
- else
- return ((hi * 2 - (hi >= 0xe0 ? 0xe1 : 0x61)) << 8) |
- (lo + (lo >= 0x7f ? 0x60 : 0x61));
+ int minidx = 0;
+ int maxidx = SJISCONVTBLSIZ -1; /* max index 1 less than number of entries */
+ int i = ( 0 + SJISCONVTBLSIZ ) % 2;
+ int w = (int)((hi << 8) | lo);
+
+ if ( (sjisconv[0].start < w) && (w < sjisconv[SJISCONVTBLSIZ-1].end) ) {
+ while (maxidx >= minidx) {
+ if ( sjisconv[i].start > w ) {
+ maxidx = i-1;
+ } else if (w > sjisconv[i].end) {
+ minidx = i+1;
+ } else {
+ w -= sjisconv[i].start;
+ w += sjisconv[i].rstart;
+ break;
+ }
+ i = (int)( minidx + (maxidx-minidx)%2 );
+ }
+ hi = (int) ((w >> 8) & 0xff);
+ lo = (int) (w & 0xff);
+ }
+ if (hi >= 0xf0) {
+ hi = GETAHI;
+ lo = GETALO;
+ }
+ if (lo >= 0x9f)
+ return ((hi * 2 - (hi >= 0xe0 ? 0xe0 : 0x60)) << 8) | (lo + 2);
+ else
+ return ((hi * 2 - (hi >= 0xe0 ? 0xe1 : 0x61)) << 8) |
+ (lo + (lo >= 0x7f ? 0x60 : 0x61));
}
/*******************************************************************
Convert FROM contain SHIFT JIS codes to EUC codes
return converted buffer
********************************************************************/
+
static char *sj_to_euc(char *from, BOOL overwrite)
{
- char *out;
- char *save;
-
- save = (char *) from;
- for (out = cvtbuf; *from;) {
- if (is_shift_jis (*from)) {
- int code = sjis2euc ((int) from[0] & 0xff, (int) from[1] & 0xff);
- *out++ = (code >> 8) & 0xff;
- *out++ = code;
- from += 2;
- } else if (is_kana (*from)) {
- *out++ = (char)euc_kana;
- *out++ = *from++;
- } else {
- *out++ = *from++;
- }
- }
- *out = 0;
- if (overwrite) {
- pstrcpy((char *) save, (char *) cvtbuf);
- return (char *) save;
+ char *out;
+ char *save;
+
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3);) {
+ if (is_shift_jis (*from)) {
+ int code = sjis2euc ((int) from[0] & 0xff, (int) from[1] & 0xff);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code & 0xff;
+ from += 2;
+ } else if (is_kana (*from)) {
+ *out++ = (char)euc_kana;
+ *out++ = *from++;
} else {
- return cvtbuf;
+ *out++ = *from++;
}
+ }
+ *out = 0;
+ if (overwrite) {
+ pstrcpy((char *) save, (char *) cvtbuf);
+ return (char *) save;
+ } else {
+ return cvtbuf;
+ }
}
/*******************************************************************
Convert FROM contain EUC codes to SHIFT JIS codes
return converted buffer
********************************************************************/
+
static char *euc_to_sj(char *from, BOOL overwrite)
{
- char *out;
- char *save;
+ char *out;
+ char *save;
+
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3); ) {
+ if (is_euc (*from)) {
+ int code = euc2sjis ((int) from[0] & 0xff, (int) from[1] & 0xff);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code & 0xff;
+ from += 2;
+ } else if (is_euc_kana (*from)) {
+ *out++ = from[1];
+ from += 2;
+ } else {
+ *out++ = *from++;
+ }
+ }
+ *out = 0;
- save = (char *) from;
- for (out = cvtbuf; *from; ) {
- if (is_euc (*from)) {
- int code = euc2sjis ((int) from[0] & 0xff, (int) from[1] & 0xff);
- *out++ = (code >> 8) & 0xff;
- *out++ = code;
- from += 2;
- } else if (is_euc_kana (*from)) {
- *out++ = from[1];
- from += 2;
+ if (overwrite) {
+ pstrcpy(save, (char *) cvtbuf);
+ return save;
+ } else {
+ return cvtbuf;
+ }
+}
+
+/*******************************************************************
+ EUC3 <-> SJIS
+********************************************************************/
+static int sjis3euc (int hi, int lo, int *len)
+{
+ int i,w;
+ int minidx;
+ int maxidx;
+
+ w = (int)((hi << 8) | lo);
+
+ /* no sjis */
+ if ( ( 0x40 >= lo ) && (lo >= 0xfc) && (lo == 0x7f )) {
+ w = (GETAHI << 8) | GETALO;
+
+ /* IBM Extended Kanji */
+ } else if (( w == 0xfa54 )||( w == 0x81ca )) {
+ *len = 2;
+ return (0xa2cc);
+
+ } else if (( w == 0xfa5b )||( w == 0x81e6)) {
+ *len = 2;
+ return (0xa2e8);
+
+ } else if (( 0xfa <= hi ) && ( hi <= 0xfc ) ) {
+ i = w - 0xfa40 - ( hi - 0xfa )*( 0xfb40 - 0xfafc) - ((lo < 0x7f)? 0 : 1 );
+ if ( i <= EUC3CONVTBLSIZ ){
+ *len = 3;
+ return euc3conv[i];
+ }
+
+/* NEC selected IBM Extend Kanji */
+ /* there are 3 code that is not good for conv */
+ } else if (( 0x8754 <= w ) && ( w <= 0x878a)) {
+ minidx = 0;
+ maxidx = EUC3CONV2TBLSIZ;
+ i = minidx + (maxidx - minidx) % 2;
+ while ( maxidx >= minidx ) {
+ if ( euc3conv2[i].sjis > w ) {
+ maxidx = i-1;
+ } else if ( w > euc3conv2[i].sjis ) {
+ minidx = i+1;
+ } else {
+ *len = 3;
+ return (euc3conv2[i].euc);
+ }
+ i = (int)( minidx + (maxidx - minidx) % 2 );
+ }
+ /* else normal EUC */
+
+ } else if (( w == 0xeef9 ) || ( w == 0x81ca )) {
+ *len = 2;
+ return (0xa2cc);
+
+ } else if (( 0xed <= hi ) && ( hi <= 0xef )) {
+ minidx = 0;
+ maxidx = SJISREVTBLSIZ;
+ i = 10;
+ while ( maxidx >= minidx ) {
+ if ( sjisrev[i].start > w ) {
+ maxidx = i-1;
+ } else if ( w > sjisrev[i].end ) {
+ minidx = i+1;
+ } else {
+ w -= sjisrev[i].start;
+ w += sjisrev[i].rstart;
+ break;
+ }
+ i = (int)( minidx + (maxidx - minidx) % 2 );
+ }
+ if ( w >= 0xfa40 ) {
+ i = w - 0xfa40 - ( hi - 0xfa )*( 0xfb40 - 0xfafc) - ((lo < 0x7f)? 0 : 1 );
+ if ( i <= EUC3CONVTBLSIZ ){
+ *len = 3;
+ return euc3conv[i];
+ } else {
+ w = (GETAHI << 8) | GETALO;
+ }
+ }
+ /* else normal EUC */
+
+/* UDC half low*/
+/* this area maps to the G2 UDC area: 0xf5a1 -- 0xfefe */
+ } else if ((0xf0 <= hi) && (hi <= 0xf4)) {
+ *len = 2;
+ if (lo >= 0x9f) {
+ return (((hi * 2 - 0xea) << 8) | (lo + 2));
+ } else {
+ return (((hi * 2 - 0xeb) << 8) | (lo + (lo >=0x7f ? 0x60: 0x61 )));
+ }
+
+/* UDC half high*/
+/* this area maps to the G3 UDC area: 0xf8f5a1 -- 0xf8fefe */
+ } else if ((0xf5 <= hi) && (hi <= 0xf9)) {
+ *len = 3;
+ if (lo >= 0x9f) {
+ return (((hi*2 - 0xf4) << 8) | (lo + 2));
+ } else {
+ return (((hi*2 - 0xf5) << 8) | (lo + (lo >= 0x7f ? 0x60: 0x61 )));
+ }
+ /* ....checked all special case */
+ }
+
+ /* These Normal 2 byte EUC */
+ *len = 2;
+ hi = (int) ((w >> 8) & 0xff);
+ lo = (int) (w & 0xff);
+
+ if (hi >= 0xf0) { /* Check range */
+ hi = GETAHI;
+ lo = GETALO;
+ }
+
+ if (lo >= 0x9f)
+ return ((hi * 2 - (hi >= 0xe0 ? 0xe0 : 0x60)) << 8) | (lo + 2);
+ else
+ return ((hi * 2 - (hi >= 0xe0 ? 0xe1 : 0x61)) << 8) |
+ (lo + (lo >= 0x7f ? 0x60 : 0x61));
+}
+
+static int euc3sjis (int hi, int lo, BOOL is_3byte)
+{
+ int w;
+
+ w = (int)((hi << 8) | lo);
+ if (is_3byte) {
+ if (( 0xf5 <= hi) && ( hi <= 0xfe)) {
+ /* UDC half high*/
+ /* this area maps to the G3 UDC area */
+ /* 0xf8f5a1 -- 0xf8fefe --> 0xf540 -- 0xf9fc */
+ if (hi & 1) {
+ return (((hi / 2 + 0x7b) << 8) | (lo - (lo >= 0xe0 ? 0x60 : 0x61)));
+ } else {
+ return (((hi / 2 + 0x7a) << 8) | (lo - 2));
+ }
+ } else {
+ /* Using map table */
+ int minidx = 0;
+ int maxidx = EUC3REVTBLSIZ;
+ int i = minidx + (maxidx - minidx) % 2;
+
+ while ( maxidx >= minidx ) {
+ if (euc3rev[i].euc > w) {
+ maxidx = i-1;
+ } else if (euc3rev[i].euc < w) {
+ minidx = i+1;
} else {
- *out++ = *from++;
+ return (euc3rev[i].sjis);
}
+ i = (int)( minidx + ( maxidx - minidx ) % 2);
+ }
+ return ((GETAHI << 8 ) | GETALO);
}
- *out = 0;
- if (overwrite) {
- pstrcpy(save, (char *) cvtbuf);
- return save;
+ } else { /* is_2byte */
+ if ((0xf5 <= hi) && (hi <= 0xfe)) {
+ /* UDC half low*/
+ /* this area maps to the G2 UDC area */
+ /* 0xf5a1 -- 0xfefe --> 0xf040 -- 0xf4fc */
+ if (hi & 1) {
+ return (((hi / 2 + 0x76) << 8) | (lo - (lo >= 0xe0 ? 0x60 : 0x61)));
+ } else {
+ return (((hi / 2 + 0x75) << 8) | (lo - 2));
+ }
+ } else { /* Normal EUC */
+ if (hi & 1) {
+ hi = hi / 2 + (hi < 0xdf ? 0x31 : 0x71);
+ return ((hi << 8) | (lo - (lo >= 0xe0 ? 0x60 : 0x61)));
+ } else {
+ hi = hi / 2 + (hi < 0xdf ? 0x30 : 0x70);
+ return ((hi << 8) | (lo - 2));
+ }
+ }
+ }
+}
+
+/*******************************************************************
+ Convert FROM contain SHIFT JIS codes to EUC codes (with SS2)
+ return converted buffer
+********************************************************************/
+
+static char *sj_to_euc3(char *from, BOOL overwrite)
+{
+ char *out;
+ char *save;
+ int len;
+
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-4);) {
+ if (is_shift_jis (*from)) {
+ int code = sjis3euc ((int) from[0] & 0xff, (int) from[1] & 0xff, &len);
+ if (len == 3) {
+ *out++ = (char)euc_sup;
+ }
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code & 0xff;
+ from += 2;
+ } else if (is_kana (*from)) {
+ *out++ = (char)euc_kana;
+ *out++ = *from++;
} else {
- return cvtbuf;
+ *out++ = *from++;
+ }
+ }
+ *out = 0;
+ if (overwrite) {
+ pstrcpy((char *) save, (char *) cvtbuf);
+ return (char *) save;
+ } else {
+ return cvtbuf;
+ }
+}
+
+/*******************************************************************
+ Convert FROM contain EUC codes (with Sup-Kanji) to SHIFT JIS codes
+ return converted buffer
+********************************************************************/
+static char *euc3_to_sj(char *from, BOOL overwrite)
+{
+ char *out;
+ char *save;
+
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3); ) {
+ if (is_euc_sup (*from)) {
+ int code = euc3sjis((int) from[1] & 0xff, (int) from[2] & 0xff, True);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code & 0xff;
+ from += 3;
+ } else if (is_euc (*from)) {
+ int code = euc3sjis ((int) from[0] & 0xff, (int) from[1] & 0xff,False);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code & 0xff;
+ from += 2;
+ } else if (is_euc_kana (*from)) {
+ *out++ = from[1];
+ from += 2;
+ } else {
+ *out++ = *from++;
}
+ }
+ *out = 0;
+
+ if (overwrite) {
+ pstrcpy(save, (char *) cvtbuf);
+ return save;
+ } else {
+ return cvtbuf;
+ }
}
/*******************************************************************
JIS7,JIS8,JUNET <-> SJIS
********************************************************************/
+
static int sjis2jis(int hi, int lo)
{
- if (lo >= 0x9f)
- return ((hi * 2 - (hi >= 0xe0 ? 0x160 : 0xe0)) << 8) | (lo - 0x7e);
- else
- return ((hi * 2 - (hi >= 0xe0 ? 0x161 : 0xe1)) << 8) |
- (lo - (lo >= 0x7f ? 0x20 : 0x1f));
+ int minidx = 0;
+ int maxidx = SJISCONVTBLSIZ -1; /* max index 1 less than number of entries */
+ int i = (0 + SJISCONVTBLSIZ) % 2;
+ int w = (int)((hi << 8) | lo);
+
+ if ((sjisconv[0].start < w) && (w < sjisconv[SJISCONVTBLSIZ-1].end)) {
+ while (maxidx >= minidx) {
+ if (sjisconv[i].start > w) {
+ maxidx = i-1;
+ } else if (w > sjisconv[i].end) {
+ minidx = i+1;
+ } else {
+ w -= sjisconv[i].start;
+ w += sjisconv[i].rstart;
+ break;
+ }
+ i = (int)( minidx + (maxidx-minidx) %2 );
+ }
+ hi = (int) ((w >> 8) & 0xff);
+ lo = (int) (w & 0xff);
+ }
+ if (hi >= 0xf0) {
+ hi = GETAHI;
+ lo = GETALO;
+ }
+ if (lo >= 0x9f)
+ return ((hi * 2 - (hi >= 0xe0 ? 0x160 : 0xe0)) << 8) | (lo - 0x7e);
+ else
+ return ((hi * 2 - (hi >= 0xe0 ? 0x161 : 0xe1)) << 8) |
+ (lo - (lo >= 0x7f ? 0x20 : 0x1f));
}
static int jis2sjis(int hi, int lo)
{
- if (hi & 1)
- return ((hi / 2 + (hi < 0x5f ? 0x71 : 0xb1)) << 8) |
- (lo + (lo >= 0x60 ? 0x20 : 0x1f));
- else
- return ((hi / 2 + (hi < 0x5f ? 0x70 : 0xb0)) << 8) | (lo + 0x7e);
+ int w;
+ int minidx = 0;
+ int maxidx = SJISREVTBLSIZ;
+ int i = 2;
+
+ if (hi & 1) {
+ hi = hi / 2 + (hi < 0x5f ? 0x71 : 0xb1);
+ w = (hi << 8) | (lo + (lo >= 0x60 ? 0x20 : 0x1f));
+ } else {
+ hi = hi / 2 + (hi < 0x5f ? 0x70 : 0xb0);
+ w = (hi << 8) | (lo + 0x7e);
+ }
+
+ if (( 0x87 < hi ) && ( hi < 0xed )) {
+ return w;
+ }
+ while (maxidx >= minidx) {
+ if (sjisrev[i].start > w) {
+ maxidx = i-1;
+ } else if (w > sjisrev[i].end) {
+ minidx = i+1;
+ } else {
+ w -= sjisrev[i].start;
+ w += sjisrev[i].rstart;
+ break;
+ }
+ i = (int)( minidx + (maxidx-minidx) %2 );
+ }
+ return w;
}
/*******************************************************************
Convert FROM contain JIS codes to SHIFT JIS codes
return converted buffer
********************************************************************/
+
static char *jis8_to_sj(char *from, BOOL overwrite)
{
- char *out;
- int shifted;
- char *save;
-
- shifted = _KJ_ROMAN;
- save = (char *) from;
- for (out = cvtbuf; *from;) {
- if (is_esc (*from)) {
- if (is_so1 (from[1]) && is_so2 (from[2])) {
- shifted = _KJ_KANJI;
- from += 3;
- } else if (is_si1 (from[1]) && is_si2 (from[2])) {
- shifted = _KJ_ROMAN;
- from += 3;
- } else { /* sequence error */
- goto normal;
- }
- } else {
- normal:
- switch (shifted) {
- default:
- case _KJ_ROMAN:
- *out++ = *from++;
- break;
- case _KJ_KANJI:
- {
- int code = jis2sjis ((int) from[0] & 0xff, (int) from[1] & 0xff);
- *out++ = (code >> 8) & 0xff;
- *out++ = code;
- from += 2;
- }
- break;
- }
- }
- }
- *out = 0;
- if (overwrite) {
- pstrcpy (save, (char *) cvtbuf);
- return save;
+ char *out;
+ int shifted;
+ char *save;
+
+ shifted = _KJ_ROMAN;
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3);) {
+ if (is_esc (*from)) {
+ if (is_so1 (from[1]) && is_so2 (from[2])) {
+ shifted = _KJ_KANJI;
+ from += 3;
+ } else if (is_si1 (from[1]) && is_si2 (from[2])) {
+ shifted = _KJ_ROMAN;
+ from += 3;
+ } else { /* sequence error */
+ goto normal;
+ }
} else {
- return cvtbuf;
+
+normal:
+
+ switch (shifted) {
+ default:
+ case _KJ_ROMAN:
+ *out++ = *from++;
+ break;
+ case _KJ_KANJI:
+ {
+ int code = jis2sjis ((int) from[0] & 0xff, (int) from[1] & 0xff);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code;
+ from += 2;
+ break;
+ }
+ }
}
+ }
+
+ *out = 0;
+ if (overwrite) {
+ pstrcpy (save, (char *) cvtbuf);
+ return save;
+ } else {
+ return cvtbuf;
+ }
}
/*******************************************************************
Convert FROM contain SHIFT JIS codes to JIS codes
return converted buffer
********************************************************************/
+
static char *sj_to_jis8(char *from, BOOL overwrite)
{
- char *out;
- int shifted;
- char *save;
-
- shifted = _KJ_ROMAN;
- save = (char *) from;
- for (out = cvtbuf; *from; ) {
- if (is_shift_jis (*from)) {
- int code;
- switch (shifted) {
- case _KJ_ROMAN: /* to KANJI */
- *out++ = jis_esc;
- *out++ = jis_so1;
- *out++ = jis_kso;
- shifted = _KJ_KANJI;
- break;
- }
- code = sjis2jis ((int) from[0] & 0xff, (int) from[1] & 0xff);
- *out++ = (code >> 8) & 0xff;
- *out++ = code;
- from += 2;
- } else {
- switch (shifted) {
- case _KJ_KANJI: /* to ROMAN/KANA */
- *out++ = jis_esc;
- *out++ = jis_si1;
- *out++ = jis_ksi;
- shifted = _KJ_ROMAN;
- break;
- }
- *out++ = *from++;
- }
- }
- switch (shifted) {
- case _KJ_KANJI: /* to ROMAN/KANA */
- *out++ = jis_esc;
- *out++ = jis_si1;
- *out++ = jis_ksi;
- shifted = _KJ_ROMAN;
- break;
- }
- *out = 0;
- if (overwrite) {
- pstrcpy (save, (char *) cvtbuf);
- return save;
+ char *out;
+ int shifted;
+ char *save;
+
+ shifted = _KJ_ROMAN;
+ save = (char *) from;
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-4); ) {
+ if (is_shift_jis (*from)) {
+ int code;
+ switch (shifted) {
+ case _KJ_ROMAN: /* to KANJI */
+ *out++ = jis_esc;
+ *out++ = jis_so1;
+ *out++ = jis_kso;
+ shifted = _KJ_KANJI;
+ break;
+ }
+ code = sjis2jis ((int) from[0] & 0xff, (int) from[1] & 0xff);
+ *out++ = (code >> 8) & 0xff;
+ *out++ = code;
+ from += 2;
} else {
- return cvtbuf;
+ switch (shifted) {
+ case _KJ_KANJI: /* to ROMAN/KANA */
+ *out++ = jis_esc;
+ *out++ = jis_si1;
+ *out++ = jis_ksi;
+ shifted = _KJ_ROMAN;
+ break;
+ }
+ *out++ = *from++;
}
+ }
+
+ switch (shifted) {
+ case _KJ_KANJI: /* to ROMAN/KANA */
+ *out++ = jis_esc;
+ *out++ = jis_si1;
+ *out++ = jis_ksi;
+ shifted = _KJ_ROMAN;
+ break;
+ }
+ *out = 0;
+ if (overwrite) {
+ pstrcpy (save, (char *) cvtbuf);
+ return save;
+ } else {
+ return cvtbuf;
+ }
}
/*******************************************************************
shifted = _KJ_ROMAN;
save = (char *) from;
- for (out = cvtbuf; *from;) {
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3);) {
if (is_esc (*from)) {
if (is_so1 (from[1]) && is_so2 (from[2])) {
shifted = _KJ_KANJI;
shifted = _KJ_ROMAN;
save = (char *) from;
- for (out = cvtbuf; *from; ) {
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-4); ) {
if (is_shift_jis (*from)) {
int code;
switch (shifted) {
Convert FROM contain 7 bits JIS(junet) codes to SHIFT JIS codes
return converted buffer
********************************************************************/
+
static char *junet_to_sj(char *from, BOOL overwrite)
{
char *out;
shifted = _KJ_ROMAN;
save = (char *) from;
- for (out = cvtbuf; *from;) {
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-3);) {
if (is_esc (*from)) {
if (is_so1 (from[1]) && is_so2 (from[2])) {
shifted = _KJ_KANJI;
shifted = _KJ_ROMAN;
save = (char *) from;
- for (out = cvtbuf; *from; ) {
+ for (out = cvtbuf; *from && (out - cvtbuf < sizeof(cvtbuf)-4); ) {
if (is_shift_jis (*from)) {
int code;
switch (shifted) {
sp = (char *) from;
dp = cvtbuf;
- while (*sp) {
- if (*sp == hex_tag && isxdigit (sp[1]) && isxdigit (sp[2])) {
+ while (*sp && (dp - cvtbuf < sizeof(cvtbuf)-3)) {
+ if (*sp == hex_tag && isxdigit((int)sp[1]) && isxdigit((int)sp[2])) {
*dp++ = (hex2bin (sp[1])<<4) | (hex2bin (sp[2]));
sp += 3;
} else
sp = (unsigned char*) from;
dp = (unsigned char*) cvtbuf;
- while (*sp) {
+ while (*sp && (((char *)dp)- cvtbuf < sizeof(cvtbuf)-7)) {
if (is_kana(*sp)) {
*dp++ = hex_tag;
*dp++ = bin2hex (((*sp)>>4)&0x0f);
sp = (char *) from;
dp = cvtbuf;
- while (*sp) {
+ while (*sp && (dp- cvtbuf < sizeof(cvtbuf)-2)) {
/*
* The only change between this and hex_to_sj is here. sj_to_cap only
* translates characters greater or equal to 0x80 - make sure that here
* we only do the reverse (that's why the strchr is used rather than
* isxdigit. Based on fix from ado@elsie.nci.nih.gov (Arthur David Olson).
*/
- if (*sp == hex_tag && (strchr ("89abcdefABCDEF", sp[1]) != NULL) && isxdigit (sp[2])) {
+ if (*sp == hex_tag && (strchr ("89abcdefABCDEF", sp[1]) != NULL) && isxdigit((int)sp[2])) {
*dp++ = (hex2bin (sp[1])<<4) | (hex2bin (sp[2]));
sp += 3;
} else
sp = (unsigned char*) from;
dp = (unsigned char*) cvtbuf;
- while (*sp) {
+ while (*sp && (((char *)dp) - cvtbuf < sizeof(cvtbuf)-4)) {
if (*sp >= 0x80) {
*dp++ = hex_tag;
*dp++ = bin2hex (((*sp)>>4)&0x0f);
}
}
+/*******************************************************************
+ cp to utf8
+********************************************************************/
+static char *cp_to_utf8(char *from, BOOL overwrite)
+{
+ unsigned char *dst;
+ unsigned char *src;
+ smb_ucs2_t val;
+ int w;
+ size_t len;
+
+ src = (unsigned char *)from;
+ dst = (unsigned char *)cvtbuf;
+ while (*src && (((char *)dst - cvtbuf) < sizeof(cvtbuf)-4)) {
+ len = _skip_multibyte_char(*src);
+ if ( len == 2 ) {
+ w = (int)(*src++ & 0xff);
+ w = (int)((w << 8)|(*src++ & 0xff));
+ } else {
+ w = (int)(*src++ & 0xff);
+ }
+ val = doscp2ucs2(w);
+
+ if ( val <= 0x7f ) {
+ *dst++ = (char)(val & 0xff);
+ } else if ( val <= 0x7ff ){
+ *dst++ = (char)( 0xc0 | ((val >> 6) & 0xff));
+ *dst++ = (char)( 0x80 | ( val & 0x3f ));
+ } else {
+ *dst++ = (char)( 0xe0 | ((val >> 12) & 0x0f));
+ *dst++ = (char)( 0x80 | ((val >> 6) & 0x3f));
+ *dst++ = (char)( 0x80 | (val & 0x3f));
+ }
+
+ }
+ *dst++='\0';
+ if (overwrite) {
+ pstrcpy ((char *) from, (char *) cvtbuf);
+ return (char *) from;
+ } else {
+ return cvtbuf;
+ }
+}
+
+/*******************************************************************
+ utf8 to cp
+********************************************************************/
+static char *utf8_to_cp(char *from, BOOL overwrite)
+{
+ unsigned char *src;
+ unsigned char *dst;
+ smb_ucs2_t val;
+ int w;
+
+ src = (unsigned char *)from;
+ dst = (unsigned char *)cvtbuf;
+
+ while (*src && ((char *)dst - cvtbuf < sizeof(cvtbuf)-4)) {
+ val = (*src++ & 0xff);
+ if (val < 0x80) {
+ *dst++ = (char)(val & 0x7f);
+ } else if ((0xc0 <= val) && (val <= 0xdf)
+ && (0x80 <= *src) && (*src <= 0xbf)) {
+ w = ucs2doscp( ((val & 31) << 6) | ((*src++) & 63 ));
+ *dst++ = (char)((w >> 8) & 0xff);
+ *dst++ = (char)(w & 0xff);
+ } else {
+ val = (val & 0x0f) << 12;
+ val |= ((*src++ & 0x3f) << 6);
+ val |= (*src++ & 0x3f);
+ w = ucs2doscp(val);
+ *dst++ = (char)((w >> 8) & 0xff);
+ *dst++ = (char)(w & 0xff);
+ }
+ }
+ *dst++='\0';
+ if (overwrite) {
+ pstrcpy ((char *) from, (char *) cvtbuf);
+ return (char *) from;
+ } else {
+ return cvtbuf;
+ }
+}
+
/************************************************************************
conversion:
_dos_to_unix _unix_to_dos
_dos_to_unix = sj_to_cap;
_unix_to_dos = cap_to_sj;
break;
+ case UTF8_CODE:
+ _dos_to_unix = cp_to_utf8;
+ _unix_to_dos = utf8_to_cp;
+ break;
+ case EUC3_CODE:
+ _dos_to_unix = sj_to_euc3;
+ _unix_to_dos = euc3_to_sj;
+ break;
}
}
codes = JUNET_CODE;
jis_kso = '@';
jis_ksi = 'H';
+ } else if (strequal (str, "utf8")) {
+ codes = UTF8_CODE;
+ } else if (strequal (str, "euc3")) {
+ codes = EUC3_CODE;
}
setup_string_function (codes);
}
Non multibyte char function.
*******************************************************************/
-static int skip_non_multibyte_char(char c)
+static size_t skip_non_multibyte_char(char c)
{
return 0;
}
switch( client_codepage )
{
case KANJI_CODEPAGE:
- multibyte_strchr = (char *(*)(char *, int )) sj_strchr;
- multibyte_strrchr = (char *(*)(char *, int )) sj_strrchr;
- multibyte_strstr = (char *(*)(char *, char *)) sj_strstr;
- multibyte_strtok = (char *(*)(char *, char *)) sj_strtok;
+ multibyte_strchr = sj_strchr;
+ multibyte_strrchr = sj_strrchr;
+ multibyte_strstr = sj_strstr;
+ multibyte_strtok = sj_strtok;
_skip_multibyte_char = skip_kanji_multibyte_char;
is_multibyte_char_1 = is_kanji_multibyte_char_1;
+ global_is_multibyte_codepage = True;
break;
case HANGUL_CODEPAGE:
- multibyte_strchr = (char *(*)(char *, int )) generic_multibyte_strchr;
- multibyte_strrchr = (char *(*)(char *, int )) generic_multibyte_strrchr;
- multibyte_strstr = (char *(*)(char *, char *)) generic_multibyte_strstr;
- multibyte_strtok = (char *(*)(char *, char *)) generic_multibyte_strtok;
+ multibyte_strchr = generic_multibyte_strchr;
+ multibyte_strrchr = generic_multibyte_strrchr;
+ multibyte_strstr = generic_multibyte_strstr;
+ multibyte_strtok = generic_multibyte_strtok;
_skip_multibyte_char = skip_generic_multibyte_char;
is_multibyte_char_1 = hangul_is_multibyte_char_1;
+ global_is_multibyte_codepage = True;
+ break;
case BIG5_CODEPAGE:
- multibyte_strchr = (char *(*)(char *, int )) generic_multibyte_strchr;
- multibyte_strrchr = (char *(*)(char *, int )) generic_multibyte_strrchr;
- multibyte_strstr = (char *(*)(char *, char *)) generic_multibyte_strstr;
- multibyte_strtok = (char *(*)(char *, char *)) generic_multibyte_strtok;
+ multibyte_strchr = generic_multibyte_strchr;
+ multibyte_strrchr = generic_multibyte_strrchr;
+ multibyte_strstr = generic_multibyte_strstr;
+ multibyte_strtok = generic_multibyte_strtok;
_skip_multibyte_char = skip_generic_multibyte_char;
is_multibyte_char_1 = big5_is_multibyte_char_1;
+ global_is_multibyte_codepage = True;
+ break;
case SIMPLIFIED_CHINESE_CODEPAGE:
- multibyte_strchr = (char *(*)(char *, int )) generic_multibyte_strchr;
- multibyte_strrchr = (char *(*)(char *, int )) generic_multibyte_strrchr;
- multibyte_strstr = (char *(*)(char *, char *)) generic_multibyte_strstr;
- multibyte_strtok = (char *(*)(char *, char *)) generic_multibyte_strtok;
+ multibyte_strchr = generic_multibyte_strchr;
+ multibyte_strrchr = generic_multibyte_strrchr;
+ multibyte_strstr = generic_multibyte_strstr;
+ multibyte_strtok = generic_multibyte_strtok;
_skip_multibyte_char = skip_generic_multibyte_char;
is_multibyte_char_1 = simpch_is_multibyte_char_1;
+ global_is_multibyte_codepage = True;
break;
/*
* Single char size code page.
*/
default:
- multibyte_strchr = (char *(*)(char *, int )) strchr;
- multibyte_strrchr = (char *(*)(char *, int )) strrchr;
- multibyte_strstr = (char *(*)(char *, char *)) strstr;
- multibyte_strtok = (char *(*)(char *, char *)) strtok;
+ multibyte_strchr = (const char *(*)(const char *, int )) strchr;
+ multibyte_strrchr = (const char *(*)(const char *, int )) strrchr;
+ multibyte_strstr = (const char *(*)(const char *, const char *)) strstr;
+ multibyte_strtok = (char *(*)(char *, const char *)) strtok;
_skip_multibyte_char = skip_non_multibyte_char;
is_multibyte_char_1 = not_multibyte_char_1;
+ global_is_multibyte_codepage = False;
break;
}
}