X-Git-Url: http://git.sourceforge.jp/view?p=nkf%2Fnkf.git;a=blobdiff_plain;f=nkf.c;h=5f52748e2b05a59e0eb90c81206250eec799f405;hp=d1d23ebcff58ce9f16aeaf1d330d620231d9654d;hb=caa533415c6aea5db29b4fa89e67718c7bdf2f54;hpb=5912ff367d97d60339a5e64e91496c2ec4efacc7

diff --git a/nkf.c b/nkf.c
index d1d23eb..5f52748 100644
--- a/nkf.c
+++ b/nkf.c
@@ -21,7 +21,7 @@
  * 3. This notice may not be removed or altered from any source distribution.
  */
 #define NKF_VERSION "2.0.9"
-#define NKF_RELEASE_DATE "2009-01-20"
+#define NKF_RELEASE_DATE "2009-04-26"
 #define COPY_RIGHT \
     "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa).\n" \
     "Copyright (C) 1996-2009, The nkf Project."
@@ -295,7 +295,7 @@ struct {
 			       && (c != '(') && (c != ')') && (c != '.') && (c != 0x22)))
 
 #define is_ibmext_in_sjis(c2) (CP932_TABLE_BEGIN <= c2 && c2 <= CP932_TABLE_END)
-#define nkf_byte_jisx0201_katakana_p(c) (SP <= c && c < (0xE0&0x7F))
+#define nkf_byte_jisx0201_katakana_p(c) (SP <= c && c <= 0x5F)
 
 #define         HOLD_SIZE       1024
 #if defined(INT_IS_SHORT)
@@ -468,6 +468,8 @@ struct input_code input_code_list[] = {
     {"Shift_JIS", 0, 0, 0, {0, 0, 0}, s_status, s_iconv, 0},
 #ifdef UTF8_INPUT_ENABLE
     {"UTF-8",     0, 0, 0, {0, 0, 0}, w_status, w_iconv, 0},
+    {"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}
 };
@@ -802,7 +804,7 @@ nkf_default_encoding()
 typedef struct {
     long capa;
     long len;
-    unsigned char *ptr;
+    nkf_char *ptr;
 } nkf_buf_t;
 
 static nkf_buf_t *
@@ -841,7 +843,7 @@ nkf_buf_clear(nkf_buf_t *buf)
 }
 
 static void
-nkf_buf_push(nkf_buf_t *buf, unsigned char c)
+nkf_buf_push(nkf_buf_t *buf, nkf_char c)
 {
     if (buf->capa <= buf->len) {
 	exit(EXIT_FAILURE);
@@ -873,17 +875,20 @@ usage(void)
 {
     fprintf(HELP_OUTPUT,
 	    "Usage:  nkf -[flags] [--] [in file] .. [out file for -O flag]\n"
-	    " j,s,e,w  Output code is ISO-2022-JP, Shift JIS, EUC-JP, UTF-8N\n"
 #ifdef UTF8_OUTPUT_ENABLE
-	    "          After 'w' you can add more options. -w[ 8 [0], 16 [[BL] [0]] ]\n"
+	    " j/s/e/w  Specify output encoding ISO-2022-JP, Shift_JIS, EUC-JP\n"
+	    "          UTF options is -w[8[0],{16,32}[{B,L}[0]]]\n"
+#else
 #endif
-	    " J,S,E,W  Input assumption is JIS 7 bit , Shift JIS, EUC-JP, UTF-8\n"
 #ifdef UTF8_INPUT_ENABLE
-	    "          After 'W' you can add more options. -W[ 8, 16 [BL] ] \n"
+	    " J/S/E/W  Specify input encoding ISO-2022-JP, Shift_JIS, EUC-JP\n"
+	    "          UTF option is -W[8,[16,32][B,L]]\n"
+#else
+	    " J/S/E    Specify output encoding ISO-2022-JP, Shift_JIS, EUC-JP\n"
 #endif
 	    );
     fprintf(HELP_OUTPUT,
-	    " m[BQSN0] MIME decode [B:base64,Q:quoted,S:strict,N:non-strict,0:no decode]\n"
+	    " m[BQSN0] MIME decode [B:base64,Q:quoted,S:strict,N:nonstrict,0:no decode]\n"
 	    " M[BQ]    MIME encode [B:base64 Q:quoted]\n"
 	    " f/F      Folding: -f60 or -f or -f60-10 (fold margin 10) F preserve nl\n"
 	    );
@@ -898,32 +903,31 @@ usage(void)
 	    " L[uwm]   Line mode u:LF w:CRLF m:CR (DEFAULT noconversion)\n"
 	    );
     fprintf(HELP_OUTPUT,
-	    "Long name options\n"
-	    " --ic=<input codeset>  --oc=<output codeset>\n"
-	    "                   Specify the input or output codeset\n"
-	    " --hiragana  --katakana  --katakana-hiragana\n"
-	    "                   To Hiragana/Katakana Conversion\n"
+	    " --ic=<encoding>        Specify the input encoding\n"
+	    " --oc=<encoding>        Specify the output encoding\n"
+	    " --hiragana --katakana  Hiragana/Katakana Conversion\n"
+	    " --katakana-hiragana    Converts each other\n"
 	    );
     fprintf(HELP_OUTPUT,
 #ifdef INPUT_OPTION
-	    " --cap-input, --url-input  Convert hex after ':' or '%%'\n"
+	    " --{cap, url}-input     Convert hex after ':' or '%%'\n"
 #endif
 #ifdef NUMCHAR_OPTION
-	    " --numchar-input   Convert Unicode Character Reference\n"
+	    " --numchar-input        Convert Unicode Character Reference\n"
 #endif
 #ifdef UTF8_INPUT_ENABLE
 	    " --fb-{skip, html, xml, perl, java, subchar}\n"
-	    "                   Specify how nkf handles unassigned characters\n"
+	    "                        Specify unassigned character's replacement\n"
 #endif
 	    );
     fprintf(HELP_OUTPUT,
 #ifdef OVERWRITE
-	    " --in-place[=SUF]  Overwrite original listed files by filtered result\n"
-	    " --overwrite[=SUF] in-place and preserve timestamp of original files\n"
+	    " --in-place[=SUF]       Overwrite original files\n"
+	    " --overwrite[=SUF]      Preserve timestamp of original files\n"
 #endif
-	    " -g --guess        Guess the input code\n"
-	    " -v --version      print the version\n"
-	    " --help/-V         print this help / configuration\n"
+	    " -g --guess             Guess the input code\n"
+	    " -v --version           Print the version\n"
+	    " --help/-V              Print this help / configuration\n"
 	    );
     version();
 }
@@ -1424,6 +1428,7 @@ set_output_encoding(nkf_encoding *enc)
 	output_endian = ENDIAN_LITTLE;
 	output_bom_f = TRUE;
 	break;
+    case UTF_32:
     case UTF_32BE_BOM:
 	output_bom_f = TRUE;
 	break;
@@ -1650,7 +1655,7 @@ nkf_unicode_to_utf8(nkf_char val, nkf_char *p1, nkf_char *p2, nkf_char *p3, nkf_
 	*p3 = 0x80 | ( val        & 0x3f);
 	*p4 = 0;
     } else if (nkf_char_unicode_value_p(val)) {
-	*p1 = 0xe0 |  (val >> 16);
+	*p1 = 0xf0 |  (val >> 18);
 	*p2 = 0x80 | ((val >> 12) & 0x3f);
 	*p3 = 0x80 | ((val >>  6) & 0x3f);
 	*p4 = 0x80 | ( val        & 0x3f);
@@ -3126,18 +3131,16 @@ h_conv(FILE *f, int c1, int c2)
 	    } else if ((c3 = (*i_getc)(f)) == EOF) {
 		ret = EOF;
 		break;
-	    } else {
-		code_status(c3);
-		if (hold_index < hold_count){
-		    c4 = hold_buf[hold_index++];
-		} else if ((c4 = (*i_getc)(f)) == EOF) {
-		    c3 = ret = EOF;
-		    break;
-		} else {
-		    code_status(c4);
-		    (*iconv)(c1, c2, (c3<<8)|c4);
-		}
 	    }
+	    code_status(c3);
+	    if (hold_index < hold_count){
+		c4 = hold_buf[hold_index++];
+	    } else if ((c4 = (*i_getc)(f)) == EOF) {
+		c3 = ret = EOF;
+		break;
+	    }
+	    code_status(c4);
+	    (*iconv)(c1, c2, (c3<<8)|c4);
 	    break;
 	case -1:
 	    /* 3 bytes EUC or UTF-8 */
@@ -4263,7 +4266,7 @@ nfc_getc(FILE *f)
 
     if (c == EOF || c > 0xFF || (c & 0xc0) == 0x80) return c;
 
-    nkf_buf_push(buf, (unsigned char)c);
+    nkf_buf_push(buf, c);
     do {
 	while (lower <= upper) {
 	    int mid = (lower+upper) / 2;
@@ -5442,6 +5445,12 @@ kanji_convert(FILE *f)
 	    if (input_mode == JIS_X_0208 && DEL <= c1 && c1 < 0x92) {
 		/* CP5022x */
 		MORE;
+	    }else if (input_codename && input_codename[0] == 'I' &&
+		    0xA1 <= c1 && c1 <= 0xDF) {
+		/* JIS X 0201 Katakana in 8bit JIS */
+		c2 = JIS_X_0201_1976_K;
+		c1 &= 0x7f;
+		SEND;
 	    } else if (c1 > DEL) {
 		/* 8 bit code */
 		if (!estab_f && !iso8859_f) {
@@ -5842,10 +5851,12 @@ options(unsigned char *cp)
 		cp_back = cp;
 		cp = (unsigned char *)long_option[i].alias;
 	    }else{
+#ifndef PERL_XS
 		if (strcmp(long_option[i].name, "help") == 0){
 		    usage();
 		    exit(EXIT_SUCCESS);
 		}
+#endif
 		if (strcmp(long_option[i].name, "ic=") == 0){
 		    enc = nkf_enc_find((char *)p);
 		    if (!enc) continue;
@@ -6155,9 +6166,6 @@ options(unsigned char *cp)
 		    output_endian = ENDIAN_LITTLE;
 		} else if (cp[0] == 'B') {
 		    cp++;
-		} else {
-		    output_encoding = nkf_enc_from_index(enc_idx);
-		    continue;
 		}
 		if (cp[0] == '0'){
 		    cp++;
@@ -6228,6 +6236,7 @@ options(unsigned char *cp)
 	    while ('0'<= *cp && *cp <='9') {
 		alpha_f |= 1 << (*cp++ - '0');
 	    }
+	    if (alpha_f & ((1 << 2) | (1 << 3))) alpha_f |= 1;
 	    if (!alpha_f) alpha_f = 1;
 	    continue;
 	case 'x':   /* Convert X0201 kana to X0208 or X0201 Conversion */