/*
* Copyright (c) 1987, Fujitsu LTD. (Itaru ICHIKAWA).
- * Copyright (c) 1996-2009, The nkf Project.
+ * Copyright (c) 1996-2018, The nkf Project.
*
* This software is provided 'as-is', without any express or implied
* warranty. In no event will the authors be held liable for any damages
*
* 3. This notice may not be removed or altered from any source distribution.
*/
-#define NKF_VERSION "2.1.0"
-#define NKF_RELEASE_DATE "2009-11-17"
+#define NKF_VERSION "2.1.5"
+#define NKF_RELEASE_DATE "2018-12-15"
#define COPY_RIGHT \
"Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa).\n" \
- "Copyright (C) 1996-2009, The nkf Project."
+ "Copyright (C) 1996-2018, The nkf Project."
#include "config.h"
#include "nkf.h"
} encoding_name_to_id_table[] = {
{"US-ASCII", ASCII},
{"ASCII", ASCII},
+ {"646", ASCII},
+ {"ROMAN8", ASCII},
{"ISO-2022-JP", ISO_2022_JP},
{"ISO2022JP-CP932", CP50220},
{"CP50220", CP50220},
{"ISO-2022-JP-2004", ISO_2022_JP_2004},
{"SHIFT_JIS", SHIFT_JIS},
{"SJIS", SHIFT_JIS},
+ {"MS_Kanji", SHIFT_JIS},
+ {"PCK", SHIFT_JIS},
{"WINDOWS-31J", WINDOWS_31J},
{"CSWINDOWS31J", WINDOWS_31J},
{"CP932", WINDOWS_31J},
/* ignore ZERO WIDTH NO-BREAK SPACE */
static int no_best_fit_chars_f = FALSE;
static int input_endian = ENDIAN_BIG;
+static int input_bom_f = FALSE;
static nkf_char unicode_subchar = '?'; /* the regular substitution character */
static void (*encode_fallback)(nkf_char c) = NULL;
static void w_status(struct input_code *, nkf_char);
static unsigned char stdobuf[IOBUF_SIZE];
#endif
+#define NKF_UNSPECIFIED (-TRUE)
+
/* flags */
static int unbuf_f = FALSE;
static int estab_f = FALSE;
static int broken_f = FALSE; /* convert ESC-less broken JIS */
static int iso8859_f = FALSE; /* ISO8859 through */
static int mimeout_f = FALSE; /* base64 mode */
-static int x0201_f = X0201_DEFAULT; /* convert JIS X 0201 */
+static int x0201_f = NKF_UNSPECIFIED; /* convert JIS X 0201 */
static int iso2022jp_f = FALSE; /* replace non ISO-2022-JP with GETA */
#ifdef UNICODE_NORMALIZATION
#define nkf_char_unicode_bmp_p(c) ((c & VALUE_MASK) <= UNICODE_BMP_MAX)
#define nkf_char_unicode_value_p(c) ((c & VALUE_MASK) <= UNICODE_MAX)
+#define UTF16_TO_UTF32(lead, trail) (((lead) << 10) + (trail) - NKF_INT32_C(0x35FDC00))
+
#ifdef NUMCHAR_OPTION
static int numchar_f = FALSE;
static nkf_char (*i_ngetc)(FILE *) = std_getc; /* input of ugetc */
{"UTF-16", 0, 0, 0, {0, 0, 0}, NULL, w_iconv16, 0},
{"UTF-32", 0, 0, 0, {0, 0, 0}, NULL, w_iconv32, 0},
#endif
- {0}
+ {NULL, 0, 0, 0, {0, 0, 0}, NULL, NULL, 0}
};
static int mimeout_mode = 0; /* 0, -1, 'Q', 'B', 1, 2 */
/* process default */
static nkf_char
-no_connection2(nkf_char c2, nkf_char c1, nkf_char c0)
+no_connection2(ARG_UNUSED nkf_char c2, ARG_UNUSED nkf_char c1, ARG_UNUSED nkf_char c0)
{
fprintf(stderr,"nkf internal module connection failure.\n");
exit(EXIT_FAILURE);
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00};
+/* X0201 kana to X0213 conversion table for han-daguten */
+/* 90-9F A0-DF */
+static const unsigned char ev_x0213[]= {
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x25,0x77,0x25,0x78,
+ 0x25,0x79,0x25,0x7a,0x25,0x7b,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x25,0x7c,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x25,0x7d,0x00,0x00,
+ 0x25,0x7e,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+ 0x00,0x00};
+
/* X0208 kigou conversion table */
/* 0x8140 - 0x819e */
#ifdef DEFAULT_CODE_LOCALE
static const char*
-nkf_locale_charmap()
+nkf_locale_charmap(void)
{
#ifdef HAVE_LANGINFO_H
return nl_langinfo(CODESET);
}
static nkf_encoding*
-nkf_locale_encoding()
+nkf_locale_encoding(void)
{
nkf_encoding *enc = 0;
const char *encname = nkf_locale_charmap();
#endif /* DEFAULT_CODE_LOCALE */
static nkf_encoding*
-nkf_utf8_encoding()
+nkf_utf8_encoding(void)
{
return &nkf_encoding_table[UTF_8];
}
static nkf_encoding*
-nkf_default_encoding()
+nkf_default_encoding(void)
{
nkf_encoding *enc = 0;
#ifdef DEFAULT_CODE_LOCALE
nkf_buf_new(int length)
{
nkf_buf_t *buf = nkf_xmalloc(sizeof(nkf_buf_t));
- buf->ptr = nkf_xmalloc(length);
+ buf->ptr = nkf_xmalloc(sizeof(nkf_char) * length);
buf->capa = length;
buf->len = 0;
return buf;
-}
+}
#if 0
static void
#define nkf_buf_length(buf) ((buf)->len)
#define nkf_buf_empty_p(buf) ((buf)->len == 0)
-static unsigned char
+static nkf_char
nkf_buf_at(nkf_buf_t *buf, int index)
{
assert(index <= buf->len);
buf->ptr[buf->len++] = c;
}
-static unsigned char
+static nkf_char
nkf_buf_pop(nkf_buf_t *buf)
{
assert(!nkf_buf_empty_p(buf));
" Z[0-4] Default/0: Convert JISX0208 Alphabet to ASCII\n"
" 1: Kankaku to one space 2: to two spaces 3: HTML Entity\n"
" 4: JISX0208 Katakana to JISX0201 Katakana\n"
- " X,x Assume X0201 kana in MS-Kanji, -x preserves X0201\n"
+ " X,x Convert Halfwidth Katakana to Fullwidth or preserve it\n"
);
fprintf(HELP_OUTPUT,
" O Output to File (DEFAULT 'nkf.out')\n"
int shift = 20;
c &= VALUE_MASK;
while(shift >= 0){
- if(c >= 1<<shift){
+ if(c >= NKF_INT32_C(1)<<shift){
while(shift >= 0){
(*f)(0, bin2hex(c>>shift));
shift -= 4;
(*oconv)(0, '\\');
c &= VALUE_MASK;
if(!nkf_char_unicode_bmp_p(c)){
- (*oconv)(0, 'U');
- (*oconv)(0, '0');
- (*oconv)(0, '0');
- (*oconv)(0, bin2hex(c>>20));
- (*oconv)(0, bin2hex(c>>16));
+ int high = (c >> 10) + NKF_INT32_C(0xD7C0); /* high surrogate */
+ int low = (c & 0x3FF) + NKF_INT32_C(0xDC00); /* low surrogate */
+ (*oconv)(0, 'u');
+ (*oconv)(0, bin2hex(high>>12));
+ (*oconv)(0, bin2hex(high>> 8));
+ (*oconv)(0, bin2hex(high>> 4));
+ (*oconv)(0, bin2hex(high ));
+ (*oconv)(0, '\\');
+ (*oconv)(0, 'u');
+ (*oconv)(0, bin2hex(low>>12));
+ (*oconv)(0, bin2hex(low>> 8));
+ (*oconv)(0, bin2hex(low>> 4));
+ (*oconv)(0, bin2hex(low ));
}else{
(*oconv)(0, 'u');
+ (*oconv)(0, bin2hex(c>>12));
+ (*oconv)(0, bin2hex(c>> 8));
+ (*oconv)(0, bin2hex(c>> 4));
+ (*oconv)(0, bin2hex(c ));
}
- (*oconv)(0, bin2hex(c>>12));
- (*oconv)(0, bin2hex(c>> 8));
- (*oconv)(0, bin2hex(c>> 4));
- (*oconv)(0, bin2hex(c ));
return;
}
case ISO_8859_1:
iso8859_f = TRUE;
break;
- case CP50220:
case CP50221:
case CP50222:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
+ case CP50220:
#ifdef SHIFTJIS_CP932
cp51932_f = TRUE;
#endif
case SHIFT_JIS:
break;
case WINDOWS_31J:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
cp51932_f = TRUE;
#endif
case EUCJP_NKF:
break;
case CP51932:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
cp51932_f = TRUE;
#endif
#endif
break;
case EUCJP_MS:
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
cp51932_f = FALSE;
#endif
#endif
break;
case EUCJP_ASCII:
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
cp51932_f = FALSE;
#endif
x0213_f = TRUE;
#ifdef SHIFTJIS_CP932
cp51932_f = FALSE;
+ if (cp932inv_f == TRUE) cp932inv_f = FALSE;
#endif
break;
case EUC_JISX0213:
{
switch (nkf_enc_to_index(enc)) {
case CP50220:
- x0201_f = TRUE;
#ifdef SHIFTJIS_CP932
if (cp932inv_f == TRUE) cp932inv_f = FALSE;
#endif
#endif
break;
case CP50221:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
if (cp932inv_f == TRUE) cp932inv_f = FALSE;
#endif
#endif
break;
case ISO_2022_JP_3:
+ case ISO_2022_JP_2004:
x0212_f = TRUE;
x0213_f = TRUE;
#ifdef SHIFTJIS_CP932
case SHIFT_JIS:
break;
case WINDOWS_31J:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef UTF8_OUTPUT_ENABLE
ms_ucs_map_f = UCS_MAP_CP932;
#endif
#endif
break;
case CP51932:
- x0201_f = TRUE;
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
#ifdef SHIFTJIS_CP932
if (cp932inv_f == TRUE) cp932inv_f = FALSE;
#endif
#endif
break;
case EUCJP_MS:
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
x0212_f = TRUE;
#ifdef UTF8_OUTPUT_ENABLE
ms_ucs_map_f = UCS_MAP_MS;
#endif
break;
case EUCJP_ASCII:
+ if (x0201_f == NKF_UNSPECIFIED) x0201_f = FALSE; /* -x specified implicitly */
x0212_f = TRUE;
#ifdef UTF8_OUTPUT_ENABLE
ms_ucs_map_f = UCS_MAP_ASCII;
}
#endif /* X0212_ENABLE */
+static int
+is_x0213_2_in_x0212(nkf_char c1)
+{
+ static const char x0213_2_table[] =
+ {0, 1, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 1, 1};
+ int ku = c1 - 0x20;
+ if (ku <= 15)
+ return x0213_2_table[ku]; /* 1, 3-5, 8, 12-15 */
+ if (78 <= ku && ku <= 94)
+ return 1;
+ return 0;
+}
+
static nkf_char
e2s_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1)
{
nkf_char ndx;
if (is_eucg3(c2)){
ndx = c2 & 0x7f;
- if (x0213_f){
+ if (x0213_f && is_x0213_2_in_x0212(ndx)){
if((0x21 <= ndx && ndx <= 0x2F)){
if (p2) *p2 = ((ndx - 1) >> 1) + 0xec - ndx / 8 * 3;
if (p1) *p1 = c1 + ((ndx & 1) ? ((c1 < 0x60) ? 0x1f : 0x20) : 0x7e);
static const char shift_jisx0213_s1a3_table[5][2] ={ { 1, 8}, { 3, 4}, { 5,12}, {13,14}, {15, 0} };
if (0xFC < c1) return 1;
#ifdef SHIFTJIS_CP932
- if (!cp932inv_f && is_ibmext_in_sjis(c2)){
+ if (!cp932inv_f && !x0213_f && is_ibmext_in_sjis(c2)){
val = shiftjis_cp932[c2 - CP932_TABLE_BEGIN][c1 - 0x40];
if (val){
c2 = val >> 8;
/* single byte */
wc = c1;
}
- else if (c1 <= 0xC3) {
+ else if (c1 <= 0xC1) {
/* trail byte or invalid */
return -1;
}
ms_ucs_map_f == UCS_MAP_CP932 ? utf8_to_euc_2bytes_932 :
ms_ucs_map_f == UCS_MAP_MS ? utf8_to_euc_2bytes_ms :
ms_ucs_map_f == UCS_MAP_CP10001 ? utf8_to_euc_2bytes_mac :
+ x0213_f ? utf8_to_euc_2bytes_x0213 :
utf8_to_euc_2bytes;
ret = unicode_to_jis_common2(c2, c1, pp, sizeof_utf8_to_euc_2bytes, p2, p1);
}else if(c0 < 0xF0){
ms_ucs_map_f == UCS_MAP_CP932 ? utf8_to_euc_3bytes_932 :
ms_ucs_map_f == UCS_MAP_MS ? utf8_to_euc_3bytes_ms :
ms_ucs_map_f == UCS_MAP_CP10001 ? utf8_to_euc_3bytes_mac :
+ x0213_f ? utf8_to_euc_3bytes_x0213 :
utf8_to_euc_3bytes;
ret = unicode_to_jis_common2(c1, c0, ppp[c2 - 0xE0], sizeof_utf8_to_euc_C2, p2, p1);
}else return -1;
#ifdef SHIFTJIS_CP932
- if (!ret && !cp932inv_f && is_eucg3(*p2)) {
- nkf_char s2, s1;
- if (e2s_conv(*p2, *p1, &s2, &s1) == 0) {
- s2e_conv(s2, s1, p2, p1);
- }else{
- ret = 1;
+ if (!ret&& is_eucg3(*p2)) {
+ if (cp932inv_f) {
+ if (encode_fallback) ret = 1;
+ }
+ else {
+ nkf_char s2, s1;
+ if (e2s_conv(*p2, *p1, &s2, &s1) == 0) {
+ s2e_conv(s2, s1, p2, p1);
+ }else{
+ ret = 1;
+ }
}
}
#endif
}
#ifdef UTF8_OUTPUT_ENABLE
+#define X0213_SURROGATE_FIND(tbl, size, euc) do { \
+ int i; \
+ for (i = 0; i < size; i++) \
+ if (tbl[i][0] == euc) { \
+ low = tbl[i][2]; \
+ break; \
+ } \
+ } while (0)
+
static nkf_char
e2w_conv(nkf_char c2, nkf_char c1)
{
}
c2 = (c2&0x7f) - 0x21;
if (0<=c2 && c2<sizeof_euc_to_utf8_2bytes)
- p = x0212_to_utf8_2bytes[c2];
+ p =
+ x0213_f ? x0212_to_utf8_2bytes_x0213[c2] :
+ x0212_to_utf8_2bytes[c2];
else
return 0;
#endif
c2 = (c2&0x7f) - 0x21;
if (0<=c2 && c2<sizeof_euc_to_utf8_2bytes)
p =
+ x0213_f ? euc_to_utf8_2bytes_x0213[c2] :
ms_ucs_map_f == UCS_MAP_ASCII ? euc_to_utf8_2bytes[c2] :
ms_ucs_map_f == UCS_MAP_CP10001 ? euc_to_utf8_2bytes_mac[c2] :
euc_to_utf8_2bytes_ms[c2];
}
if (!p) return 0;
c1 = (c1 & 0x7f) - 0x21;
- if (0<=c1 && c1<sizeof_euc_to_utf8_1byte)
- return p[c1];
+ if (0<=c1 && c1<sizeof_euc_to_utf8_1byte) {
+ nkf_char val = p[c1];
+ if (x0213_f && 0xD800<=val && val<=0xDBFF) {
+ nkf_char euc = (c2+0x21)<<8 | (c1+0x21);
+ nkf_char low = 0;
+ if (p==x0212_to_utf8_2bytes_x0213[c2]) {
+ X0213_SURROGATE_FIND(x0213_2_surrogate_table, sizeof_x0213_2_surrogate_table, euc);
+ } else {
+ X0213_SURROGATE_FIND(x0213_1_surrogate_table, sizeof_x0213_1_surrogate_table, euc);
+ }
+ if (!low) return 0;
+ return UTF16_TO_UTF32(val, low);
+ } else {
+ return val;
+ }
+ }
+ return 0;
+}
+
+static nkf_char
+e2w_combining(nkf_char comb, nkf_char c2, nkf_char c1)
+{
+ nkf_char euc;
+ int i;
+ for (i = 0; i < sizeof_x0213_combining_chars; i++)
+ if (x0213_combining_chars[i] == comb)
+ break;
+ if (i >= sizeof_x0213_combining_chars)
+ return 0;
+ euc = (c2&0x7f)<<8 | (c1&0x7f);
+ for (i = 0; i < sizeof_x0213_combining_table; i++)
+ if (x0213_combining_table[i][0] == euc)
+ return x0213_combining_table[i][1];
return 0;
}
#endif
}
}
else {
+ int i;
+ if (x0213_f) {
+ c1 = (val >> 10) + NKF_INT32_C(0xD7C0); /* high surrogate */
+ c2 = (val & 0x3FF) + NKF_INT32_C(0xDC00); /* low surrogate */
+ for (i = 0; i < sizeof_x0213_1_surrogate_table; i++)
+ if (x0213_1_surrogate_table[i][1] == c1 && x0213_1_surrogate_table[i][2] == c2) {
+ val = x0213_1_surrogate_table[i][0];
+ *p2 = val >> 8;
+ *p1 = val & 0xFF;
+ return 0;
+ }
+ for (i = 0; i < sizeof_x0213_2_surrogate_table; i++)
+ if (x0213_2_surrogate_table[i][1] == c1 && x0213_2_surrogate_table[i][2] == c2) {
+ val = x0213_2_surrogate_table[i][0];
+ *p2 = PREFIX_EUCG3 | (val >> 8);
+ *p1 = val & 0xFF;
+ return 0;
+ }
+ }
*p2 = 0;
*p1 = nkf_char_unicode_new(val);
}
}
static nkf_char
-s_iconv(nkf_char c2, nkf_char c1, nkf_char c0)
+s_iconv(ARG_UNUSED nkf_char c2, nkf_char c1, ARG_UNUSED nkf_char c0)
{
if (c2 == JIS_X_0201_1976_K || (0xA1 <= c2 && c2 <= 0xDF)) {
if (iso2022jp_f && !x0201_f) {
return 0;
}
+static int
+x0213_wait_combining_p(nkf_char wc)
+{
+ int i;
+ for (i = 0; i < sizeof_x0213_combining_table; i++) {
+ if (x0213_combining_table[i][1] == wc) {
+ return TRUE;
+ }
+ }
+ return FALSE;
+}
+
+static int
+x0213_combining_p(nkf_char wc)
+{
+ int i;
+ for (i = 0; i < sizeof_x0213_combining_chars; i++) {
+ if (x0213_combining_chars[i] == wc) {
+ return TRUE;
+ }
+ }
+ return FALSE;
+}
+
static nkf_char
w_iconv(nkf_char c1, nkf_char c2, nkf_char c3)
{
c2 = nkf_char_unicode_new(nkf_utf8_to_unicode(c1, c2, c3, c4));
c1 = 0;
} else {
+ if (x0213_f && x0213_wait_combining_p(nkf_utf8_to_unicode(c1, c2, c3, c4)))
+ return -3;
ret = w2e_conv(c1, c2, c3, &c1, &c2);
}
if (ret == 0){
return ret;
}
+static nkf_char
+w_iconv_nocombine(nkf_char c1, nkf_char c2, nkf_char c3)
+{
+ /* continue from the line below 'return -3;' in w_iconv() */
+ nkf_char ret = w2e_conv(c1, c2, c3, &c1, &c2);
+ if (ret == 0){
+ (*oconv)(c1, c2);
+ }
+ return ret;
+}
+
#define NKF_ICONV_INVALID_CODE_RANGE -13
+#define NKF_ICONV_WAIT_COMBINING_CHAR -14
+#define NKF_ICONV_NOT_COMBINED -15
static size_t
-unicode_iconv(nkf_char wc)
+unicode_iconv(nkf_char wc, int nocombine)
{
nkf_char c1, c2;
int ret = 0;
/* unpaired surrogate */
return NKF_ICONV_INVALID_CODE_RANGE;
}else if (wc < 0xFFFF) {
+ if (!nocombine && x0213_f && x0213_wait_combining_p(wc))
+ return NKF_ICONV_WAIT_COMBINING_CHAR;
ret = w16e_conv(wc, &c2, &c1);
if (ret) return ret;
}else if (wc < 0x10FFFF) {
return 0;
}
-#define NKF_ICONV_NEED_ONE_MORE_BYTE -1
-#define NKF_ICONV_NEED_TWO_MORE_BYTES -2
-#define UTF16_TO_UTF32(lead, trail) (((lead) << 10) + (trail) - NKF_INT32_C(0x35FDC00))
+static nkf_char
+unicode_iconv_combine(nkf_char wc, nkf_char wc2)
+{
+ nkf_char c1, c2;
+ int i;
+
+ if (wc2 < 0x80) {
+ return NKF_ICONV_NOT_COMBINED;
+ }else if ((wc2>>11) == 27) {
+ /* unpaired surrogate */
+ return NKF_ICONV_INVALID_CODE_RANGE;
+ }else if (wc2 < 0xFFFF) {
+ if (!x0213_combining_p(wc2))
+ return NKF_ICONV_NOT_COMBINED;
+ for (i = 0; i < sizeof_x0213_combining_table; i++) {
+ if (x0213_combining_table[i][1] == wc &&
+ x0213_combining_table[i][2] == wc2) {
+ c2 = x0213_combining_table[i][0] >> 8;
+ c1 = x0213_combining_table[i][0] & 0x7f;
+ (*oconv)(c2, c1);
+ return 0;
+ }
+ }
+ }else if (wc2 < 0x10FFFF) {
+ return NKF_ICONV_NOT_COMBINED;
+ } else {
+ return NKF_ICONV_INVALID_CODE_RANGE;
+ }
+ return NKF_ICONV_NOT_COMBINED;
+}
+
+static nkf_char
+w_iconv_combine(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4, nkf_char c5, nkf_char c6)
+{
+ nkf_char wc, wc2;
+ wc = nkf_utf8_to_unicode(c1, c2, c3, 0);
+ wc2 = nkf_utf8_to_unicode(c4, c5, c6, 0);
+ if (wc2 < 0)
+ return wc2;
+ return unicode_iconv_combine(wc, wc2);
+}
+
+#define NKF_ICONV_NEED_ONE_MORE_BYTE (size_t)-1
+#define NKF_ICONV_NEED_TWO_MORE_BYTES (size_t)-2
static size_t
nkf_iconv_utf_16(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
{
}
}
- return (*unicode_iconv)(wc);
+ return (*unicode_iconv)(wc, FALSE);
+}
+
+static size_t
+nkf_iconv_utf_16_combine(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
+{
+ nkf_char wc, wc2;
+
+ if (input_endian == ENDIAN_BIG) {
+ if (0xD8 <= c3 && c3 <= 0xDB) {
+ return NKF_ICONV_NOT_COMBINED;
+ } else {
+ wc = c1 << 8 | c2;
+ wc2 = c3 << 8 | c4;
+ }
+ } else {
+ if (0xD8 <= c2 && c2 <= 0xDB) {
+ return NKF_ICONV_NOT_COMBINED;
+ } else {
+ wc = c2 << 8 | c1;
+ wc2 = c4 << 8 | c3;
+ }
+ }
+
+ return unicode_iconv_combine(wc, wc2);
+}
+
+static size_t
+nkf_iconv_utf_16_nocombine(nkf_char c1, nkf_char c2)
+{
+ nkf_char wc;
+ if (input_endian == ENDIAN_BIG)
+ wc = c1 << 8 | c2;
+ else
+ wc = c2 << 8 | c1;
+ return (*unicode_iconv)(wc, TRUE);
}
static nkf_char
-w_iconv16(nkf_char c2, nkf_char c1, nkf_char c0)
+w_iconv16(nkf_char c2, nkf_char c1, ARG_UNUSED nkf_char c0)
{
- return 0;
+ (*oconv)(c2, c1);
+ return 16; /* different from w_iconv32 */
}
static nkf_char
-w_iconv32(nkf_char c2, nkf_char c1, nkf_char c0)
+w_iconv32(nkf_char c2, nkf_char c1, ARG_UNUSED nkf_char c0)
{
- return 0;
+ (*oconv)(c2, c1);
+ return 32; /* different from w_iconv16 */
}
-static size_t
-nkf_iconv_utf_32(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
+static nkf_char
+utf32_to_nkf_char(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
{
nkf_char wc;
- if (c1 == EOF) {
- (*oconv)(EOF, 0);
- return 0;
- }
-
switch(input_endian){
case ENDIAN_BIG:
wc = c2 << 16 | c3 << 8 | c4;
default:
return NKF_ICONV_INVALID_CODE_RANGE;
}
+ return wc;
+}
+
+static size_t
+nkf_iconv_utf_32(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
+{
+ nkf_char wc;
+
+ if (c1 == EOF) {
+ (*oconv)(EOF, 0);
+ return 0;
+ }
+
+ wc = utf32_to_nkf_char(c1, c2, c3, c4);
+ if (wc < 0)
+ return wc;
+
+ return (*unicode_iconv)(wc, FALSE);
+}
- return (*unicode_iconv)(wc);
+static nkf_char
+nkf_iconv_utf_32_combine(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4, nkf_char c5, nkf_char c6, nkf_char c7, nkf_char c8)
+{
+ nkf_char wc, wc2;
+
+ wc = utf32_to_nkf_char(c1, c2, c3, c4);
+ if (wc < 0)
+ return wc;
+ wc2 = utf32_to_nkf_char(c5, c6, c7, c8);
+ if (wc2 < 0)
+ return wc2;
+
+ return unicode_iconv_combine(wc, wc2);
+}
+
+static size_t
+nkf_iconv_utf_32_nocombine(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
+{
+ nkf_char wc;
+
+ wc = utf32_to_nkf_char(c1, c2, c3, c4);
+ return (*unicode_iconv)(wc, TRUE);
}
#endif
}
#ifdef UTF8_OUTPUT_ENABLE
+#define OUTPUT_UTF8(val) do { \
+ nkf_unicode_to_utf8(val, &c1, &c2, &c3, &c4); \
+ (*o_putc)(c1); \
+ if (c2) (*o_putc)(c2); \
+ if (c3) (*o_putc)(c3); \
+ if (c4) (*o_putc)(c4); \
+ } while (0)
+
static void
w_oconv(nkf_char c2, nkf_char c1)
{
nkf_char c3, c4;
- nkf_char val;
+ nkf_char val, val2;
if (output_bom_f) {
output_bom_f = FALSE;
if (c2 == 0 && nkf_char_unicode_p(c1)){
val = c1 & VALUE_MASK;
- nkf_unicode_to_utf8(val, &c1, &c2, &c3, &c4);
- (*o_putc)(c1);
- if (c2) (*o_putc)(c2);
- if (c3) (*o_putc)(c3);
- if (c4) (*o_putc)(c4);
+ OUTPUT_UTF8(val);
return;
}
} else {
val = e2w_conv(c2, c1);
if (val){
- nkf_unicode_to_utf8(val, &c1, &c2, &c3, &c4);
- (*o_putc)(c1);
- if (c2) (*o_putc)(c2);
- if (c3) (*o_putc)(c3);
- if (c4) (*o_putc)(c4);
+ val2 = e2w_combining(val, c2, c1);
+ if (val2)
+ OUTPUT_UTF8(val2);
+ OUTPUT_UTF8(val);
}
}
}
+#define OUTPUT_UTF16_BYTES(c1, c2) do { \
+ if (output_endian == ENDIAN_LITTLE){ \
+ (*o_putc)(c1); \
+ (*o_putc)(c2); \
+ }else{ \
+ (*o_putc)(c2); \
+ (*o_putc)(c1); \
+ } \
+ } while (0)
+
+#define OUTPUT_UTF16(val) do { \
+ if (nkf_char_unicode_bmp_p(val)) { \
+ c2 = (val >> 8) & 0xff; \
+ c1 = val & 0xff; \
+ OUTPUT_UTF16_BYTES(c1, c2); \
+ } else { \
+ val &= VALUE_MASK; \
+ if (val <= UNICODE_MAX) { \
+ c2 = (val >> 10) + NKF_INT32_C(0xD7C0); /* high surrogate */ \
+ c1 = (val & 0x3FF) + NKF_INT32_C(0xDC00); /* low surrogate */ \
+ OUTPUT_UTF16_BYTES(c2 & 0xff, (c2 >> 8) & 0xff); \
+ OUTPUT_UTF16_BYTES(c1 & 0xff, (c1 >> 8) & 0xff); \
+ } \
+ } \
+ } while (0)
+
static void
w_oconv16(nkf_char c2, nkf_char c1)
{
if (output_bom_f) {
output_bom_f = FALSE;
- if (output_endian == ENDIAN_LITTLE){
- (*o_putc)(0xFF);
- (*o_putc)(0xFE);
- }else{
- (*o_putc)(0xFE);
- (*o_putc)(0xFF);
- }
+ OUTPUT_UTF16_BYTES(0xFF, 0xFE);
}
if (c2 == EOF) {
}
if (c2 == 0 && nkf_char_unicode_p(c1)) {
- if (nkf_char_unicode_bmp_p(c1)) {
- c2 = (c1 >> 8) & 0xff;
- c1 &= 0xff;
- } else {
- c1 &= VALUE_MASK;
- if (c1 <= UNICODE_MAX) {
- c2 = (c1 >> 10) + NKF_INT32_C(0xD7C0); /* high surrogate */
- c1 = (c1 & 0x3FF) + NKF_INT32_C(0xDC00); /* low surrogate */
- if (output_endian == ENDIAN_LITTLE){
- (*o_putc)(c2 & 0xff);
- (*o_putc)((c2 >> 8) & 0xff);
- (*o_putc)(c1 & 0xff);
- (*o_putc)((c1 >> 8) & 0xff);
- }else{
- (*o_putc)((c2 >> 8) & 0xff);
- (*o_putc)(c2 & 0xff);
- (*o_putc)((c1 >> 8) & 0xff);
- (*o_putc)(c1 & 0xff);
- }
- }
- return;
- }
+ OUTPUT_UTF16(c1);
} else if (c2) {
- nkf_char val = e2w_conv(c2, c1);
- c2 = (val >> 8) & 0xff;
- c1 = val & 0xff;
+ nkf_char val, val2;
+ val = e2w_conv(c2, c1);
if (!val) return;
- }
-
- if (output_endian == ENDIAN_LITTLE){
- (*o_putc)(c1);
- (*o_putc)(c2);
- }else{
- (*o_putc)(c2);
- (*o_putc)(c1);
+ val2 = e2w_combining(val, c2, c1);
+ if (val2)
+ OUTPUT_UTF16(val2);
+ OUTPUT_UTF16(val);
+ } else {
+ OUTPUT_UTF16_BYTES(c1, c2);
}
}
+#define OUTPUT_UTF32(c) do { \
+ if (output_endian == ENDIAN_LITTLE){ \
+ (*o_putc)( (c) & 0xFF); \
+ (*o_putc)(((c) >> 8) & 0xFF); \
+ (*o_putc)(((c) >> 16) & 0xFF); \
+ (*o_putc)(0); \
+ }else{ \
+ (*o_putc)(0); \
+ (*o_putc)(((c) >> 16) & 0xFF); \
+ (*o_putc)(((c) >> 8) & 0xFF); \
+ (*o_putc)( (c) & 0xFF); \
+ } \
+ } while (0)
+
static void
w_oconv32(nkf_char c2, nkf_char c1)
{
} else if (c2 == 0 && nkf_char_unicode_p(c1)) {
c1 &= VALUE_MASK;
} else if (c2) {
- c1 = e2w_conv(c2, c1);
- if (!c1) return;
- }
- if (output_endian == ENDIAN_LITTLE){
- (*o_putc)( c1 & 0xFF);
- (*o_putc)((c1 >> 8) & 0xFF);
- (*o_putc)((c1 >> 16) & 0xFF);
- (*o_putc)(0);
- }else{
- (*o_putc)(0);
- (*o_putc)((c1 >> 16) & 0xFF);
- (*o_putc)((c1 >> 8) & 0xFF);
- (*o_putc)( c1 & 0xFF);
+ nkf_char val, val2;
+ val = e2w_conv(c2, c1);
+ if (!val) return;
+ val2 = e2w_combining(val, c2, c1);
+ if (val2)
+ OUTPUT_UTF32(val2);
+ c1 = val;
}
+ OUTPUT_UTF32(c1);
}
#endif
#define SCORE_DEPEND (SCORE_KANA << 1) /* MD Characters */
#define SCORE_CP932 (SCORE_DEPEND << 1) /* IBM extended characters */
#define SCORE_X0212 (SCORE_CP932 << 1) /* JIS X 0212 */
-#define SCORE_NO_EXIST (SCORE_X0212 << 1) /* Undefined Characters */
+#define SCORE_X0213 (SCORE_X0212 << 1) /* JIS X 0213 */
+#define SCORE_NO_EXIST (SCORE_X0213 << 1) /* Undefined Characters */
#define SCORE_iMIME (SCORE_NO_EXIST << 1) /* MIME selected */
#define SCORE_ERROR (SCORE_iMIME << 1) /* Error */
0, 0, 0, 0,
0, 0, 0, 0,
0, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND,
- SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_NO_EXIST,
+ SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_X0213,
};
static const nkf_char score_table_F0[] = {
SCORE_L2, SCORE_L2, SCORE_L2, SCORE_L2,
- SCORE_L2, SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST,
+ SCORE_L2, SCORE_DEPEND, SCORE_X0213, SCORE_X0213,
SCORE_DEPEND, SCORE_DEPEND, SCORE_CP932, SCORE_CP932,
- SCORE_CP932, SCORE_NO_EXIST, SCORE_NO_EXIST, SCORE_ERROR,
+ SCORE_CP932, SCORE_X0213, SCORE_X0213, SCORE_ERROR,
+};
+
+static const nkf_char score_table_8FA0[] = {
+ 0, SCORE_X0213, SCORE_X0212, SCORE_X0213,
+ SCORE_X0213, SCORE_X0213, SCORE_X0212, SCORE_X0212,
+ SCORE_X0213, SCORE_X0212, SCORE_X0212, SCORE_X0212,
+ SCORE_X0213, SCORE_X0213, SCORE_X0213, SCORE_X0213,
+};
+
+static const nkf_char score_table_8FE0[] = {
+ SCORE_X0212, SCORE_X0212, SCORE_X0212, SCORE_X0212,
+ SCORE_X0212, SCORE_X0212, SCORE_X0212, SCORE_X0212,
+ SCORE_X0212, SCORE_X0212, SCORE_X0212, SCORE_X0212,
+ SCORE_X0212, SCORE_X0212, SCORE_X0213, SCORE_X0213,
+};
+
+static const nkf_char score_table_8FF0[] = {
+ SCORE_X0213, SCORE_X0213, SCORE_X0213, SCORE_X0212,
+ SCORE_X0212, SCORE_X0213, SCORE_X0213, SCORE_X0213,
+ SCORE_X0213, SCORE_X0213, SCORE_X0213, SCORE_X0213,
+ SCORE_X0213, SCORE_X0213, SCORE_X0213, SCORE_X0213,
};
static void
code_score(struct input_code *ptr)
{
nkf_char c2 = ptr->buf[0];
-#ifdef UTF8_OUTPUT_ENABLE
nkf_char c1 = ptr->buf[1];
-#endif
if (c2 < 0){
set_code_score(ptr, SCORE_ERROR);
}else if (c2 == SS2){
set_code_score(ptr, SCORE_KANA);
}else if (c2 == 0x8f){
- set_code_score(ptr, SCORE_X0212);
+ if ((c1 & 0x70) == 0x20){
+ set_code_score(ptr, score_table_8FA0[c1 & 0x0f]);
+ }else if ((c1 & 0x70) == 0x60){
+ set_code_score(ptr, score_table_8FE0[c1 & 0x0f]);
+ }else if ((c1 & 0x70) == 0x70){
+ set_code_score(ptr, score_table_8FF0[c1 & 0x0f]);
+ }else{
+ set_code_score(ptr, SCORE_X0212);
+ }
#ifdef UTF8_OUTPUT_ENABLE
}else if (!e2w_conv(c2, c1)){
set_code_score(ptr, SCORE_NO_EXIST);
#endif /*WIN32DLL*/
static nkf_char
-std_ungetc(nkf_char c, FILE *f)
+std_ungetc(nkf_char c, ARG_UNUSED FILE *f)
{
nkf_buf_push(nkf_state->std_gc_buf, c);
return c;
}
#endif /*WIN32DLL*/
-static unsigned char hold_buf[HOLD_SIZE*2];
+static nkf_char hold_buf[HOLD_SIZE*2];
static int hold_count = 0;
static nkf_char
push_hold_buf(nkf_char c2)
{
if (hold_count >= HOLD_SIZE*2)
return (EOF);
- hold_buf[hold_count++] = (unsigned char)c2;
+ hold_buf[hold_count++] = c2;
return ((hold_count >= HOLD_SIZE*2) ? EOF : hold_count);
}
static int
-h_conv(FILE *f, int c1, int c2)
+h_conv(FILE *f, nkf_char c1, nkf_char c2)
{
- int ret, c4, c3;
+ int ret;
int hold_index;
-
+ int fromhold_count;
+ nkf_char c3, c4;
/** it must NOT be in the kanji shifte sequence */
/** it must NOT be written in JIS7 */
hold_index = 0;
while (hold_index < hold_count){
c1 = hold_buf[hold_index++];
- if (c1 <= DEL){
+ if (nkf_char_unicode_p(c1)) {
+ (*oconv)(0, c1);
+ continue;
+ }
+ else if (c1 <= DEL){
(*iconv)(0, c1, 0);
continue;
}else if (iconv == s_iconv && 0xa1 <= c1 && c1 <= 0xdf){
(*iconv)(JIS_X_0201_1976_K, c1, 0);
continue;
}
+ fromhold_count = 1;
if (hold_index < hold_count){
c2 = hold_buf[hold_index++];
+ fromhold_count++;
}else{
c2 = (*i_getc)(f);
if (c2 == EOF){
code_status(c4);
(*iconv)(c1, c2, (c3<<8)|c4);
break;
+ case -3:
+ /* 4 bytes UTF-8 (check combining character) */
+ if (hold_index < hold_count){
+ c3 = hold_buf[hold_index++];
+ fromhold_count++;
+ } else if ((c3 = (*i_getc)(f)) == EOF) {
+ w_iconv_nocombine(c1, c2, 0);
+ break;
+ }
+ if (hold_index < hold_count){
+ c4 = hold_buf[hold_index++];
+ fromhold_count++;
+ } else if ((c4 = (*i_getc)(f)) == EOF) {
+ w_iconv_nocombine(c1, c2, 0);
+ if (fromhold_count <= 2)
+ (*i_ungetc)(c3,f);
+ else
+ hold_index--;
+ continue;
+ }
+ if (w_iconv_combine(c1, c2, 0, c3, c4, 0)) {
+ w_iconv_nocombine(c1, c2, 0);
+ if (fromhold_count <= 2) {
+ (*i_ungetc)(c4,f);
+ (*i_ungetc)(c3,f);
+ } else if (fromhold_count == 3) {
+ (*i_ungetc)(c4,f);
+ hold_index--;
+ } else {
+ hold_index -= 2;
+ }
+ }
+ break;
case -1:
/* 3 bytes EUC or UTF-8 */
if (hold_index < hold_count){
c3 = hold_buf[hold_index++];
+ fromhold_count++;
} else if ((c3 = (*i_getc)(f)) == EOF) {
ret = EOF;
break;
} else {
code_status(c3);
}
- (*iconv)(c1, c2, c3);
+ if ((*iconv)(c1, c2, c3) == -3) {
+ /* 6 bytes UTF-8 (check combining character) */
+ nkf_char c5, c6;
+ if (hold_index < hold_count){
+ c4 = hold_buf[hold_index++];
+ fromhold_count++;
+ } else if ((c4 = (*i_getc)(f)) == EOF) {
+ w_iconv_nocombine(c1, c2, c3);
+ continue;
+ }
+ if (hold_index < hold_count){
+ c5 = hold_buf[hold_index++];
+ fromhold_count++;
+ } else if ((c5 = (*i_getc)(f)) == EOF) {
+ w_iconv_nocombine(c1, c2, c3);
+ if (fromhold_count == 4)
+ hold_index--;
+ else
+ (*i_ungetc)(c4,f);
+ continue;
+ }
+ if (hold_index < hold_count){
+ c6 = hold_buf[hold_index++];
+ fromhold_count++;
+ } else if ((c6 = (*i_getc)(f)) == EOF) {
+ w_iconv_nocombine(c1, c2, c3);
+ if (fromhold_count == 5) {
+ hold_index -= 2;
+ } else if (fromhold_count == 4) {
+ hold_index--;
+ (*i_ungetc)(c5,f);
+ } else {
+ (*i_ungetc)(c5,f);
+ (*i_ungetc)(c4,f);
+ }
+ continue;
+ }
+ if (w_iconv_combine(c1, c2, c3, c4, c5, c6)) {
+ w_iconv_nocombine(c1, c2, c3);
+ if (fromhold_count == 6) {
+ hold_index -= 3;
+ } else if (fromhold_count == 5) {
+ hold_index -= 2;
+ (*i_ungetc)(c6,f);
+ } else if (fromhold_count == 4) {
+ hold_index--;
+ (*i_ungetc)(c6,f);
+ (*i_ungetc)(c5,f);
+ } else {
+ (*i_ungetc)(c6,f);
+ (*i_ungetc)(c5,f);
+ (*i_ungetc)(c4,f);
+ }
+ }
+ }
break;
}
if (c3 == EOF) break;
check_bom(FILE *f)
{
int c2;
+ input_bom_f = FALSE;
switch(c2 = (*i_getc)(f)){
case 0x00:
if((c2 = (*i_getc)(f)) == 0x00){
set_iconv(TRUE, w_iconv32);
}
if (iconv == w_iconv32) {
+ input_bom_f = TRUE;
input_endian = ENDIAN_BIG;
return;
}
set_iconv(TRUE, w_iconv);
}
if (iconv == w_iconv) {
+ input_bom_f = TRUE;
return;
}
(*i_ungetc)(0xBF,f);
}
if (iconv == w_iconv16) {
input_endian = ENDIAN_BIG;
+ input_bom_f = TRUE;
return;
}
(*i_ungetc)(0xFF,f);
}
if (iconv == w_iconv32) {
input_endian = ENDIAN_LITTLE;
+ input_bom_f = TRUE;
return;
}
(*i_ungetc)(0x00,f);
}
if (iconv == w_iconv16) {
input_endian = ENDIAN_LITTLE;
+ input_bom_f = TRUE;
return;
}
(*i_ungetc)(0xFE,f);
}
static nkf_char
-broken_ungetc(nkf_char c, FILE *f)
+broken_ungetc(nkf_char c, ARG_UNUSED FILE *f)
{
if (nkf_buf_length(nkf_state->broken_buf) < 2)
nkf_buf_push(nkf_state->broken_buf, c);
else if (c2 != 0 || c1 != LF) (*o_eol_conv)(c2, c1);
}
+static void
+put_newline(void (*func)(nkf_char))
+{
+ switch (eolmode_f ? eolmode_f : DEFAULT_NEWLINE) {
+ case CRLF:
+ (*func)(0x0D);
+ (*func)(0x0A);
+ break;
+ case CR:
+ (*func)(0x0D);
+ break;
+ case LF:
+ (*func)(0x0A);
+ break;
+ }
+}
+
+static void
+oconv_newline(void (*func)(nkf_char, nkf_char))
+{
+ switch (eolmode_f ? eolmode_f : DEFAULT_NEWLINE) {
+ case CRLF:
+ (*func)(0, 0x0D);
+ (*func)(0, 0x0A);
+ break;
+ case CR:
+ (*func)(0, 0x0D);
+ break;
+ case LF:
+ (*func)(0, 0x0A);
+ break;
+ }
+}
+
/*
Return value of fold_conv()
f_prev = c1;
f_line = 0;
fold_state = CR;
- } else if ((f_prev == c1 && !fold_preserve_f)
- || (f_prev == LF && fold_preserve_f)
+ } else if ((f_prev == c1)
+ || (f_prev == LF)
) { /* duplicate newline */
if (f_line) {
f_line = 0;
f_prev = c1;
if (c2 || c2 == JIS_X_0201_1976_K)
f_prev |= 0x80; /* this is Japanese */
- f_line += char_size(c2,c1);
+ f_line += c2 == JIS_X_0201_1976_K ? 1: char_size(c2,c1);
if (f_line<=fold_len) { /* normal case */
fold_state = 1;
} else {
/* terminator process */
switch(fold_state) {
case LF:
- OCONV_NEWLINE((*o_fconv));
+ oconv_newline(o_fconv);
(*o_fconv)(c2,c1);
break;
case 0:
return;
case CR:
- OCONV_NEWLINE((*o_fconv));
+ oconv_newline(o_fconv);
break;
case TAB:
case SP:
z_prev2 = 0;
(*o_zconv)(ev[(z_prev1-SP)*2], ev[(z_prev1-SP)*2+1]);
return;
+ } else if (x0213_f && c1 == (0xdf&0x7f) && ev_x0213[(z_prev1-SP)*2]) { /* \e$BH>ByE@\e(B */
+ z_prev2 = 0;
+ (*o_zconv)(ev_x0213[(z_prev1-SP)*2], ev_x0213[(z_prev1-SP)*2+1]);
+ return;
}
}
z_prev2 = 0;
(*o_zconv)(cv[(z_prev1-SP)*2], cv[(z_prev1-SP)*2+1]);
}
if (c2 == JIS_X_0201_1976_K) {
- if (dv[(c1-SP)*2] || ev[(c1-SP)*2]) {
+ if (dv[(c1-SP)*2] || ev[(c1-SP)*2] || (x0213_f && ev_x0213[(c1-SP)*2])) {
/* wait for \e$BByE@\e(B or \e$BH>ByE@\e(B */
z_prev1 = c1;
z_prev2 = c2;
0x4D00, 0x4D5E, 0x4D5F, 0x4E00, 0x4E5E, 0x4E5F, 0x4F00, 0x5000,
0x5100, 0x5200, 0x5300, 0x2C00, 0x5400, 0x2D00, 0x5500, 0x2E00,
0x5600, 0x5700, 0x5800, 0x5900, 0x5A00, 0x5B00, 0x0000, 0x5C00,
- 0x0000, 0x0000, 0x2600, 0x5D00, 0x335E, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
+ 0x0000, 0x0000, 0x2600, 0x5D00, 0x335E, 0x0000, 0x0000, 0x365F,
+ 0x375F, 0x385F, 0x395F, 0x3A5F, 0x3E5F, 0x425F, 0x445F, 0x0000
};
if (fullwidth_to_halfwidth[c1-0x20]){
c2 = fullwidth_to_halfwidth[c1-0x20];
}
return;
}
+ } else if (c2 == 0 && nkf_char_unicode_p(c1) &&
+ ((c1&VALUE_MASK) == 0x3099 || (c1&VALUE_MASK) == 0x309A)) { /* \e$B9g@.MQByE@!&H>ByE@\e(B */
+ (*o_zconv)(JIS_X_0201_1976_K, 0x5E + (c1&VALUE_MASK) - 0x3099);
+ return;
}
}
(*o_zconv)(c2,c1);
(const unsigned char *)"\075?ISO-8859-1?Q?",
(const unsigned char *)"\075?ISO-8859-1?B?",
(const unsigned char *)"\075?ISO-2022-JP?B?",
+ (const unsigned char *)"\075?ISO-2022-JP?B?",
(const unsigned char *)"\075?ISO-2022-JP?Q?",
#if defined(UTF8_INPUT_ENABLE)
(const unsigned char *)"\075?UTF-8?B?",
/* \e$B3:Ev$9$k%3!<%I$NM%@hEY$r>e$2$k$?$a$NL\0u\e(B */
nkf_char (*mime_priority_func[])(nkf_char c2, nkf_char c1, nkf_char c0) = {
- e_iconv, s_iconv, 0, 0, 0, 0,
+ e_iconv, s_iconv, 0, 0, 0, 0, 0,
#if defined(UTF8_INPUT_ENABLE)
w_iconv, w_iconv,
#endif
};
static const nkf_char mime_encode[] = {
- EUC_JP, SHIFT_JIS, ISO_8859_1, ISO_8859_1, JIS_X_0208, JIS_X_0201_1976_K,
+ EUC_JP, SHIFT_JIS, ISO_8859_1, ISO_8859_1, JIS_X_0208, JIS_X_0201_1976_K, JIS_X_0201_1976_K,
#if defined(UTF8_INPUT_ENABLE)
UTF_8, UTF_8,
#endif
};
static const nkf_char mime_encode_method[] = {
- 'B', 'B','Q', 'B', 'B', 'Q',
+ 'B', 'B','Q', 'B', 'B', 'B', 'Q',
#if defined(UTF8_INPUT_ENABLE)
'B', 'Q',
#endif
}
static nkf_char
-mime_ungetc(nkf_char c, FILE *f)
+mime_ungetc(nkf_char c, ARG_UNUSED FILE *f)
{
mime_input_buf_unshift(c);
return c;
static nkf_char
mime_getc_buf(FILE *f)
{
- /* we don't keep eof of mime_input_buf, becase it contains ?= as
+ /* we don't keep eof of mime_input_buf, because it contains ?= as
a terminator. It was checked in mime_integrity. */
return ((mimebuf_f)?
(*i_mgetc_buf)(f):mime_input_buf(mime_input_state.input++));
static nkf_char
mime_begin(FILE *f)
{
- nkf_char c1;
+ nkf_char c1 = 0;
int i,k;
/* In NONSTRICT mode, only =? is checked. In case of failure, we */
#ifdef CHECK_OPTION
static void
-no_putc(nkf_char c)
+no_putc(ARG_UNUSED nkf_char c)
{
;
}
if (p->score & (SCORE_DEPEND|SCORE_CP932))
input_codename = "CP932";
} else if (strcmp(input_codename, "EUC-JP") == 0) {
- if (p->score & (SCORE_X0212))
+ if (p->score & SCORE_X0213)
+ input_codename = "EUC-JIS-2004";
+ else if (p->score & (SCORE_X0212))
input_codename = "EUCJP-MS";
else if (p->score & (SCORE_DEPEND|SCORE_CP932))
input_codename = "CP51932";
if (guess_f == 1) {
printf("%s\n", input_codename);
} else {
- printf("%s%s\n",
+ printf("%s%s%s%s\n",
input_codename,
+ iconv != w_iconv16 && iconv != w_iconv32 ? "" :
+ input_endian == ENDIAN_LITTLE ? " LE" :
+ input_endian == ENDIAN_BIG ? " BE" :
+ "[BUG]",
+ input_bom_f ? " (BOM)" : "",
input_eol == CR ? " (CR)" :
input_eol == LF ? " (LF)" :
input_eol == CRLF ? " (CRLF)" :
nkf_char (*u)(nkf_char c ,FILE *f) = i_nungetc;
int i = 0, j;
nkf_char buf[12];
- long c = -1;
+ nkf_char c = -1;
buf[i] = (*g)(f);
if (buf[i] == '&'){
#define MIMEOUT_BUF_LENGTH 74
static struct {
- char buf[MIMEOUT_BUF_LENGTH+1];
+ unsigned char buf[MIMEOUT_BUF_LENGTH+1];
int count;
} mimeout_state;
(*o_mputc)(mimeout_state.buf[i]);
i++;
}
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
(*o_mputc)(SP);
base64_count = 1;
if (mimeout_state.count>0 && nkf_isspace(mimeout_state.buf[i])) {
if (c2 == EOF){
if (base64_count + mimeout_state.count/3*4> 73){
(*o_base64conv)(EOF,0);
- OCONV_NEWLINE((*o_base64conv));
+ oconv_newline(o_base64conv);
(*o_base64conv)(0,SP);
base64_count = 1;
}
} else {
- if (base64_count + mimeout_state.count/3*4> 66) {
+ if ((c2 != 0 || c1 > DEL) && base64_count + mimeout_state.count/3*4> 66) {
(*o_base64conv)(EOF,0);
- OCONV_NEWLINE((*o_base64conv));
+ oconv_newline(o_base64conv);
(*o_base64conv)(0,SP);
base64_count = 1;
mimeout_mode = -1;
mimeout_mode = (output_mode==ASCII ||output_mode == ISO_8859_1) ? 'Q' : 'B';
open_mime(output_mode);
(*o_base64conv)(EOF,0);
- OCONV_NEWLINE((*o_base64conv));
+ oconv_newline(o_base64conv);
(*o_base64conv)(0,SP);
base64_count = 1;
mimeout_mode = -1;
if (base64_count > 71){
if (c!=CR && c!=LF) {
(*o_mputc)('=');
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
}
base64_count = 0;
}
}else{
if (base64_count > 71){
eof_mime();
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
base64_count = 0;
}
if (c == EOF) { /* c==EOF */
} else if (c <= SP) {
close_mime();
if (base64_count > 70) {
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
base64_count = 0;
}
if (!nkf_isblank(c)) {
} else {
if (base64_count > 70) {
close_mime();
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
(*o_mputc)(SP);
base64_count = 1;
open_mime(output_mode);
return;
}
}
- (*o_mputc)(c);
- base64_count++;
+ if (c != 0x1B) {
+ (*o_mputc)(c);
+ base64_count++;
+ return;
+ }
}
- return;
}
if (mimeout_mode <= 0) {
- if (c <= DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) {
+ if (c <= DEL && (output_mode==ASCII || output_mode == ISO_8859_1 ||
+ output_mode == UTF_8)) {
if (nkf_isspace(c)) {
int flag = 0;
if (mimeout_mode == -1) {
i = 0;
for (; i < mimeout_state.count - len; ++i) {
- if (!strncmp(mimeout_state.buf+i, str, len)) {
+ if (!strncmp((char *)(mimeout_state.buf+i), str, len)) {
i += len - 2;
break;
}
}
if (i == 0 || i == mimeout_state.count - len) {
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
base64_count = 0;
if (!nkf_isspace(mimeout_state.buf[0])){
(*o_mputc)(SP);
for (j = 0; j <= i; ++j) {
(*o_mputc)(mimeout_state.buf[j]);
}
- PUT_NEWLINE((*o_mputc));
+ put_newline(o_mputc);
base64_count = 1;
for (; j <= mimeout_state.count; ++j) {
mimeout_state.buf[j - i] = mimeout_state.buf[j];
}
}else{
/* mimeout_mode == 'B', 1, 2 */
- if ( c<=DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) {
+ if (c <= DEL && (output_mode==ASCII || output_mode == ISO_8859_1 ||
+ output_mode == UTF_8)) {
if (lastchar == CR || lastchar == LF){
if (nkf_isblank(c)) {
for (i=0;i<mimeout_state.count;i++) {
mimeout_addchar(mimeout_state.buf[i]);
}
mimeout_state.count = 0;
- } else if (SP<c && c<DEL) {
+ } else {
eof_mime();
for (i=0;i<mimeout_state.count;i++) {
(*o_mputc)(mimeout_state.buf[i]);
mimeout_state.buf[mimeout_state.count++] = (char)c;
if (mimeout_state.count>MIMEOUT_BUF_LENGTH) {
eof_mime();
- for (i=0;i<mimeout_state.count;i++) {
- (*o_mputc)(mimeout_state.buf[i]);
+ for (j=0;j<mimeout_state.count;j++) {
+ (*o_mputc)(mimeout_state.buf[j]);
base64_count++;
}
mimeout_state.count = 0;
size_t input_buffer_size;
char *output_buffer;
size_t output_buffer_size;
-}
+};
static nkf_iconv_t
nkf_iconv_new(char *tocode, char *fromcode)
broken_f = FALSE;
iso8859_f = FALSE;
mimeout_f = FALSE;
- x0201_f = X0201_DEFAULT;
+ x0201_f = NKF_UNSPECIFIED;
iso2022jp_f = FALSE;
#if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
ms_ucs_map_f = UCS_MAP_ASCII;
set_output_encoding(output_encoding);
oconv = nkf_enc_to_oconv(output_encoding);
o_putc = std_putc;
+ if (nkf_enc_unicode_p(output_encoding))
+ output_mode = UTF_8;
+
+ if (x0201_f == NKF_UNSPECIFIED) {
+ x0201_f = X0201_DEFAULT;
+ }
- /* replace continucation module, from output side */
+ /* replace continuation module, from output side */
- /* output redicrection */
+ /* output redirection */
#ifdef CHECK_OPTION
if (noout_f || guess_f){
o_putc = no_putc;
i_getc = std_getc;
i_ungetc = std_ungetc;
- /* input redicrection */
+ /* input redirection */
#ifdef INPUT_OPTION
if (cap_f){
i_cgetc = i_getc; i_getc = cap_getc;
#define NEXT continue /* no output, get next */
#define SKIP c2=0;continue /* no output, get next */
#define MORE c2=c1;continue /* need one more byte */
-#define SEND ; /* output c1 and c2, get next */
+#define SEND (void)0 /* output c1 and c2, get next */
#define LAST break /* end of loop, go closing */
#define set_input_mode(mode) do { \
input_mode = mode; \
(c2 = (*i_getc)(f)) != EOF &&
(c3 = (*i_getc)(f)) != EOF &&
(c4 = (*i_getc)(f)) != EOF) {
- nkf_iconv_utf_32(c1, c2, c3, c4);
+ nkf_char c5, c6, c7, c8;
+ if (nkf_iconv_utf_32(c1, c2, c3, c4) == (size_t)NKF_ICONV_WAIT_COMBINING_CHAR) {
+ if ((c5 = (*i_getc)(f)) != EOF &&
+ (c6 = (*i_getc)(f)) != EOF &&
+ (c7 = (*i_getc)(f)) != EOF &&
+ (c8 = (*i_getc)(f)) != EOF) {
+ if (nkf_iconv_utf_32_combine(c1, c2, c3, c4, c5, c6, c7, c8)) {
+ (*i_ungetc)(c8, f);
+ (*i_ungetc)(c7, f);
+ (*i_ungetc)(c6, f);
+ (*i_ungetc)(c5, f);
+ nkf_iconv_utf_32_nocombine(c1, c2, c3, c4);
+ }
+ } else {
+ nkf_iconv_utf_32_nocombine(c1, c2, c3, c4);
+ }
+ }
}
- (*i_ungetc)(EOF, f);
+ goto finished;
}
else if (iconv == w_iconv16) {
while ((c1 = (*i_getc)(f)) != EOF &&
(c2 = (*i_getc)(f)) != EOF) {
- if (nkf_iconv_utf_16(c1, c2, 0, 0) == -2 &&
+ size_t ret = nkf_iconv_utf_16(c1, c2, 0, 0);
+ if (ret == NKF_ICONV_NEED_TWO_MORE_BYTES &&
(c3 = (*i_getc)(f)) != EOF &&
(c4 = (*i_getc)(f)) != EOF) {
nkf_iconv_utf_16(c1, c2, c3, c4);
+ } else if (ret == (size_t)NKF_ICONV_WAIT_COMBINING_CHAR) {
+ if ((c3 = (*i_getc)(f)) != EOF &&
+ (c4 = (*i_getc)(f)) != EOF) {
+ if (nkf_iconv_utf_16_combine(c1, c2, c3, c4)) {
+ (*i_ungetc)(c4, f);
+ (*i_ungetc)(c3, f);
+ nkf_iconv_utf_16_nocombine(c1, c2);
+ }
+ } else {
+ nkf_iconv_utf_16_nocombine(c1, c2);
+ }
}
}
- (*i_ungetc)(EOF, f);
+ goto finished;
}
#endif
code_status(c1);
if (c2) {
/* second byte */
- if (c2 > DEL) {
+ if (c2 > ((input_encoding && nkf_enc_cp5022x_p(input_encoding)) ? 0x92 : DEL)) {
/* in case of 8th bit is on */
if (!estab_f&&!mime_decode_mode) {
/* in case of not established yet */
- /* It is still ambiguious */
+ /* It is still ambiguous */
if (h_conv(f, c2, c1)==EOF) {
LAST;
}
else if (c1 == 'I') {
/* JIS X 0201 Katakana */
set_input_mode(JIS_X_0201_1976_K);
+ shift_mode = 1;
SKIP;
}
else if (c1 == 'B' || c1 == 'J' || c1 == 'H') {
}
}
else {
+ i_ungetc(c1,f);
/* lonely ESC */
(*oconv)(0, ESC);
- SEND;
+ SKIP;
}
} else if (c1 == ESC && iconv == s_iconv) {
/* ESC in Shift_JIS */
}
}
else {
+ i_ungetc(c1,f);
/* lonely ESC */
(*oconv)(0, ESC);
- SEND;
+ SKIP;
}
} else if (c1 == LF || c1 == CR) {
if (broken_f&4) {
}
}
break;
+ case -3:
+ /* 4 bytes UTF-8 (check combining character) */
+ if ((c3 = (*i_getc)(f)) != EOF) {
+ if ((c4 = (*i_getc)(f)) != EOF) {
+ if (w_iconv_combine(c2, c1, 0, c3, c4, 0)) {
+ (*i_ungetc)(c4, f);
+ (*i_ungetc)(c3, f);
+ w_iconv_nocombine(c2, c1, 0);
+ }
+ } else {
+ (*i_ungetc)(c3, f);
+ w_iconv_nocombine(c2, c1, 0);
+ }
+ } else {
+ w_iconv_nocombine(c2, c1, 0);
+ }
+ break;
case -1:
/* 3 bytes EUC or UTF-8 */
if ((c3 = (*i_getc)(f)) != EOF) {
code_status(c3);
- (*iconv)(c2, c1, c3);
+ if ((*iconv)(c2, c1, c3) == -3) {
+ /* 6 bytes UTF-8 (check combining character) */
+ nkf_char c5, c6;
+ if ((c4 = (*i_getc)(f)) != EOF) {
+ if ((c5 = (*i_getc)(f)) != EOF) {
+ if ((c6 = (*i_getc)(f)) != EOF) {
+ if (w_iconv_combine(c2, c1, c3, c4, c5, c6)) {
+ (*i_ungetc)(c6, f);
+ (*i_ungetc)(c5, f);
+ (*i_ungetc)(c4, f);
+ w_iconv_nocombine(c2, c1, c3);
+ }
+ } else {
+ (*i_ungetc)(c5, f);
+ (*i_ungetc)(c4, f);
+ w_iconv_nocombine(c2, c1, c3);
+ }
+ } else {
+ (*i_ungetc)(c4, f);
+ w_iconv_nocombine(c2, c1, c3);
+ }
+ } else {
+ w_iconv_nocombine(c2, c1, c3);
+ }
+ }
}
break;
}
/* goto next_word */
}
+finished:
/* epilogue */
(*iconv)(EOF, 0, 0);
if (!input_codename)
/*
* int options(unsigned char *cp)
- *
+ *
* return values:
* 0: success
* -1: ArgumentError
option_mode = 1;
return 0;
}
- for (i=0;i<sizeof(long_option)/sizeof(long_option[0]);i++) {
+ for (i=0;i<(int)(sizeof(long_option)/sizeof(long_option[0]));i++) {
p = (unsigned char *)long_option[i].name;
for (j=0;*p && *p != '=' && *p == cp[j];p++, j++);
if (*p == cp[j] || cp[j] == SP){
break;
#endif
#ifdef UTF8_OUTPUT_ENABLE
- case 'w': /* UTF-8 output */
+ case 'w': /* UTF-{8,16,32} output */
if (cp[0] == '8') {
cp++;
if (cp[0] == '0'){
if (cp[0]=='L') {
cp++;
output_endian = ENDIAN_LITTLE;
+ output_bom_f = TRUE;
} else if (cp[0] == 'B') {
cp++;
+ output_bom_f = TRUE;
}
if (cp[0] == '0'){
+ output_bom_f = FALSE;
cp++;
enc_idx = enc_idx == UTF_16
? (output_endian == ENDIAN_LITTLE ? UTF_16LE : UTF_16BE)
: (output_endian == ENDIAN_LITTLE ? UTF_32LE : UTF_32BE);
} else {
- output_bom_f = TRUE;
enc_idx = enc_idx == UTF_16
? (output_endian == ENDIAN_LITTLE ? UTF_16LE_BOM : UTF_16BE_BOM)
: (output_endian == ENDIAN_LITTLE ? UTF_32LE_BOM : UTF_32BE_BOM);
continue;
#endif
case SP:
- /* module muliple options in a string are allowed for Perl moudle */
+ /* module multiple options in a string are allowed for Perl module */
while(*cp && *cp++!='-');
continue;
default: