int file_knjchk(FILE *fp)
{
int c;
int f_sjis,f_euc;
int n_sjis,n_sjis_i,n_euc,n_euc_i;
n_sjis=0;
n_sjis_i=0;
n_euc=0;
n_euc_i=0;
f_sjis=FALSE;
f_euc=FALSE;
for(;;)
{
c=fgetc(fp);
if (c==EOF|| n_euc>32||n_sjis>32||n_euc_i>8||n_sjis_i>8)
return n_euc- n_euc_i*4>= n_sjis- n_sjis_i*4? KC_euc:KC_sjis;
if (c==ESC)
{
c=fgetc(fp);
if (c=='K')
return KC_jis;
if (c=='$')
{
c=fgetc(fp);
if (c=='B'||c=='@')
return KC_jis;
}
}
if (f_euc)
{
if (iseuc(c))
++n_euc; else
++n_euc_i;
f_euc=FALSE;
} else
{
if (iseuc(c))
f_euc=TRUE;
}
if (f_sjis)
{
if (issjis2(c))
++n_sjis; else
++n_sjis_i;
f_sjis=FALSE;
} else
{
if (issjis1(c))
f_sjis=TRUE;
}
}
}
gint guess_kanji(gint imax, guchar *buf)
{
int i, bad_euc, bad_sjis;
for (i = 0; i < imax; i++) {
if(buf[i+5] == '\0')
break;
if((strncmp(&buf[i], JIS0208_1978, strlen(JIS0208_1978)) == 0) ||
(strncmp(&buf[i], JIS0208_1983, strlen(JIS0208_1983)) == 0) ||
(strncmp(&buf[i], JIS0208_1990, strlen(JIS0208_1990)) == 0) ||
(strncmp(&buf[i], JIS0212, strlen(JIS0212)) == 0) ||
(strncmp(&buf[i], JIS_ASC, strlen(JIS_ASC)) == 0) ||
(strncmp(&buf[i], JIS_ASC2, strlen(JIS_ASC2)) == 0) ||
(strncmp(&buf[i], JIS_KANA, strlen(JIS_KANA)) == 0))
return(KCODE_JIS);
}
bad_euc = 0;
for (i = 0; i < imax; i++) {
if(buf[i+2] == '\0')
break;
if (iseuc(buf[i]) && ++i < imax) {
if (! iseuc(buf[i])) { bad_euc += 10; i--; }
else if (buf[i-1] >= 0xd0) bad_euc++; /* Dai 2 Suijun */
/* 1999-02-01 bug fixed. Thanks: massangeana */
} else if (buf[i] == 0x8e && ++i < imax) {
if (ishankana(buf[i])) bad_euc++;
else { bad_euc += 10; i--; }
} else if (buf[i] >= 0x80) bad_euc += 10;
}
bad_sjis = 0;
for (i = 0; i < imax; i++) {
if(buf[i+2] == '\0')
break;
if (issjis1(buf[i]) && ++i < imax) {
if (! issjis2(buf[i])) { bad_sjis += 10; i--; }
else if ((unsigned) (buf[i-1] * 256U + buf[i]) >= 0x989f)
bad_sjis++; /* Dai 2 Suijun */
} else if (buf[i] >= 0x80) {
if (ishankana(buf[i])) bad_sjis++;
else bad_sjis += 10;
}
}
if(bad_sjis < bad_euc)
return(KCODE_SJIS);
else if (bad_sjis > bad_euc)
return(KCODE_EUC);
else if ((bad_euc == 0) && (bad_sjis == 0))
return(KCODE_ASCII);
else
return(KCODE_UNKNOWN);
}
int kanji_countbuf(char c)
{
if (c==0)
return 0;
if (iseuc(c))
return 2;
if ((u_char)c==0x8e)
return 2;
if ((u_char)c==0x8f)
return 3;
return 1;
}
int kanji_countdsp(char c, int n)
{
if (c==0)
return 0;
if (c=='\t' && n!=-1)
return (n/sysinfo.tabstop+1)*sysinfo.tabstop - n;
if ((u_char)c==0x8e) // 半角かな
return 2;
if (iseuc(c)|| (u_char)c==0x8f|| iscntrl(c))
return 2;
return 1;
}
static gboolean validate_euc_str(guchar *str){
guchar *p;
p = str;
while(*p){
if (iseuc(p)) {
p +=2;
} else if(isprint(*p)) {
p ++;
} else if(isspace(*p)) {
*p = 0x20;
p++;
} else {
return(FALSE);
}
}
return(TRUE);
}
void hiragana_to_katakana(gchar *word)
{
gint i=0;
g_assert(word != NULL);
while(word[i] != '\0'){
if(isalpha(word[i])) {
i++;
continue;
}
if(iseuc(&word[i])) {
if(iseuchiragana(&word[i])) {
word[i] = 0xa5;
}
i += 2;
continue;
}
i++;
}
}
Encoding GetEncoding(PCSTR str, size_t len)
{
size_t i;
int ascii, eucjp, sjis, utf8, bad_eucjp, bad_sjis, bad_utf8;
int jis, hankana;
const unsigned char* buf = (const unsigned char*)str;
ascii = 1;
bad_eucjp = eucjp = 0;
bad_sjis = sjis = 0;
bad_utf8 = utf8 = 0;
jis = 0;
// check BOM
if (len >= 2)
{
if (buf[0] == 0xff && buf[1] == 0xfe)
return ENCODING_UTF16_LE;
else if (buf[0] == 0xfe && buf[1] == 0xff)
return ENCODING_UTF16_BE;
}
if (len >= 3 && !memcmp(buf, "\xef\xbb\xbf", 3))
return ENCODING_UTF8_BOM;
// check ENCODING_SJIS
hankana = 0;
for (i = 0; i < len; i++)
{
if (buf[i] >= 0x80)
ascii = 0;
if (buf[i] == 0x1b)
jis = 1;
if (buf[i] == 0x8e &&
i + 2 < len &&
buf[i + 2] == 0x8e &&
ishankana(buf[i + 1]))
{
bad_sjis += 1;
}
if (ishankana(buf[i]))
{
sjis += 0x10/2 - 1;
hankana++;
}
else
{
if (hankana == 1)
bad_sjis++;
hankana = 0;
if (issjis1(buf[i]))
{
if (i + 1 < len)
{
if (issjis2(buf[i + 1]))
{
sjis += 0x10;
i++;
}
else
bad_sjis += 0x100;
}
}
else if (buf[i] >= 0x80)
bad_sjis += 0x100;
}
}
if (ascii)
return jis ? ENCODING_JIS : ENCODING_ASCII;
// check ENCODING_EUCJP - JP
hankana = 0;
for (i = 0; i < len; i++)
{
if (buf[i] == 0x8e)
{
if (i + 1 < len)
{
if (ishankana(buf[i + 1]))
{
eucjp += 10;
i++;
hankana++;
}
else
bad_eucjp += 0x100;
}
}
else
{
if (hankana == 1)
bad_eucjp++;
hankana = 0;
if (iseuc(buf[i]))
{
if (i + 1 < len)
{
if (iseuc(buf[i + 1]))
{
i++;
eucjp += 0x10;
}
else
bad_eucjp += 0x100;
}
}
else if (buf[i] == 0x8f)
{
if (i + 2 < len)
{
if (iseuc(buf[i + 1]) && iseuc(buf[i + 2]))
{
i += 2;
eucjp += 0x10;
}
else
bad_eucjp += 100;
}
}
else if (buf[i] >= 0x80)
bad_eucjp += 0x100;
}
}
// check UTF-8
for (i = 0; i < len; i++)
{
if (isutf8_2byte(buf[i]))
{
if (i + 1 < len)
{
if (isutf8_trail(buf[i + 1]))
{
utf8 += 10;
i++;
}
else
bad_utf8 += 100;
}
}
else if (isutf8_3byte(buf[i]))
{
if (i + 2 < len)
{
if (isutf8_trail(buf[i + 1]) && isutf8_trail(buf[i + 2]))
{
utf8 += 15;
i += 2;
}
else
bad_utf8 += 1000;
}
}
else if (buf[i] >= 0x80)
bad_utf8 += 1000;
}
if (sjis - bad_sjis > eucjp - bad_eucjp)
{
if (sjis - bad_sjis > utf8 - bad_utf8)
return ENCODING_SJIS;
else if (sjis - bad_sjis < utf8 - bad_utf8)
return ENCODING_UTF8;
}
if (sjis - bad_sjis < eucjp - bad_eucjp)
{
if (eucjp - bad_eucjp > utf8 - bad_utf8)
return ENCODING_EUCJP;
else if (eucjp - bad_eucjp < utf8 - bad_utf8)
return ENCODING_UTF8;
}
return ENCODING_UNKNOWN;
}
const char *kanji_fromeuc(char *s,size_t bytes,const char *t,int kc)
{
u_char c;
const char *p;
enum {KM_ank, KM_kanji, KM_kana} km;
p=s;
switch(kc)
{
case KC_euc:
return t;
case KC_sjis:
for(;*t!='\0'&& bytes>0;)
{
c=*t++;
if (c==0x8e&& iskana(*t))
c=*t++; else
{
if (iseuc(c)&& iseuc(*t))
{
*s++= EUCtoSJIStable1[c - 0xa1];
--bytes;
c= *(u_char *)t - ((c&1)==0 ? 2 : *(u_char *)t>=0xe0? 0x60: 0x61);
++t;
}
}
*s++=c;
--bytes;
}
*s='\0';
break;
case KC_jis:
km=KM_ank;
for(;*t!='\0'&& bytes>0;)
{
c=*t++;
if (iseuc(c)&& km!=KM_kanji)
{
if (bytes<3)
break;
*s++='\x1b';
*s++='$';
*s++='B';
bytes-=3;
km=KM_kanji;
}
if (c==0x8e && iskana(*t))
{
if (km!=KM_kana)
{
if (bytes<3)
break;
*s++='\x1b';
*s++='(';
*s++='I';
bytes-=3;
km=KM_kana;
}
c=*t++;
} else
{
if (!iseuc(c) && km!=KM_ank)
{
if (bytes<3)
break;
*s++='\x1b';
*s++='(';
*s++='B';
bytes-=3;
km=KM_ank;
}
}
*s++= (c&0x7f);
--bytes;
}
if (km!=KM_ank&& bytes>=3)
strcpy(s,"\x1b(B"); else
*s='\0';
}
return p;
}
