1 /** Network Kanji Filter. (PDS Version)
2 ************************************************************************
3 ** Copyright (C) 1987, Fujitsu LTD. (Itaru ICHIKAWA)
4 **
\e$BO"Mm@h!'
\e(B
\e$B!J3t!KIY;NDL8&5f=j!!%=%U%H#38&!!;T@n!!;j
\e(B
5 **
\e$B!J
\e(BE-Mail Address: ichikawa@flab.fujitsu.co.jp
\e$B!K
\e(B
6 ** Copyright (C) 1996,1998
8 **
\e$BO"Mm@h!'
\e(B
\e$BN05eBg3X>pJs9)3X2J
\e(B
\e$B2OLn
\e(B
\e$B??<#
\e(B mime/X0208 support
9 **
\e$B!J
\e(BE-Mail Address: kono@ie.u-ryukyu.ac.jp
\e$B!K
\e(B
10 **
\e$BO"Mm@h!'
\e(B COW for DOS & Win16 & Win32 & OS/2
11 **
\e$B!J
\e(BE-Mail Address: GHG00637@niftyserve.or.p
\e$B!K
\e(B
13 **
\e$B$3$N%=!<%9$N$$$+$J$kJ#<L!$2~JQ!$=$@5$b5vBz$7$^$9!#$?$@$7!"
\e(B
14 **
\e$B$=$N:]$K$O!"C/$,9W8%$7$?$r<($9$3$NItJ,$r;D$9$3$H!#
\e(B
15 **
\e$B:FG[I[$d;(;o$NIUO?$J$I$NLd$$9g$o$;$bI,MW$"$j$^$;$s!#
\e(B
16 **
\e$B1DMxMxMQ$b>e5-$KH?$7$J$$HO0O$G5v2D$7$^$9!#
\e(B
17 **
\e$B%P%$%J%j$NG[I[$N:]$K$O
\e(Bversion message
\e$B$rJ]B8$9$k$3$H$r>r7o$H$7$^$9!#
\e(B
18 **
\e$B$3$N%W%m%0%i%`$K$D$$$F$OFC$K2?$NJ]>Z$b$7$J$$!"0-$7$+$i$:!#
\e(B
20 ** Everyone is permitted to do anything on this program
21 ** including copying, modifying, improving,
22 ** as long as you don't try to pretend that you wrote it.
23 ** i.e., the above copyright notice has to appear in all copies.
24 ** Binary distribution requires original version messages.
25 ** You don't have to ask before copying, redistribution or publishing.
26 ** THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE.
27 ***********************************************************************/
29 /***********************************************************************
30 ** UTF-8
\e$B%5%]!<%H$K$D$$$F
\e(B
31 **
\e$B=>Mh$N
\e(B nkf
\e$B$HF~$l$+$($F$=$N$^$^;H$($k$h$&$K$J$C$F$$$^$9
\e(B
32 ** nkf -e
\e$B$J$I$H$7$F5/F0$9$k$H!"<+F0H=JL$G
\e(B UTF-8
\e$B$HH=Dj$5$l$l$P!"
\e(B
33 **
\e$B$=$N$^$^
\e(B euc-jp
\e$B$KJQ49$5$l$^$9
\e(B
35 **
\e$B$^$@%P%0$,$"$k2DG=@-$,9b$$$G$9!#
\e(B
36 ** (
\e$BFC$K<+F0H=JL!"%3!<%I:.:_!"%(%i!<=hM}7O
\e(B)
38 **
\e$B2?$+LdBj$r8+$D$1$?$i!"
\e(B
39 ** E-Mail: furukawa@tcp-ip.or.jp
40 **
\e$B$^$G8fO"Mm$r$*4j$$$7$^$9!#
\e(B
41 ***********************************************************************/
42 /* $Id: nkf.c,v 1.48 2004/12/18 10:33:53 rei_furukawa Exp $ */
43 #define NKF_VERSION "2.0.4"
44 #define NKF_RELEASE_DATE "2004-12-01"
47 static char *CopyRight =
48 "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa),2000 S. Kono, COW, 2002-2004 Kono, Furukawa";
55 ** USAGE: nkf [flags] [file]
58 ** b Output is buffered (DEFAULT)
59 ** u Output is unbuffered
63 ** j Outout code is JIS 7 bit (DEFAULT SELECT)
64 ** s Output code is MS Kanji (DEFAULT SELECT)
65 ** e Output code is AT&T JIS (DEFAULT SELECT)
66 ** w Output code is AT&T JIS (DEFAULT SELECT)
67 ** l Output code is JIS 7bit and ISO8859-1 Latin-1
69 ** m MIME conversion for ISO-2022-JP
70 ** I Convert non ISO-2022-JP charactor to GETA by Pekoe <pekoe@lair.net>
71 ** i_ Output sequence to designate JIS-kanji (DEFAULT_J)
72 ** o_ Output sequence to designate single-byte roman characters (DEFAULT_R)
73 ** M MIME output conversion
75 ** r {de/en}crypt ROT13/47
79 ** T Text mode output (for MS-DOS)
81 ** x Do not convert X0201 kana into X0208
82 ** Z Convert X0208 alphabet to ASCII
87 ** B try to fix broken JIS, missing Escape
88 ** B[1-9] broken level
90 ** O Output to 'nkf.out' file or last file name
91 ** d Delete \r in line feed
92 ** c Add \r in line feed
93 ** -- other long option
94 ** -- ignore following option (don't use with -O )
98 #if (defined(__TURBOC__) || defined(_MSC_VER) || defined(LSI_C) || defined(__MINGW32__)) && !defined(MSDOS)
100 #if (defined(__Win32__) || defined(_WIN32)) && !defined(__WIN32__)
115 #if defined(MSDOS) || defined(__OS2__)
122 #define setbinmode(fp) fsetbin(fp)
123 #else /* Microsoft C, Turbo C */
124 #define setbinmode(fp) setmode(fileno(fp), O_BINARY)
126 #else /* UNIX,OS/2 */
127 #define setbinmode(fp)
130 #ifdef _IOFBF /* SysV and MSDOS, Windows */
131 #define setvbuffer(fp, buf, size) setvbuf(fp, buf, _IOFBF, size)
133 #define setvbuffer(fp, buf, size) setbuffer(fp, buf, size)
136 /*Borland C++ 4.5 EasyWin*/
137 #if defined(__TURBOC__) && defined(_Windows) && !defined(__WIN32__) /*Easy Win */
146 /* added by satoru@isoternet.org */
148 #include <sys/stat.h>
149 #ifndef MSDOS /* UNIX, OS/2 */
153 #if defined(_MSC_VER) || defined(__MINGW32__) /* VC++, MinGW */
154 #include <sys/utime.h>
155 #elif defined(__TURBOC__) /* BCC */
157 #elif defined(LSI_C) /* LSI C */
169 /* state of output_mode and input_mode
187 /* Input Assumption */
191 #define LATIN1_INPUT 6
193 #define STRICT_MIME 8
198 #define JAPANESE_EUC 10
202 #define UTF8_INPUT 13
203 #define UTF16LE_INPUT 14
204 #define UTF16BE_INPUT 15
224 #define is_alnum(c) \
225 (('a'<=c && c<='z')||('A'<= c && c<='Z')||('0'<=c && c<='9'))
227 #define HOLD_SIZE 1024
228 #define IOBUF_SIZE 16384
230 #define DEFAULT_J 'B'
231 #define DEFAULT_R 'B'
233 #define SJ0162 0x00e1 /* 01 - 62 ku offset */
234 #define SJ6394 0x0161 /* 63 - 94 ku offset */
236 #define RANGE_NUM_MAX 18
241 #if defined( UTF8_OUTPUT_ENABLE ) || defined( UTF8_INPUT_ENABLE )
242 #define sizeof_euc_utf8 94
243 #define sizeof_euc_to_utf8_1byte 94
244 #define sizeof_euc_to_utf8_2bytes 94
245 #define sizeof_utf8_to_euc_C2 64
246 #define sizeof_utf8_to_euc_E5B8 64
247 #define sizeof_utf8_to_euc_2bytes 112
248 #define sizeof_utf8_to_euc_3bytes 112
251 /* MIME preprocessor */
254 #ifdef EASYWIN /*Easy Win */
255 extern POINT _BufferSize;
258 /* function prototype */
260 #ifdef ANSI_C_PROTOTYPE
262 #define STATIC static
274 void (*status_func)PROTO((struct input_code *, int));
275 int (*iconv_func)PROTO((int c2, int c1, int c0));
279 STATIC char *input_codename = "";
281 STATIC int noconvert PROTO((FILE *f));
282 STATIC int kanji_convert PROTO((FILE *f));
283 STATIC int h_conv PROTO((FILE *f,int c2,int c1));
284 STATIC int push_hold_buf PROTO((int c2));
285 STATIC void set_iconv PROTO((int f, int (*iconv_func)()));
286 STATIC int s_iconv PROTO((int c2,int c1,int c0));
287 STATIC int s2e_conv PROTO((int c2, int c1, int *p2, int *p1));
288 STATIC int e_iconv PROTO((int c2,int c1,int c0));
289 #ifdef UTF8_INPUT_ENABLE
290 STATIC int w2e_conv PROTO((int c2,int c1,int c0,int *p2,int *p1));
291 STATIC int w_iconv PROTO((int c2,int c1,int c0));
292 STATIC int w_iconv16 PROTO((int c2,int c1,int c0));
293 STATIC int w_iconv_common PROTO((int c1,int c0,unsigned short **pp,int psize,int *p2,int *p1));
294 STATIC int ww16_conv PROTO((int c2, int c1, int c0));
296 #ifdef UTF8_OUTPUT_ENABLE
297 STATIC int e2w_conv PROTO((int c2,int c1));
298 STATIC void w_oconv PROTO((int c2,int c1));
299 STATIC void w_oconv16 PROTO((int c2,int c1));
301 STATIC void e_oconv PROTO((int c2,int c1));
302 STATIC int e2s_conv PROTO((int c2, int c1, int *p2, int *p1));
303 STATIC void s_oconv PROTO((int c2,int c1));
304 STATIC void j_oconv PROTO((int c2,int c1));
305 STATIC void fold_conv PROTO((int c2,int c1));
306 STATIC void cr_conv PROTO((int c2,int c1));
307 STATIC void z_conv PROTO((int c2,int c1));
308 STATIC void rot_conv PROTO((int c2,int c1));
309 STATIC void hira_conv PROTO((int c2,int c1));
310 STATIC void base64_conv PROTO((int c2,int c1));
311 STATIC void iso2022jp_check_conv PROTO((int c2,int c1));
312 STATIC void no_connection PROTO((int c2,int c1));
313 STATIC int no_connection2 PROTO((int c2,int c1,int c0));
315 STATIC void code_score PROTO((struct input_code *ptr));
316 STATIC void code_status PROTO((int c));
318 STATIC void std_putc PROTO((int c));
319 STATIC int std_getc PROTO((FILE *f));
320 STATIC int std_ungetc PROTO((int c,FILE *f));
322 STATIC int broken_getc PROTO((FILE *f));
323 STATIC int broken_ungetc PROTO((int c,FILE *f));
325 STATIC int mime_begin PROTO((FILE *f));
326 STATIC int mime_getc PROTO((FILE *f));
327 STATIC int mime_ungetc PROTO((int c,FILE *f));
329 STATIC int mime_begin_strict PROTO((FILE *f));
330 STATIC int mime_getc_buf PROTO((FILE *f));
331 STATIC int mime_ungetc_buf PROTO((int c,FILE *f));
332 STATIC int mime_integrity PROTO((FILE *f,unsigned char *p));
334 STATIC int base64decode PROTO((int c));
335 STATIC void mime_putc PROTO((int c));
336 STATIC void open_mime PROTO((int c));
337 STATIC void close_mime PROTO(());
338 STATIC void usage PROTO(());
339 STATIC void version PROTO(());
340 STATIC void options PROTO((unsigned char *c));
342 STATIC void reinit PROTO(());
347 static unsigned char stdibuf[IOBUF_SIZE];
348 static unsigned char stdobuf[IOBUF_SIZE];
349 static unsigned char hold_buf[HOLD_SIZE*2];
350 static int hold_count;
352 /* MIME preprocessor fifo */
354 #define MIME_BUF_SIZE (1024) /* 2^n ring buffer */
355 #define MIME_BUF_MASK (MIME_BUF_SIZE-1)
356 #define Fifo(n) mime_buf[(n)&MIME_BUF_MASK]
357 static unsigned char mime_buf[MIME_BUF_SIZE];
358 static unsigned int mime_top = 0;
359 static unsigned int mime_last = 0; /* decoded */
360 static unsigned int mime_input = 0; /* undecoded */
363 static int unbuf_f = FALSE;
364 static int estab_f = FALSE;
365 static int nop_f = FALSE;
366 static int binmode_f = TRUE; /* binary mode */
367 static int rot_f = FALSE; /* rot14/43 mode */
368 static int hira_f = FALSE; /* hira/kata henkan */
369 static int input_f = FALSE; /* non fixed input code */
370 static int alpha_f = FALSE; /* convert JIx0208 alphbet to ASCII */
371 static int mime_f = STRICT_MIME; /* convert MIME B base64 or Q */
372 static int mimebuf_f = FALSE; /* MIME buffered input */
373 static int broken_f = FALSE; /* convert ESC-less broken JIS */
374 static int iso8859_f = FALSE; /* ISO8859 through */
375 static int mimeout_f = FALSE; /* base64 mode */
376 #if defined(MSDOS) || defined(__OS2__)
377 static int x0201_f = TRUE; /* Assume JISX0201 kana */
379 static int x0201_f = NO_X0201; /* Assume NO JISX0201 */
381 static int iso2022jp_f = FALSE; /* convert ISO-2022-JP */
382 #ifdef UTF8_OUTPUT_ENABLE
383 static int unicode_bom_f= 0; /* Output Unicode BOM */
384 static int w_oconv16_LE = 0; /* utf-16 little endian */
385 static int ms_ucs_map_f = FALSE; /* Microsoft UCS Mapping Compatible */
389 #ifdef NUMCHAR_OPTION
391 #define CLASS_MASK 0x0f000000
392 #define CLASS_UTF16 0x01000000
396 static int cap_f = FALSE;
397 static int (*i_cgetc)PROTO((FILE *)) = std_getc; /* input of cgetc */
398 static int (*i_cungetc)PROTO((int c ,FILE *f)) = std_ungetc;
399 STATIC int cap_getc PROTO((FILE *f));
400 STATIC int cap_ungetc PROTO((int c,FILE *f));
402 static int url_f = FALSE;
403 static int (*i_ugetc)PROTO((FILE *)) = std_getc; /* input of ugetc */
404 static int (*i_uungetc)PROTO((int c ,FILE *f)) = std_ungetc;
405 STATIC int url_getc PROTO((FILE *f));
406 STATIC int url_ungetc PROTO((int c,FILE *f));
408 static int numchar_f = FALSE;
409 static int (*i_ngetc)PROTO((FILE *)) = std_getc; /* input of ugetc */
410 static int (*i_nungetc)PROTO((int c ,FILE *f)) = std_ungetc;
411 STATIC int numchar_getc PROTO((FILE *f));
412 STATIC int numchar_ungetc PROTO((int c,FILE *f));
416 static int noout_f = FALSE;
417 STATIC void no_putc PROTO((int c));
418 static int debug_f = FALSE;
419 STATIC void debug PROTO((char *str));
422 static int guess_f = FALSE;
423 STATIC void print_guessed_code PROTO((char *filename));
424 STATIC void set_input_codename PROTO((char *codename));
425 static int is_inputcode_mixed = FALSE;
426 static int is_inputcode_set = FALSE;
429 static int exec_f = 0;
432 #ifdef SHIFTJIS_CP932
433 STATIC int cp932_f = TRUE;
434 #define CP932_TABLE_BEGIN (0xfa)
435 #define CP932_TABLE_END (0xfc)
437 STATIC int cp932inv_f = TRUE;
438 #define CP932INV_TABLE_BEGIN (0xed)
439 #define CP932INV_TABLE_END (0xee)
441 /* STATIC int cp932_conv PROTO((int c2, int c1)); */
442 #endif /* SHIFTJIS_CP932 */
445 STATIC int x0212_f = FALSE;
446 static int x0212_shift PROTO((int c));
447 static int x0212_unshift PROTO((int c));
450 STATIC unsigned char prefix_table[256];
452 STATIC void e_status PROTO((struct input_code *, int));
453 STATIC void s_status PROTO((struct input_code *, int));
455 #ifdef UTF8_INPUT_ENABLE
456 STATIC void w_status PROTO((struct input_code *, int));
457 STATIC void w16_status PROTO((struct input_code *, int));
458 static int utf16_mode = UTF16LE_INPUT;
461 struct input_code input_code_list[] = {
462 {"EUC-JP", 0, 0, 0, {0, 0, 0}, e_status, e_iconv, 0},
463 {"Shift_JIS", 0, 0, 0, {0, 0, 0}, s_status, s_iconv, 0},
464 {"UTF-8", 0, 0, 0, {0, 0, 0}, w_status, w_iconv, 0},
465 {"UTF-16", 0, 0, 0, {0, 0, 0}, w16_status, w_iconv16, 0},
469 static int mimeout_mode = 0;
470 static int base64_count = 0;
472 /* X0208 -> ASCII converter */
475 static int f_line = 0; /* chars in line */
476 static int f_prev = 0;
477 static int fold_preserve_f = FALSE; /* preserve new lines */
478 static int fold_f = FALSE;
479 static int fold_len = 0;
482 static unsigned char kanji_intro = DEFAULT_J;
483 static unsigned char ascii_intro = DEFAULT_R;
487 #define FOLD_MARGIN 10
488 #define DEFAULT_FOLD 60
490 static int fold_margin = FOLD_MARGIN;
494 #ifdef DEFAULT_CODE_JIS
495 # define DEFAULT_CONV j_oconv
497 #ifdef DEFAULT_CODE_SJIS
498 # define DEFAULT_CONV s_oconv
500 #ifdef DEFAULT_CODE_EUC
501 # define DEFAULT_CONV e_oconv
503 #ifdef DEFAULT_CODE_UTF8
504 # define DEFAULT_CONV w_oconv
507 /* process default */
508 static void (*output_conv)PROTO((int c2,int c1)) = DEFAULT_CONV;
510 static void (*oconv)PROTO((int c2,int c1)) = no_connection;
511 /* s_iconv or oconv */
512 static int (*iconv)PROTO((int c2,int c1,int c0)) = no_connection2;
514 static void (*o_zconv)PROTO((int c2,int c1)) = no_connection;
515 static void (*o_fconv)PROTO((int c2,int c1)) = no_connection;
516 static void (*o_crconv)PROTO((int c2,int c1)) = no_connection;
517 static void (*o_rot_conv)PROTO((int c2,int c1)) = no_connection;
518 static void (*o_hira_conv)PROTO((int c2,int c1)) = no_connection;
519 static void (*o_base64conv)PROTO((int c2,int c1)) = no_connection;
520 static void (*o_iso2022jp_check_conv)PROTO((int c2,int c1)) = no_connection;
522 /* static redirections */
524 static void (*o_putc)PROTO((int c)) = std_putc;
526 static int (*i_getc)PROTO((FILE *f)) = std_getc; /* general input */
527 static int (*i_ungetc)PROTO((int c,FILE *f)) =std_ungetc;
529 static int (*i_bgetc)PROTO((FILE *)) = std_getc; /* input of mgetc */
530 static int (*i_bungetc)PROTO((int c ,FILE *f)) = std_ungetc;
532 static void (*o_mputc)PROTO((int c)) = std_putc ; /* output of mputc */
534 static int (*i_mgetc)PROTO((FILE *)) = std_getc; /* input of mgetc */
535 static int (*i_mungetc)PROTO((int c ,FILE *f)) = std_ungetc;
537 /* for strict mime */
538 static int (*i_mgetc_buf)PROTO((FILE *)) = std_getc; /* input of mgetc_buf */
539 static int (*i_mungetc_buf)PROTO((int c,FILE *f)) = std_ungetc;
542 static int output_mode = ASCII, /* output kanji mode */
543 input_mode = ASCII, /* input kanji mode */
544 shift_mode = FALSE; /* TRUE shift out, or X0201 */
545 static int mime_decode_mode = FALSE; /* MIME mode B base64, Q hex */
547 /* X0201 / X0208 conversion tables */
549 /* X0201 kana conversion table */
552 unsigned char cv[]= {
553 0x21,0x21,0x21,0x23,0x21,0x56,0x21,0x57,
554 0x21,0x22,0x21,0x26,0x25,0x72,0x25,0x21,
555 0x25,0x23,0x25,0x25,0x25,0x27,0x25,0x29,
556 0x25,0x63,0x25,0x65,0x25,0x67,0x25,0x43,
557 0x21,0x3c,0x25,0x22,0x25,0x24,0x25,0x26,
558 0x25,0x28,0x25,0x2a,0x25,0x2b,0x25,0x2d,
559 0x25,0x2f,0x25,0x31,0x25,0x33,0x25,0x35,
560 0x25,0x37,0x25,0x39,0x25,0x3b,0x25,0x3d,
561 0x25,0x3f,0x25,0x41,0x25,0x44,0x25,0x46,
562 0x25,0x48,0x25,0x4a,0x25,0x4b,0x25,0x4c,
563 0x25,0x4d,0x25,0x4e,0x25,0x4f,0x25,0x52,
564 0x25,0x55,0x25,0x58,0x25,0x5b,0x25,0x5e,
565 0x25,0x5f,0x25,0x60,0x25,0x61,0x25,0x62,
566 0x25,0x64,0x25,0x66,0x25,0x68,0x25,0x69,
567 0x25,0x6a,0x25,0x6b,0x25,0x6c,0x25,0x6d,
568 0x25,0x6f,0x25,0x73,0x21,0x2b,0x21,0x2c,
572 /* X0201 kana conversion table for daguten */
575 unsigned char dv[]= {
576 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
577 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
578 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
579 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
580 0x00,0x00,0x00,0x00,0x00,0x00,0x25,0x74,
581 0x00,0x00,0x00,0x00,0x25,0x2c,0x25,0x2e,
582 0x25,0x30,0x25,0x32,0x25,0x34,0x25,0x36,
583 0x25,0x38,0x25,0x3a,0x25,0x3c,0x25,0x3e,
584 0x25,0x40,0x25,0x42,0x25,0x45,0x25,0x47,
585 0x25,0x49,0x00,0x00,0x00,0x00,0x00,0x00,
586 0x00,0x00,0x00,0x00,0x25,0x50,0x25,0x53,
587 0x25,0x56,0x25,0x59,0x25,0x5c,0x00,0x00,
588 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
589 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
590 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
591 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
594 /* X0201 kana conversion table for han-daguten */
597 unsigned char ev[]= {
598 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
599 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
600 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
601 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
602 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
603 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
604 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
605 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
606 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
607 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
608 0x00,0x00,0x00,0x00,0x25,0x51,0x25,0x54,
609 0x25,0x57,0x25,0x5a,0x25,0x5d,0x00,0x00,
610 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
611 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
612 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
613 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
617 /* X0208 kigou conversion table */
618 /* 0x8140 - 0x819e */
620 unsigned char fv[] = {
622 0x00,0x00,0x00,0x00,0x2c,0x2e,0x00,0x3a,
623 0x3b,0x3f,0x21,0x00,0x00,0x27,0x60,0x00,
624 0x5e,0x00,0x5f,0x00,0x00,0x00,0x00,0x00,
625 0x00,0x00,0x00,0x00,0x00,0x2d,0x00,0x2f,
626 0x5c,0x00,0x00,0x7c,0x00,0x00,0x60,0x27,
627 0x22,0x22,0x28,0x29,0x00,0x00,0x5b,0x5d,
628 0x7b,0x7d,0x3c,0x3e,0x00,0x00,0x00,0x00,
629 0x00,0x00,0x00,0x00,0x2b,0x2d,0x00,0x00,
630 0x00,0x3d,0x00,0x3c,0x3e,0x00,0x00,0x00,
631 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
632 0x24,0x00,0x00,0x25,0x23,0x26,0x2a,0x40,
633 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00
639 static int file_out = FALSE;
641 static int overwrite = FALSE;
644 static int crmode_f = 0; /* CR, NL, CRLF */
645 #ifdef EASYWIN /*Easy Win */
646 static int end_check;
658 char *outfname = NULL;
661 #ifdef EASYWIN /*Easy Win */
662 _BufferSize.y = 400;/*Set Scroll Buffer Size*/
665 for (argc--,argv++; (argc > 0) && **argv == '-'; argc--, argv++) {
666 cp = (unsigned char *)*argv;
671 if (pipe(fds) < 0 || (pid = fork()) < 0){
682 execvp(argv[1], &argv[1]);
696 if(x0201_f == WISH_TRUE)
697 x0201_f = ((!iso2022jp_f)? TRUE : NO_X0201);
699 if (binmode_f == TRUE)
701 if (freopen("","wb",stdout) == NULL)
708 setbuf(stdout, (char *) NULL);
710 setvbuffer(stdout, stdobuf, IOBUF_SIZE);
713 if (binmode_f == TRUE)
715 if (freopen("","rb",stdin) == NULL) return (-1);
719 setvbuffer(stdin, stdibuf, IOBUF_SIZE);
723 kanji_convert(stdin);
724 if (guess_f) print_guessed_code(NULL);
729 if ((fin = fopen((origfname = *argv++), "r")) == NULL) {
738 /* reopen file for stdout */
739 if (file_out == TRUE) {
742 outfname = malloc(strlen(origfname)
743 + strlen(".nkftmpXXXXXX")
749 strcpy(outfname, origfname);
753 for (i = strlen(outfname); i; --i){
754 if (outfname[i - 1] == '/'
755 || outfname[i - 1] == '\\'){
761 strcat(outfname, "ntXXXXXX");
763 fd = open(outfname, O_WRONLY | O_CREAT | O_TRUNC,
766 strcat(outfname, ".nkftmpXXXXXX");
767 fd = mkstemp(outfname);
770 || (fd_backup = dup(fileno(stdout))) < 0
771 || dup2(fd, fileno(stdout)) < 0
782 outfname = "nkf.out";
785 if(freopen(outfname, "w", stdout) == NULL) {
789 if (binmode_f == TRUE) {
791 if (freopen("","wb",stdout) == NULL)
798 if (binmode_f == TRUE)
800 if (freopen("","rb",fin) == NULL)
805 setvbuffer(fin, stdibuf, IOBUF_SIZE);
809 char *filename = NULL;
811 if (nfiles > 1) filename = origfname;
812 if (guess_f) print_guessed_code(filename);
818 #if defined(MSDOS) && !defined(__MINGW32__)
826 if (dup2(fd_backup, fileno(stdout)) < 0){
829 if (stat(origfname, &sb)) {
830 fprintf(stderr, "Can't stat %s\n", origfname);
832 /*
\e$B%Q!<%_%C%7%g%s$rI|85
\e(B */
833 if (chmod(outfname, sb.st_mode)) {
834 fprintf(stderr, "Can't set permission %s\n", outfname);
837 /*
\e$B%?%$%`%9%?%s%W$rI|85
\e(B */
838 #if defined(MSDOS) && !defined(__MINGW32__)
839 tb[0] = tb[1] = sb.st_mtime;
840 if (utime(outfname, tb)) {
841 fprintf(stderr, "Can't set timestamp %s\n", outfname);
844 tb.actime = sb.st_atime;
845 tb.modtime = sb.st_mtime;
846 if (utime(outfname, &tb)) {
847 fprintf(stderr, "Can't set timestamp %s\n", outfname);
851 if (unlink(origfname)){
855 if (rename(outfname, origfname)) {
857 fprintf(stderr, "Can't rename %s to %s\n",
858 outfname, origfname);
866 #ifdef EASYWIN /*Easy Win */
867 if (file_out == FALSE)
868 scanf("%d",&end_check);
871 #else /* for Other OS */
872 if (file_out == TRUE)
902 {"katakana-hiragana","h3"},
909 #ifdef UTF8_OUTPUT_ENABLE
914 #ifdef UTF8_INPUT_ENABLE
916 {"utf16-input", "W16"},
925 #ifdef NUMCHAR_OPTION
926 {"numchar-input", ""},
932 #ifdef SHIFTJIS_CP932
942 static int option_mode = 0;
949 unsigned char *p = NULL;
961 case '-': /* literal options */
962 if (!*cp) { /* ignore the rest of arguments */
966 for (i=0;i<sizeof(long_option)/sizeof(long_option[0]);i++) {
968 p = (unsigned char *)long_option[i].name;
969 for (j=0;*p && (*p != '=') && *p == cp[j];p++, j++);
970 if (!*p || *p == cp[j]){
977 cp = (unsigned char *)long_option[i].alias;
980 if (strcmp(long_option[i].name, "overwrite") == 0){
987 if (strcmp(long_option[i].name, "cap-input") == 0){
991 if (strcmp(long_option[i].name, "url-input") == 0){
996 #ifdef NUMCHAR_OPTION
997 if (strcmp(long_option[i].name, "numchar-input") == 0){
1003 if (strcmp(long_option[i].name, "no-output") == 0){
1007 if (strcmp(long_option[i].name, "debug") == 0){
1012 if (strcmp(long_option[i].name, "cp932") == 0){
1013 #ifdef SHIFTJIS_CP932
1017 #ifdef UTF8_OUTPUT_ENABLE
1018 ms_ucs_map_f = TRUE;
1022 if (strcmp(long_option[i].name, "no-cp932") == 0){
1023 #ifdef SHIFTJIS_CP932
1027 #ifdef UTF8_OUTPUT_ENABLE
1028 ms_ucs_map_f = FALSE;
1032 #ifdef SHIFTJIS_CP932
1033 if (strcmp(long_option[i].name, "cp932inv") == 0){
1040 if (strcmp(long_option[i].name, "x0212") == 0){
1047 if (strcmp(long_option[i].name, "exec-in") == 0){
1051 if (strcmp(long_option[i].name, "exec-out") == 0){
1056 #ifdef UTF8_OUTPUT_ENABLE
1057 if (strcmp(long_option[i].name, "ms-ucs-map") == 0){
1058 ms_ucs_map_f = TRUE;
1062 if (strcmp(long_option[i].name, "prefix=") == 0){
1063 if (*p == '=' && ' ' < p[1] && p[1] < 128){
1064 for (i = 2; ' ' < p[i] && p[i] < 128; i++){
1065 prefix_table[p[i]] = p[1];
1072 case 'b': /* buffered mode */
1075 case 'u': /* non bufferd mode */
1078 case 't': /* transparent mode */
1081 case 'j': /* JIS output */
1083 output_conv = j_oconv;
1085 case 'e': /* AT&T EUC output */
1086 output_conv = e_oconv;
1088 case 's': /* SJIS output */
1089 output_conv = s_oconv;
1091 case 'l': /* ISO8859 Latin-1 support, no conversion */
1092 iso8859_f = TRUE; /* Only compatible with ISO-2022-JP */
1093 input_f = LATIN1_INPUT;
1095 case 'i': /* Kanji IN ESC-$-@/B */
1096 if (*cp=='@'||*cp=='B')
1097 kanji_intro = *cp++;
1099 case 'o': /* ASCII IN ESC-(-J/B */
1100 if (*cp=='J'||*cp=='B'||*cp=='H')
1101 ascii_intro = *cp++;
1108 if ('9'>= *cp && *cp>='0')
1109 hira_f |= (*cp++ -'0');
1116 #if defined(MSDOS) || defined(__OS2__)
1131 #ifdef UTF8_OUTPUT_ENABLE
1132 case 'w': /* UTF-8 output */
1133 if ('1'== cp[0] && '6'==cp[1]) {
1134 output_conv = w_oconv16; cp+=2;
1136 unicode_bom_f=2; cp++;
1139 unicode_bom_f=1; cp++;
1141 } else if (cp[0] == 'B') {
1142 unicode_bom_f=2; cp++;
1144 unicode_bom_f=1; cp++;
1147 } else if (cp[0] == '8') {
1148 output_conv = w_oconv; cp++;
1151 unicode_bom_f=1; cp++;
1154 output_conv = w_oconv;
1157 #ifdef UTF8_INPUT_ENABLE
1158 case 'W': /* UTF-8 input */
1159 if ('1'== cp[0] && '6'==cp[1]) {
1160 input_f = UTF16LE_INPUT;
1163 } else if (cp[0] == 'B') {
1165 input_f = UTF16BE_INPUT;
1167 } else if (cp[0] == '8') {
1169 input_f = UTF8_INPUT;
1171 input_f = UTF8_INPUT;
1174 /* Input code assumption */
1175 case 'J': /* JIS input */
1176 case 'E': /* AT&T EUC input */
1177 input_f = JIS_INPUT;
1179 case 'S': /* MS Kanji input */
1180 input_f = SJIS_INPUT;
1181 if (x0201_f==NO_X0201) x0201_f=TRUE;
1183 case 'Z': /* Convert X0208 alphabet to asii */
1184 /* bit:0 Convert X0208
1185 bit:1 Convert Kankaku to one space
1186 bit:2 Convert Kankaku to two spaces
1187 bit:3 Convert HTML Entity
1189 if ('9'>= *cp && *cp>='0')
1190 alpha_f |= 1<<(*cp++ -'0');
1194 case 'x': /* Convert X0201 kana to X0208 or X0201 Conversion */
1195 x0201_f = FALSE; /* No X0201->X0208 conversion */
1197 ESC-(-I in JIS, EUC, MS Kanji
1198 SI/SO in JIS, EUC, MS Kanji
1199 SSO in EUC, JIS, not in MS Kanji
1200 MS Kanji (0xa0-0xdf)
1202 ESC-(-I in JIS (0x20-0x5f)
1203 SSO in EUC (0xa0-0xdf)
1204 0xa0-0xd in MS Kanji (0xa0-0xdf)
1207 case 'X': /* Assume X0201 kana */
1208 /* Default value is NO_X0201 for EUC/MS-Kanji mix */
1211 case 'F': /* prserve new lines */
1212 fold_preserve_f = TRUE;
1213 case 'f': /* folding -f60 or -f */
1216 while('0'<= *cp && *cp <='9') { /* we don't use atoi here */
1218 fold_len += *cp++ - '0';
1220 if (!(0<fold_len && fold_len<BUFSIZ))
1221 fold_len = DEFAULT_FOLD;
1225 while('0'<= *cp && *cp <='9') { /* we don't use atoi here */
1227 fold_margin += *cp++ - '0';
1231 case 'm': /* MIME support */
1232 if (*cp=='B'||*cp=='Q') {
1233 mime_decode_mode = *cp++;
1234 mimebuf_f = FIXED_MIME;
1235 } else if (*cp=='N') {
1236 mime_f = TRUE; cp++;
1237 } else if (*cp=='S') {
1238 mime_f = STRICT_MIME; cp++;
1239 } else if (*cp=='0') {
1240 mime_f = FALSE; cp++;
1243 case 'M': /* MIME output */
1246 mimeout_f = FIXED_MIME; cp++;
1247 } else if (*cp=='Q') {
1249 mimeout_f = FIXED_MIME; cp++;
1254 case 'B': /* Broken JIS support */
1256 bit:1 allow any x on ESC-(-x or ESC-$-x
1257 bit:2 reset to ascii on NL
1259 if ('9'>= *cp && *cp>='0')
1260 broken_f |= 1<<(*cp++ -'0');
1265 case 'O':/* for Output file */
1269 case 'c':/* add cr code */
1272 case 'd':/* delete cr code */
1275 case 'I': /* ISO-2022-JP output */
1278 case 'L': /* line mode */
1279 if (*cp=='u') { /* unix */
1280 crmode_f = NL; cp++;
1281 } else if (*cp=='m') { /* mac */
1282 crmode_f = CR; cp++;
1283 } else if (*cp=='w') { /* windows */
1284 crmode_f = CRLF; cp++;
1285 } else if (*cp=='0') { /* no conversion */
1295 /* module muliple options in a string are allowed for Perl moudle */
1296 while(*cp && *cp!='-') cp++;
1300 /* bogus option but ignored */
1306 #ifdef ANSI_C_PROTOTYPE
1307 struct input_code * find_inputcode_byfunc(int (*iconv_func)(int c2,int c1,int c0))
1309 struct input_code * find_inputcode_byfunc(iconv_func)
1310 int (*iconv_func)();
1314 struct input_code *p = input_code_list;
1316 if (iconv_func == p->iconv_func){
1325 #ifdef ANSI_C_PROTOTYPE
1326 void set_iconv(int f, int (*iconv_func)(int c2,int c1,int c0))
1328 void set_iconv(f, iconv_func)
1330 int (*iconv_func)();
1334 static int (*iconv_for_check)() = 0;
1336 #ifdef INPUT_CODE_FIX
1344 #ifdef INPUT_CODE_FIX
1345 && (f == -TRUE || !input_f) /* -TRUE means "FORCE" */
1351 if (estab_f && iconv_for_check != iconv){
1352 struct input_code *p = find_inputcode_byfunc(iconv);
1354 set_input_codename(p->name);
1355 debug(input_codename);
1357 iconv_for_check = iconv;
1362 #define SCORE_L2 (1) /*
\e$BBh
\e(B2
\e$B?e=`4A;z
\e(B */
1363 #define SCORE_KANA (SCORE_L2 << 1) /*
\e$B$$$o$f$kH>3Q%+%J
\e(B */
1364 #define SCORE_DEPEND (SCORE_KANA << 1) /*
\e$B5!<o0MB8J8;z
\e(B */
1365 #ifdef SHIFTJIS_CP932
1366 #define SCORE_CP932 (SCORE_DEPEND << 1) /* CP932
\e$B$K$h$kFI$_49$(
\e(B */
1367 #define SCORE_NO_EXIST (SCORE_CP932 << 1) /*
\e$BB8:_$7$J$$J8;z
\e(B */
1369 #define SCORE_NO_EXIST (SCORE_DEPEND << 1) /*
\e$BB8:_$7$J$$J8;z
\e(B */
1371 #define SCORE_iMIME (SCORE_NO_EXIST << 1) /* MIME
\e$B$K$h$k;XDj
\e(B */
1372 #define SCORE_ERROR (SCORE_iMIME << 1) /*
\e$B%(%i!<
\e(B */
1374 #define SCORE_INIT (SCORE_iMIME)
1376 int score_table_A0[] = {
1379 0, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND,
1380 SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_NO_EXIST,
1383 int score_table_F0[] = {
1384 SCORE_L2, SCORE_L2, SCORE_L2, SCORE_L2,
1385 SCORE_L2, SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST,
1386 SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND,
1387 SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST, SCORE_ERROR,
1390 void set_code_score(ptr, score)
1391 struct input_code *ptr;
1395 ptr->score |= score;
1399 void clr_code_score(ptr, score)
1400 struct input_code *ptr;
1404 ptr->score &= ~score;
1408 void code_score(ptr)
1409 struct input_code *ptr;
1411 int c2 = ptr->buf[0];
1412 int c1 = ptr->buf[1];
1414 set_code_score(ptr, SCORE_ERROR);
1415 }else if (c2 == SSO){
1416 set_code_score(ptr, SCORE_KANA);
1417 #ifdef UTF8_OUTPUT_ENABLE
1418 }else if (!e2w_conv(c2, c1)){
1419 set_code_score(ptr, SCORE_NO_EXIST);
1421 }else if ((c2 & 0x70) == 0x20){
1422 set_code_score(ptr, score_table_A0[c2 & 0x0f]);
1423 }else if ((c2 & 0x70) == 0x70){
1424 set_code_score(ptr, score_table_F0[c2 & 0x0f]);
1425 }else if ((c2 & 0x70) >= 0x50){
1426 set_code_score(ptr, SCORE_L2);
1430 void status_disable(ptr)
1431 struct input_code *ptr;
1436 if (iconv == ptr->iconv_func) set_iconv(FALSE, 0);
1439 void status_push_ch(ptr, c)
1440 struct input_code *ptr;
1443 ptr->buf[ptr->index++] = c;
1446 void status_clear(ptr)
1447 struct input_code *ptr;
1453 void status_reset(ptr)
1454 struct input_code *ptr;
1457 ptr->score = SCORE_INIT;
1460 void status_reinit(ptr)
1461 struct input_code *ptr;
1464 ptr->_file_stat = 0;
1467 void status_check(ptr, c)
1468 struct input_code *ptr;
1471 if (c <= DEL && estab_f){
1476 void s_status(ptr, c)
1477 struct input_code *ptr;
1482 status_check(ptr, c);
1487 #ifdef NUMCHAR_OPTION
1488 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1491 }else if (0xa1 <= c && c <= 0xdf){
1492 status_push_ch(ptr, SSO);
1493 status_push_ch(ptr, c);
1496 }else if ((0x81 <= c && c < 0xa0) || (0xe0 <= c && c <= 0xef)){
1498 status_push_ch(ptr, c);
1499 #ifdef SHIFTJIS_CP932
1501 && CP932_TABLE_BEGIN <= c && c <= CP932_TABLE_END){
1503 status_push_ch(ptr, c);
1504 #endif /* SHIFTJIS_CP932 */
1506 }else if (x0212_f && 0xf0 <= c && c <= 0xfc){
1508 status_push_ch(ptr, c);
1509 #endif /* X0212_ENABLE */
1511 status_disable(ptr);
1515 if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)){
1516 status_push_ch(ptr, c);
1517 s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]);
1521 status_disable(ptr);
1525 #ifdef SHIFTJIS_CP932
1526 if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)){
1527 status_push_ch(ptr, c);
1528 if (s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]) == 0){
1529 set_code_score(ptr, SCORE_CP932);
1534 #endif /* SHIFTJIS_CP932 */
1535 #ifndef X0212_ENABLE
1536 status_disable(ptr);
1542 void e_status(ptr, c)
1543 struct input_code *ptr;
1548 status_check(ptr, c);
1553 #ifdef NUMCHAR_OPTION
1554 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1557 }else if (SSO == c || (0xa1 <= c && c <= 0xfe)){
1559 status_push_ch(ptr, c);
1561 }else if (0x8f == c){
1563 status_push_ch(ptr, c);
1564 #endif /* X0212_ENABLE */
1566 status_disable(ptr);
1570 if (0xa1 <= c && c <= 0xfe){
1571 status_push_ch(ptr, c);
1575 status_disable(ptr);
1580 if (0xa1 <= c && c <= 0xfe){
1582 status_push_ch(ptr, c);
1584 status_disable(ptr);
1586 #endif /* X0212_ENABLE */
1590 #ifdef UTF8_INPUT_ENABLE
1591 void w16_status(ptr, c)
1592 struct input_code *ptr;
1599 if (ptr->_file_stat == 0){
1600 if (c == 0xfe || c == 0xff){
1602 status_push_ch(ptr, c);
1603 ptr->_file_stat = 1;
1605 status_disable(ptr);
1606 ptr->_file_stat = -1;
1608 }else if (ptr->_file_stat > 0){
1610 status_push_ch(ptr, c);
1611 }else if (ptr->_file_stat < 0){
1612 status_disable(ptr);
1618 status_disable(ptr);
1619 ptr->_file_stat = -1;
1621 status_push_ch(ptr, c);
1628 if (ptr->stat != c && (c == 0xfe || c == 0xff)){
1629 status_push_ch(ptr, c);
1632 status_disable(ptr);
1633 ptr->_file_stat = -1;
1639 void w_status(ptr, c)
1640 struct input_code *ptr;
1645 status_check(ptr, c);
1650 #ifdef NUMCHAR_OPTION
1651 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1654 }else if (0xc0 <= c && c <= 0xdf){
1656 status_push_ch(ptr, c);
1657 }else if (0xe0 <= c && c <= 0xef){
1659 status_push_ch(ptr, c);
1661 status_disable(ptr);
1666 if (0x80 <= c && c <= 0xbf){
1667 status_push_ch(ptr, c);
1668 if (ptr->index > ptr->stat){
1669 int bom = (ptr->buf[0] == 0xef && ptr->buf[1] == 0xbb
1670 && ptr->buf[2] == 0xbf);
1671 w2e_conv(ptr->buf[0], ptr->buf[1], ptr->buf[2],
1672 &ptr->buf[0], &ptr->buf[1]);
1679 status_disable(ptr);
1690 int action_flag = 1;
1691 struct input_code *result = 0;
1692 struct input_code *p = input_code_list;
1694 (p->status_func)(p, c);
1697 }else if(p->stat == 0){
1708 if (result && !estab_f){
1709 set_iconv(TRUE, result->iconv_func);
1710 }else if (c <= DEL){
1711 struct input_code *ptr = input_code_list;
1720 #define STD_GC_BUFSIZE (256)
1721 int std_gc_buf[STD_GC_BUFSIZE];
1729 return std_gc_buf[--std_gc_ndx];
1739 if (std_gc_ndx == STD_GC_BUFSIZE){
1742 std_gc_buf[std_gc_ndx++] = c;
1760 while ((c = (*i_getc)(f)) != EOF)
1769 oconv = output_conv;
1772 /* replace continucation module, from output side */
1774 /* output redicrection */
1776 if (noout_f || guess_f){
1783 if (mimeout_f == TRUE) {
1784 o_base64conv = oconv; oconv = base64_conv;
1786 /* base64_count = 0; */
1790 o_crconv = oconv; oconv = cr_conv;
1793 o_rot_conv = oconv; oconv = rot_conv;
1796 o_iso2022jp_check_conv = oconv; oconv = iso2022jp_check_conv;
1799 o_hira_conv = oconv; oconv = hira_conv;
1802 o_fconv = oconv; oconv = fold_conv;
1805 if (alpha_f || x0201_f) {
1806 o_zconv = oconv; oconv = z_conv;
1810 i_ungetc = std_ungetc;
1811 /* input redicrection */
1814 i_cgetc = i_getc; i_getc = cap_getc;
1815 i_cungetc = i_ungetc; i_ungetc= cap_ungetc;
1818 i_ugetc = i_getc; i_getc = url_getc;
1819 i_uungetc = i_ungetc; i_ungetc= url_ungetc;
1822 #ifdef NUMCHAR_OPTION
1824 i_ngetc = i_getc; i_getc = numchar_getc;
1825 i_nungetc = i_ungetc; i_ungetc= numchar_ungetc;
1828 if (mime_f && mimebuf_f==FIXED_MIME) {
1829 i_mgetc = i_getc; i_getc = mime_getc;
1830 i_mungetc = i_ungetc; i_ungetc = mime_ungetc;
1833 i_bgetc = i_getc; i_getc = broken_getc;
1834 i_bungetc = i_ungetc; i_ungetc = broken_ungetc;
1836 if (input_f == JIS_INPUT || input_f == LATIN1_INPUT) {
1837 set_iconv(-TRUE, e_iconv);
1838 } else if (input_f == SJIS_INPUT) {
1839 set_iconv(-TRUE, s_iconv);
1840 #ifdef UTF8_INPUT_ENABLE
1841 } else if (input_f == UTF8_INPUT) {
1842 set_iconv(-TRUE, w_iconv);
1843 } else if (input_f == UTF16LE_INPUT) {
1844 set_iconv(-TRUE, w_iconv16);
1847 set_iconv(FALSE, e_iconv);
1851 struct input_code *p = input_code_list;
1859 Conversion main loop. Code detection only.
1869 module_connection();
1874 output_mode = ASCII;
1877 #define NEXT continue /* no output, get next */
1878 #define SEND ; /* output c1 and c2, get next */
1879 #define LAST break /* end of loop, go closing */
1881 while ((c1 = (*i_getc)(f)) != EOF) {
1886 /* in case of 8th bit is on */
1888 /* in case of not established yet */
1889 /* It is still ambiguious */
1890 if (h_conv(f, c2, c1)==EOF)
1896 /* in case of already established */
1898 /* ignore bogus code */
1904 /* second byte, 7 bit code */
1905 /* it might be kanji shitfted */
1906 if ((c1 == DEL) || (c1 <= SPACE)) {
1907 /* ignore bogus first code */
1915 #ifdef UTF8_INPUT_ENABLE
1924 #ifdef NUMCHAR_OPTION
1925 } else if ((c1 & CLASS_MASK) == CLASS_UTF16){
1928 } else if (c1 > DEL) {
1930 if (!estab_f && !iso8859_f) {
1931 /* not established yet */
1934 } else { /* estab_f==TRUE */
1939 } else if (SSP<=c1 && c1<0xe0 && iconv == s_iconv) {
1940 /* SJIS X0201 Case... */
1941 if(iso2022jp_f && x0201_f==NO_X0201) {
1942 (*oconv)(GETA1, GETA2);
1949 } else if (c1==SSO && iconv != s_iconv) {
1950 /* EUC X0201 Case */
1951 c1 = (*i_getc)(f); /* skip SSO */
1953 if (SSP<=c1 && c1<0xe0) {
1954 if(iso2022jp_f && x0201_f==NO_X0201) {
1955 (*oconv)(GETA1, GETA2);
1962 } else { /* bogus code, skip SSO and one byte */
1966 /* already established */
1971 } else if ((c1 > SPACE) && (c1 != DEL)) {
1972 /* in case of Roman characters */
1974 /* output 1 shifted byte */
1978 } else if (SPACE<=c1 && c1<(0xe0&0x7f) ){
1979 /* output 1 shifted byte */
1980 if(iso2022jp_f && x0201_f==NO_X0201) {
1981 (*oconv)(GETA1, GETA2);
1988 /* look like bogus code */
1991 } else if (input_mode == X0208) {
1992 /* in case of Kanji shifted */
1995 } else if (c1 == '=' && mime_f && !mime_decode_mode ) {
1996 /* Check MIME code */
1997 if ((c1 = (*i_getc)(f)) == EOF) {
2000 } else if (c1 == '?') {
2001 /* =? is mime conversion start sequence */
2002 if(mime_f == STRICT_MIME) {
2003 /* check in real detail */
2004 if (mime_begin_strict(f) == EOF)
2008 } else if (mime_begin(f) == EOF)
2018 /* normal ASCII code */
2021 } else if (c1 == SI) {
2024 } else if (c1 == SO) {
2027 } else if (c1 == ESC ) {
2028 if ((c1 = (*i_getc)(f)) == EOF) {
2029 /* (*oconv)(0, ESC); don't send bogus code */
2031 } else if (c1 == '$') {
2032 if ((c1 = (*i_getc)(f)) == EOF) {
2034 (*oconv)(0, ESC); don't send bogus code
2035 (*oconv)(0, '$'); */
2037 } else if (c1 == '@'|| c1 == 'B') {
2038 /* This is kanji introduction */
2041 set_input_codename("ISO-2022-JP");
2042 debug(input_codename);
2044 } else if (c1 == '(') {
2045 if ((c1 = (*i_getc)(f)) == EOF) {
2046 /* don't send bogus code
2052 } else if (c1 == '@'|| c1 == 'B') {
2053 /* This is kanji introduction */
2058 } else if (c1 == 'D'){
2062 #endif /* X0212_ENABLE */
2064 /* could be some special code */
2071 } else if (broken_f&0x2) {
2072 /* accept any ESC-(-x as broken code ... */
2082 } else if (c1 == '(') {
2083 if ((c1 = (*i_getc)(f)) == EOF) {
2084 /* don't send bogus code
2086 (*oconv)(0, '('); */
2090 /* This is X0201 kana introduction */
2091 input_mode = X0201; shift_mode = X0201;
2093 } else if (c1 == 'B' || c1 == 'J' || c1 == 'H') {
2094 /* This is X0208 kanji introduction */
2095 input_mode = ASCII; shift_mode = FALSE;
2097 } else if (broken_f&0x2) {
2098 input_mode = ASCII; shift_mode = FALSE;
2103 /* maintain various input_mode here */
2107 } else if ( c1 == 'N' || c1 == 'n' ){
2109 c3 = (*i_getc)(f); /* skip SS2 */
2110 if ( (SPACE<=c3 && c3 < 0x60) || (0xa0<=c3 && c3 < 0xe0)){
2125 } else if ((c1 == NL || c1 == CR) && broken_f&4) {
2126 input_mode = ASCII; set_iconv(FALSE, 0);
2129 } else if (c1 == NL && mime_f && !mime_decode_mode ) {
2130 if ((c1=(*i_getc)(f))!=EOF && c1 == SPACE) {
2138 } else if (c1 == CR && mime_f && !mime_decode_mode ) {
2139 if ((c1=(*i_getc)(f))!=EOF) {
2143 } else if (c1 == NL && (c1=(*i_getc)(f))!=EOF && c1 == SPACE) {
2160 if (input_mode == X0208)
2161 (*oconv)(c2, c1); /* this is JIS, not SJIS/EUC case */
2163 else if (input_mode == X0212)
2164 (*oconv)((0x8f << 8) | c2, c1);
2165 #endif /* X0212_ENABLE */
2166 else if (input_mode)
2167 (*oconv)(input_mode, c1); /* other special case */
2168 else if ((*iconv)(c2, c1, 0) < 0){ /* can be EUC/SJIS */
2169 int c0 = (*i_getc)(f);
2172 (*iconv)(c2, c1, c0);
2178 /* goto next_word */
2182 (*iconv)(EOF, 0, 0);
2195 /** it must NOT be in the kanji shifte sequence */
2196 /** it must NOT be written in JIS7 */
2197 /** and it must be after 2 byte 8bit code */
2204 while ((c1 = (*i_getc)(f)) != EOF) {
2210 if (push_hold_buf(c1) == EOF || estab_f){
2216 struct input_code *p = input_code_list;
2217 struct input_code *result = p;
2222 if (p->score < result->score){
2227 set_iconv(FALSE, result->iconv_func);
2232 ** 1) EOF is detected, or
2233 ** 2) Code is established, or
2234 ** 3) Buffer is FULL (but last word is pushed)
2236 ** in 1) and 3) cases, we continue to use
2237 ** Kanji codes by oconv and leave estab_f unchanged.
2242 while (wc < hold_count){
2243 c2 = hold_buf[wc++];
2245 #ifdef NUMCHAR_OPTION
2246 || (c2 & CLASS_MASK) == CLASS_UTF16
2251 }else if (iconv == s_iconv && 0xa1 <= c2 && c2 <= 0xdf){
2252 (*iconv)(X0201, c2, 0);
2255 if (wc < hold_count){
2256 c1 = hold_buf[wc++];
2265 if ((*iconv)(c2, c1, 0) < 0){
2267 if (wc < hold_count){
2268 c0 = hold_buf[wc++];
2277 (*iconv)(c2, c1, c0);
2290 if (hold_count >= HOLD_SIZE*2)
2292 hold_buf[hold_count++] = c2;
2293 return ((hold_count >= HOLD_SIZE*2) ? EOF : hold_count);
2296 int s2e_conv(c2, c1, p2, p1)
2301 #ifdef SHIFTJIS_CP932
2302 if (cp932_f && CP932_TABLE_BEGIN <= c2 && c2 <= CP932_TABLE_END){
2303 extern unsigned short shiftjis_cp932[3][189];
2304 val = shiftjis_cp932[c2 - CP932_TABLE_BEGIN][c1 - 0x40];
2310 #endif /* SHIFTJIS_CP932 */
2312 if (x0212_f && 0xfa <= c2 && c2 <= 0xfc){
2313 extern unsigned short shiftjis_x0212[3][189];
2314 val = shiftjis_x0212[c2 - 0xfa][c1 - 0x40];
2317 c2 = (0x8f << 8) | (val >> 8);
2329 c2 = c2 + c2 - ((c2 <= 0x9f) ? SJ0162 : SJ6394);
2331 c1 = c1 - ((c1 > DEL) ? SPACE : 0x1f);
2339 c2 = x0212_unshift(c2);
2354 } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) {
2357 int ret = s2e_conv(c2, c1, &c2, &c1);
2358 if (ret) return ret;
2372 }else if (c2 == 0x8f){
2376 c2 = (c2 << 8) | (c1 & 0x7f);
2378 #ifdef SHIFTJIS_CP932
2381 if (e2s_conv(c2, c1, &s2, &s1) == 0){
2382 s2e_conv(s2, s1, &c2, &c1);
2383 if ((c2 & 0xff00) == 0){
2389 #endif /* SHIFTJIS_CP932 */
2390 #endif /* X0212_ENABLE */
2391 } else if (c2 == SSO){
2394 } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) {
2404 #ifdef UTF8_INPUT_ENABLE
2406 w2e_conv(c2, c1, c0, p2, p1)
2410 extern unsigned short * utf8_to_euc_2bytes[];
2411 extern unsigned short ** utf8_to_euc_3bytes[];
2414 if (0xc0 <= c2 && c2 <= 0xef) {
2415 unsigned short **pp;
2418 if (c0 == 0) return -1;
2419 pp = utf8_to_euc_3bytes[c2 - 0x80];
2420 ret = w_iconv_common(c1, c0, pp, sizeof_utf8_to_euc_C2, p2, p1);
2422 ret = w_iconv_common(c2, c1, utf8_to_euc_2bytes, sizeof_utf8_to_euc_2bytes, p2, p1);
2424 #ifdef NUMCHAR_OPTION
2427 if (p1) *p1 = CLASS_UTF16 | ww16_conv(c2, c1, c0);
2432 } else if (c2 == X0201) {
2445 int ret = w2e_conv(c2, c1, c0, &c2, &c1);
2453 w16w_conv(val, p2, p1, p0)
2461 }else if (val < 0x800){
2462 *p2 = 0xc0 | (val >> 6);
2463 *p1 = 0x80 | (val & 0x3f);
2466 *p2 = 0xe0 | (val >> 12);
2467 *p1 = 0x80 | ((val >> 6) & 0x3f);
2468 *p0 = 0x80 | (val & 0x3f);
2473 ww16_conv(c2, c1, c0)
2478 val = (c2 & 0x0f) << 12;
2479 val |= (c1 & 0x3f) << 6;
2481 }else if (c2 >= 0xc0){
2482 val = (c2 & 0x1f) << 6;
2491 w16e_conv(val, p2, p1)
2495 extern unsigned short * utf8_to_euc_2bytes[];
2496 extern unsigned short ** utf8_to_euc_3bytes[];
2498 unsigned short **pp;
2502 w16w_conv(val, &c2, &c1, &c0);
2505 pp = utf8_to_euc_3bytes[c2 - 0x80];
2506 psize = sizeof_utf8_to_euc_C2;
2507 ret = w_iconv_common(c1, c0, pp, psize, p2, p1);
2509 pp = utf8_to_euc_2bytes;
2510 psize = sizeof_utf8_to_euc_2bytes;
2511 ret = w_iconv_common(c2, c1, pp, psize, p2, p1);
2513 #ifdef NUMCHAR_OPTION
2516 *p1 = CLASS_UTF16 | val;
2528 w_iconv16(c2, c1, c0)
2533 if (c2==0376 && c1==0377){
2534 utf16_mode = UTF16LE_INPUT;
2536 } else if (c2==0377 && c1==0376){
2537 utf16_mode = UTF16BE_INPUT;
2540 if (c2 != EOF && utf16_mode == UTF16BE_INPUT) {
2542 tmp=c1; c1=c2; c2=tmp;
2544 if ((c2==0 && c1 < 0x80) || c2==EOF) {
2548 ret = w16e_conv(((c2<<8)&0xff00) + c1, &c2, &c1);
2549 if (ret) return ret;
2555 w_iconv_common(c1, c0, pp, psize, p2, p1)
2557 unsigned short **pp;
2565 if (pp == 0) return 1;
2568 if (c1 < 0 || psize <= c1) return 1;
2570 if (p == 0) return 1;
2573 if (c0 < 0 || sizeof_utf8_to_euc_E5B8 <= c0) return 1;
2575 if (val == 0) return 1;
2582 if (c2 == SO) c2 = X0201;
2591 #ifdef UTF8_OUTPUT_ENABLE
2596 extern unsigned short euc_to_utf8_1byte[];
2597 extern unsigned short * euc_to_utf8_2bytes[];
2598 extern unsigned short * euc_to_utf8_2bytes_ms[];
2602 p = euc_to_utf8_1byte;
2604 } else if (c2 >> 8 == 0x8f){
2605 extern unsigned short * x0212_to_utf8_2bytes[];
2606 c2 = (c2&0x7f) - 0x21;
2607 if (0<=c2 && c2<sizeof_euc_to_utf8_2bytes)
2608 p = x0212_to_utf8_2bytes[c2];
2614 c2 = (c2&0x7f) - 0x21;
2615 if (0<=c2 && c2<sizeof_euc_to_utf8_2bytes)
2616 p = ms_ucs_map_f ? euc_to_utf8_2bytes_ms[c2] : euc_to_utf8_2bytes[c2];
2621 c1 = (c1 & 0x7f) - 0x21;
2622 if (0<=c1 && c1<sizeof_euc_to_utf8_1byte)
2638 if (unicode_bom_f==2) {
2645 #ifdef NUMCHAR_OPTION
2646 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2647 w16w_conv(c1, &c2, &c1, &c0);
2651 if (c0) (*o_putc)(c0);
2658 output_mode = ASCII;
2660 } else if (c2 == ISO8859_1) {
2661 output_mode = ISO8859_1;
2662 (*o_putc)(c1 | 0x080);
2666 val = e2w_conv(c2, c1);
2668 w16w_conv(val, &c2, &c1, &c0);
2672 if (c0) (*o_putc)(c0);
2688 if (unicode_bom_f==2) {
2690 (*o_putc)((unsigned char)'\377');
2694 (*o_putc)((unsigned char)'\377');
2699 if (c2 == ISO8859_1) {
2702 #ifdef NUMCHAR_OPTION
2703 } else if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16) {
2704 c2 = (c1 >> 8) & 0xff;
2708 unsigned short val = e2w_conv(c2, c1);
2709 c2 = (val >> 8) & 0xff;
2728 #ifdef NUMCHAR_OPTION
2729 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2730 w16e_conv(c1, &c2, &c1);
2731 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2739 } else if (c2 == 0) {
2740 output_mode = ASCII;
2742 } else if (c2 == X0201) {
2743 output_mode = JAPANESE_EUC;
2744 (*o_putc)(SSO); (*o_putc)(c1|0x80);
2745 } else if (c2 == ISO8859_1) {
2746 output_mode = ISO8859_1;
2747 (*o_putc)(c1 | 0x080);
2749 } else if ((c2 & 0xff00) >> 8 == 0x8f){
2750 output_mode = JAPANESE_EUC;
2751 #ifdef SHIFTJIS_CP932
2754 if (e2s_conv(c2, c1, &s2, &s1) == 0){
2755 s2e_conv(s2, s1, &c2, &c1);
2759 if ((c2 & 0xff00) >> 8 == 0x8f){
2762 (*o_putc)((c2 & 0x7f) | 0x080);
2763 (*o_putc)(c1 | 0x080);
2766 (*o_putc)((c2 & 0x7f) | 0x080);
2767 (*o_putc)(c1 | 0x080);
2771 if ((c1<0x21 || 0x7e<c1) ||
2772 (c2<0x21 || 0x7e<c2)) {
2773 set_iconv(FALSE, 0);
2774 return; /* too late to rescue this char */
2776 output_mode = JAPANESE_EUC;
2777 (*o_putc)(c2 | 0x080);
2778 (*o_putc)(c1 | 0x080);
2788 if ((ret & 0xff00) == 0x8f00){
2789 if (0x75 <= c && c <= 0x7f){
2790 ret = c + (0x109 - 0x75);
2793 if (0x75 <= c && c <= 0x7f){
2794 ret = c + (0x113 - 0x75);
2801 int x0212_unshift(c)
2805 if (0x7f <= c && c <= 0x88){
2806 ret = c + (0x75 - 0x7f);
2807 }else if (0x89 <= c && c <= 0x92){
2808 ret = (0x8f << 8) | 0x80 | (c + (0x75 - 0x89));
2812 #endif /* X0212_ENABLE */
2815 e2s_conv(c2, c1, p2, p1)
2816 int c2, c1, *p2, *p1;
2819 unsigned short *ptr;
2821 if ((c2 & 0xff00) == 0x8f00){
2822 extern unsigned short *x0212_shiftjis[];
2823 ptr = x0212_shiftjis[(c2 & 0x7f) - 0x21];
2825 val = ptr[(c1 & 0x7f) - 0x21];
2834 c2 = x0212_shift(c2);
2836 #endif /* X0212_ENABLE */
2837 if ((c2 & 0xff00) == 0x8f00){
2840 if (p2) *p2 = ((c2 - 1) >> 1) + ((c2 <= 0x5e) ? 0x71 : 0xb1);
2841 if (p1) *p1 = c1 + ((c2 & 1) ? ((c1 < 0x60) ? 0x1f : 0x20) : 0x7e);
2850 #ifdef NUMCHAR_OPTION
2851 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2852 w16e_conv(c1, &c2, &c1);
2858 } else if (c2 == 0) {
2859 output_mode = ASCII;
2861 } else if (c2 == X0201) {
2862 output_mode = SHIFT_JIS;
2864 } else if (c2 == ISO8859_1) {
2865 output_mode = ISO8859_1;
2866 (*o_putc)(c1 | 0x080);
2868 } else if ((c2 & 0xff00) >> 8 == 0x8f){
2869 output_mode = SHIFT_JIS;
2870 if (e2s_conv(c2, c1, &c2, &c1) == 0){
2876 if ((c1<0x20 || 0x7e<c1) ||
2877 (c2<0x20 || 0x7e<c2)) {
2878 set_iconv(FALSE, 0);
2879 return; /* too late to rescue this char */
2881 output_mode = SHIFT_JIS;
2882 e2s_conv(c2, c1, &c2, &c1);
2884 #ifdef SHIFTJIS_CP932
2886 && CP932INV_TABLE_BEGIN <= c2 && c2 <= CP932INV_TABLE_END){
2887 extern unsigned short cp932inv[2][189];
2888 int c = cp932inv[c2 - CP932INV_TABLE_BEGIN][c1 - 0x40];
2894 #endif /* SHIFTJIS_CP932 */
2897 if (prefix_table[(unsigned char)c1]){
2898 (*o_putc)(prefix_table[(unsigned char)c1]);
2909 #ifdef NUMCHAR_OPTION
2910 if ((c1 & CLASS_MASK) == CLASS_UTF16){
2911 w16e_conv(c1, &c2, &c1);
2915 if (output_mode !=ASCII && output_mode!=ISO8859_1) {
2918 (*o_putc)(ascii_intro);
2919 output_mode = ASCII;
2923 } else if ((c2 & 0xff00) >> 8 == 0x8f){
2924 if (output_mode!=X0212) {
2925 output_mode = X0212;
2931 (*o_putc)(c2 & 0x7f);
2934 } else if (c2==X0201) {
2935 if (output_mode!=X0201) {
2936 output_mode = X0201;
2942 } else if (c2==ISO8859_1) {
2943 /* iso8859 introduction, or 8th bit on */
2944 /* Can we convert in 7bit form using ESC-'-'-A ?
2946 output_mode = ISO8859_1;
2948 } else if (c2 == 0) {
2949 if (output_mode !=ASCII && output_mode!=ISO8859_1) {
2952 (*o_putc)(ascii_intro);
2953 output_mode = ASCII;
2957 if (output_mode != X0208) {
2958 output_mode = X0208;
2961 (*o_putc)(kanji_intro);
2963 if (c1<0x20 || 0x7e<c1)
2965 if (c2<0x20 || 0x7e<c2)
2977 if (base64_count>50 && !mimeout_mode && c2==0 && c1==SPACE) {
2980 } else if (base64_count>66 && mimeout_mode) {
2981 (*o_base64conv)(EOF,0);
2982 (*o_base64conv)(NL,0);
2983 (*o_base64conv)(SPACE,0);
2985 (*o_base64conv)(c2,c1);
2989 static int broken_buf[3];
2990 static int broken_counter = 0;
2991 static int broken_last = 0;
2998 if (broken_counter>0) {
2999 return broken_buf[--broken_counter];
3002 if (c=='$' && broken_last != ESC
3003 && (input_mode==ASCII || input_mode==X0201)) {
3006 if (c1=='@'|| c1=='B') {
3007 broken_buf[0]=c1; broken_buf[1]=c;
3014 } else if (c=='(' && broken_last != ESC
3015 && (input_mode==X0208 || input_mode==X0201)) { /* ) */
3018 if (c1=='J'|| c1=='B') {
3019 broken_buf[0]=c1; broken_buf[1]=c;
3037 if (broken_counter<2)
3038 broken_buf[broken_counter++]=c;
3042 static int prev_cr = 0;
3050 if (! (c2==0&&c1==NL) ) {
3056 } else if (c1=='\r') {
3058 } else if (c1=='\n') {
3059 if (crmode_f==CRLF) {
3060 (*o_crconv)(0,'\r');
3061 } else if (crmode_f==CR) {
3062 (*o_crconv)(0,'\r');
3066 } else if (c1!='\032' || crmode_f!=NL){
3072 Return value of fold_conv()
3074 \n add newline and output char
3075 \r add newline and output nothing
3078 1 (or else) normal output
3080 fold state in prev (previous character)
3082 >0x80 Japanese (X0208/X0201)
3087 This fold algorthm does not preserve heading space in a line.
3088 This is the main difference from fmt.
3091 #define char_size(c2,c1) (c2?2:1)
3100 if (c1== '\r' && !fold_preserve_f) {
3101 fold_state=0; /* ignore cr */
3102 }else if (c1== '\n'&&f_prev=='\r' && fold_preserve_f) {
3104 fold_state=0; /* ignore cr */
3105 } else if (c1== BS) {
3106 if (f_line>0) f_line--;
3108 } else if (c2==EOF && f_line != 0) { /* close open last line */
3110 } else if ((c1=='\n' && !fold_preserve_f)
3111 || ((c1=='\r'||(c1=='\n'&&f_prev!='\r'))
3112 && fold_preserve_f)) {
3114 if (fold_preserve_f) {
3118 } else if ((f_prev == c1 && !fold_preserve_f)
3119 || (f_prev == '\n' && fold_preserve_f)
3120 ) { /* duplicate newline */
3123 fold_state = '\n'; /* output two newline */
3129 if (f_prev&0x80) { /* Japanese? */
3131 fold_state = 0; /* ignore given single newline */
3132 } else if (f_prev==' ') {
3136 if (++f_line<=fold_len)
3140 fold_state = '\r'; /* fold and output nothing */
3144 } else if (c1=='\f') {
3149 fold_state = '\n'; /* output newline and clear */
3150 } else if ( (c2==0 && c1==' ')||
3151 (c2==0 && c1=='\t')||
3152 (c2=='!'&& c1=='!')) {
3153 /* X0208 kankaku or ascii space */
3154 if (f_prev == ' ') {
3155 fold_state = 0; /* remove duplicate spaces */
3158 if (++f_line<=fold_len)
3159 fold_state = ' '; /* output ASCII space only */
3161 f_prev = ' '; f_line = 0;
3162 fold_state = '\r'; /* fold and output nothing */
3166 prev0 = f_prev; /* we still need this one... , but almost done */
3168 if (c2 || c2==X0201)
3169 f_prev |= 0x80; /* this is Japanese */
3170 f_line += char_size(c2,c1);
3171 if (f_line<=fold_len) { /* normal case */
3174 if (f_line>=fold_len+fold_margin) { /* too many kinsou suspension */
3175 f_line = char_size(c2,c1);
3176 fold_state = '\n'; /* We can't wait, do fold now */
3177 } else if (c2==X0201) {
3178 /* simple kinsoku rules return 1 means no folding */
3179 if (c1==(0xde&0x7f)) fold_state = 1; /*
\e$B!+
\e(B*/
3180 else if (c1==(0xdf&0x7f)) fold_state = 1; /*
\e$B!,
\e(B*/
3181 else if (c1==(0xa4&0x7f)) fold_state = 1; /*
\e$B!#
\e(B*/
3182 else if (c1==(0xa3&0x7f)) fold_state = 1; /*
\e$B!$
\e(B*/
3183 else if (c1==(0xa1&0x7f)) fold_state = 1; /*
\e$B!W
\e(B*/
3184 else if (c1==(0xb0&0x7f)) fold_state = 1; /* - */
3185 else if (SPACE<=c1 && c1<=(0xdf&0x7f)) { /* X0201 */
3187 fold_state = '\n';/* add one new f_line before this character */
3190 fold_state = '\n';/* add one new f_line before this character */
3193 /* kinsoku point in ASCII */
3194 if ( c1==')'|| /* { [ ( */
3205 /* just after special */
3206 } else if (!is_alnum(prev0)) {
3207 f_line = char_size(c2,c1);
3209 } else if ((prev0==' ') || /* ignored new f_line */
3210 (prev0=='\n')|| /* ignored new f_line */
3211 (prev0&0x80)) { /* X0208 - ASCII */
3212 f_line = char_size(c2,c1);
3213 fold_state = '\n';/* add one new f_line before this character */
3215 fold_state = 1; /* default no fold in ASCII */
3219 if (c1=='"') fold_state = 1; /*
\e$B!"
\e(B */
3220 else if (c1=='#') fold_state = 1; /*
\e$B!#
\e(B */
3221 else if (c1=='W') fold_state = 1; /*
\e$B!W
\e(B */
3222 else if (c1=='K') fold_state = 1; /*
\e$B!K
\e(B */
3223 else if (c1=='$') fold_state = 1; /*
\e$B!$
\e(B */
3224 else if (c1=='%') fold_state = 1; /*
\e$B!%
\e(B */
3225 else if (c1=='\'') fold_state = 1; /*
\e$B!\
\e(B */
3226 else if (c1=='(') fold_state = 1; /*
\e$B!(
\e(B */
3227 else if (c1==')') fold_state = 1; /*
\e$B!)
\e(B */
3228 else if (c1=='*') fold_state = 1; /*
\e$B!*
\e(B */
3229 else if (c1=='+') fold_state = 1; /*
\e$B!+
\e(B */
3230 else if (c1==',') fold_state = 1; /*
\e$B!,
\e(B */
3231 /* default no fold in kinsoku */
3234 f_line = char_size(c2,c1);
3235 /* add one new f_line before this character */
3238 f_line = char_size(c2,c1);
3240 /* add one new f_line before this character */
3245 /* terminator process */
3246 switch(fold_state) {
3265 int z_prev2=0,z_prev1=0;
3272 /* if (c2) c1 &= 0x7f; assertion */
3274 if (x0201_f && z_prev2==X0201) { /* X0201 */
3275 if (c1==(0xde&0x7f)) { /*
\e$BByE@
\e(B */
3277 (*o_zconv)(dv[(z_prev1-SPACE)*2],dv[(z_prev1-SPACE)*2+1]);
3279 } else if (c1==(0xdf&0x7f)&&ev[(z_prev1-SPACE)*2]) { /*
\e$BH>ByE@
\e(B */
3281 (*o_zconv)(ev[(z_prev1-SPACE)*2],ev[(z_prev1-SPACE)*2+1]);
3285 (*o_zconv)(cv[(z_prev1-SPACE)*2],cv[(z_prev1-SPACE)*2+1]);
3294 if (x0201_f && c2==X0201) {
3295 if (dv[(c1-SPACE)*2]||ev[(c1-SPACE)*2]) {
3296 /* wait for
\e$BByE@
\e(B or
\e$BH>ByE@
\e(B */
3297 z_prev1 = c1; z_prev2 = c2;
3300 (*o_zconv)(cv[(c1-SPACE)*2],cv[(c1-SPACE)*2+1]);
3305 /* JISX0208 Alphabet */
3306 if (alpha_f && c2 == 0x23 ) {
3308 } else if (alpha_f && c2 == 0x21 ) {
3309 /* JISX0208 Kigou */
3314 } else if (alpha_f&0x4) {
3319 } else if (0x20<c1 && c1<0x7f && fv[c1-0x20]) {
3325 case '>': entity = ">"; break;
3326 case '<': entity = "<"; break;
3327 case '\"': entity = """; break;
3328 case '&': entity = "&"; break;
3331 while (*entity) (*o_zconv)(0, *entity++);
3341 #define rot13(c) ( \
3343 (c <= 'M') ? (c + 13): \
3344 (c <= 'Z') ? (c - 13): \
3346 (c <= 'm') ? (c + 13): \
3347 (c <= 'z') ? (c - 13): \
3351 #define rot47(c) ( \
3353 ( c <= 'O' ) ? (c + 47) : \
3354 ( c <= '~' ) ? (c - 47) : \
3362 if (c2==0 || c2==X0201 || c2==ISO8859_1) {
3368 (*o_rot_conv)(c2,c1);
3375 if ((hira_f & 1) && c2==0x25 && 0x20<c1 && c1<0x74) {
3377 } else if ((hira_f & 2) && c2==0x24 && 0x20<c1 && c1<0x74) {
3380 (*o_hira_conv)(c2,c1);
3385 iso2022jp_check_conv(c2,c1)
3388 static int range[RANGE_NUM_MAX][2] = {
3411 if(c2 >= 0x00 && c2 <= 0x20 && c1 >= 0x7f && c1 <= 0xff) {
3415 if((c2 >= 0x29 && c2 <= 0x2f) || (c2 >= 0x75 && c2 <= 0x7e)) {
3420 for (i = 0; i < RANGE_NUM_MAX; i++) {
3421 start = range[i][0];
3424 if (c >= start && c <= end) {
3429 (*o_iso2022jp_check_conv)(c2,c1);
3433 /* This converts =?ISO-2022-JP?B?HOGE HOGE?= */
3435 unsigned char *mime_pattern[] = {
3436 (unsigned char *)"\075?EUC-JP?B?",
3437 (unsigned char *)"\075?SHIFT_JIS?B?",
3438 (unsigned char *)"\075?ISO-8859-1?Q?",
3439 (unsigned char *)"\075?ISO-8859-1?B?",
3440 (unsigned char *)"\075?ISO-2022-JP?B?",
3441 (unsigned char *)"\075?ISO-2022-JP?Q?",
3442 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3443 (unsigned char *)"\075?UTF-8?B?",
3444 (unsigned char *)"\075?UTF-8?Q?",
3446 (unsigned char *)"\075?US-ASCII?Q?",
3451 /*
\e$B3:Ev$9$k%3!<%I$NM%@hEY$r>e$2$k$?$a$NL\0u
\e(B */
3452 int (*mime_priority_func[])PROTO((int c2, int c1, int c0)) = {
3453 e_iconv, s_iconv, 0, 0, 0, 0,
3454 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3460 int mime_encode[] = {
3461 JAPANESE_EUC, SHIFT_JIS,ISO8859_1, ISO8859_1, X0208, X0201,
3462 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3469 int mime_encode_method[] = {
3470 'B', 'B','Q', 'B', 'B', 'Q',
3471 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3479 #define MAXRECOVER 20
3481 /* I don't trust portablity of toupper */
3482 #define nkf_toupper(c) (('a'<=c && c<='z')?(c-('a'-'A')):c)
3483 #define nkf_isdigit(c) ('0'<=c && c<='9')
3484 #define nkf_isxdigit(c) (nkf_isdigit(c) || ('a'<=c && c<='f') || ('A'<=c && c <= 'F'))
3489 if (i_getc!=mime_getc) {
3490 i_mgetc = i_getc; i_getc = mime_getc;
3491 i_mungetc = i_ungetc; i_ungetc = mime_ungetc;
3492 if(mime_f==STRICT_MIME) {
3493 i_mgetc_buf = i_mgetc; i_mgetc = mime_getc_buf;
3494 i_mungetc_buf = i_mungetc; i_mungetc = mime_ungetc_buf;
3500 unswitch_mime_getc()
3502 if(mime_f==STRICT_MIME) {
3503 i_mgetc = i_mgetc_buf;
3504 i_mungetc = i_mungetc_buf;
3507 i_ungetc = i_mungetc;
3511 mime_begin_strict(f)
3516 unsigned char *p,*q;
3517 int r[MAXRECOVER]; /* recovery buffer, max mime pattern lenght */
3519 mime_decode_mode = FALSE;
3520 /* =? has been checked */
3522 p = mime_pattern[j];
3525 for(i=2;p[i]>' ';i++) { /* start at =? */
3526 if ( ((r[i] = c1 = (*i_getc)(f))==EOF) || nkf_toupper(c1) != p[i] ) {
3527 /* pattern fails, try next one */
3529 while ((p = mime_pattern[++j])) {
3530 for(k=2;k<i;k++) /* assume length(p) > i */
3531 if (p[k]!=q[k]) break;
3532 if (k==i && nkf_toupper(c1)==p[k]) break;
3534 if (p) continue; /* found next one, continue */
3535 /* all fails, output from recovery buffer */
3543 mime_decode_mode = p[i-2];
3545 clr_code_score(find_inputcode_byfunc(mime_priority_func[j]), SCORE_iMIME);
3547 if (mime_decode_mode=='B') {
3548 mimebuf_f = unbuf_f;
3550 /* do MIME integrity check */
3551 return mime_integrity(f,mime_pattern[j]);
3563 /* we don't keep eof of Fifo, becase it contains ?= as
3564 a terminator. It was checked in mime_integrity. */
3565 return ((mimebuf_f)?
3566 (*i_mgetc_buf)(f):Fifo(mime_input++));
3570 mime_ungetc_buf(c,f)
3575 (*i_mungetc_buf)(c,f);
3577 Fifo(--mime_input)=c;
3588 /* In NONSTRICT mode, only =? is checked. In case of failure, we */
3589 /* re-read and convert again from mime_buffer. */
3591 /* =? has been checked */
3593 Fifo(mime_last++)='='; Fifo(mime_last++)='?';
3594 for(i=2;i<MAXRECOVER;i++) { /* start at =? */
3595 /* We accept any character type even if it is breaked by new lines */
3596 c1 = (*i_getc)(f); Fifo(mime_last++)= c1 ;
3597 if (c1=='\n'||c1==' '||c1=='\r'||
3598 c1=='-'||c1=='_'||is_alnum(c1) ) continue;
3600 /* Failed. But this could be another MIME preemble */
3608 c1 = (*i_getc)(f); Fifo(mime_last++) = c1;
3609 if (!(++i<MAXRECOVER) || c1==EOF) break;
3610 if (c1=='b'||c1=='B') {
3611 mime_decode_mode = 'B';
3612 } else if (c1=='q'||c1=='Q') {
3613 mime_decode_mode = 'Q';
3617 c1 = (*i_getc)(f); Fifo(mime_last++) = c1;
3618 if (!(++i<MAXRECOVER) || c1==EOF) break;
3620 mime_decode_mode = FALSE;
3626 if (!mime_decode_mode) {
3627 /* false MIME premble, restart from mime_buffer */
3628 mime_decode_mode = 1; /* no decode, but read from the mime_buffer */
3629 /* Since we are in MIME mode until buffer becomes empty, */
3630 /* we never go into mime_begin again for a while. */
3633 /* discard mime preemble, and goto MIME mode */
3635 /* do no MIME integrity check */
3636 return c1; /* used only for checking EOF */
3651 fprintf(stderr, "%s\n", str);
3657 set_input_codename (codename)
3662 strcmp(codename, "") != 0 &&
3663 strcmp(codename, input_codename) != 0)
3665 is_inputcode_mixed = TRUE;
3667 input_codename = codename;
3668 is_inputcode_set = TRUE;
3672 print_guessed_code (filename)
3675 char *codename = "BINARY";
3676 if (!is_inputcode_mixed) {
3677 if (strcmp(input_codename, "") == 0) {
3680 codename = input_codename;
3683 if (filename != NULL) printf("%s:", filename);
3684 printf("%s\n", codename);
3691 if (nkf_isdigit(x)) return x - '0';
3692 return nkf_toupper(x) - 'A' + 10;
3697 #ifdef ANSI_C_PROTOTYPE
3698 int hex_getc(int ch, FILE *f, int (*g)(FILE *f), int (*u)(int c, FILE *f))
3701 hex_getc(ch, f, g, u)
3714 if (!nkf_isxdigit(c2)){
3719 if (!nkf_isxdigit(c3)){
3724 return (hex2bin(c2) << 4) | hex2bin(c3);
3731 return hex_getc(':', f, i_cgetc, i_cungetc);
3739 return (*i_cungetc)(c, f);
3746 return hex_getc('%', f, i_ugetc, i_uungetc);
3754 return (*i_uungetc)(c, f);
3758 #ifdef NUMCHAR_OPTION
3763 int (*g)() = i_ngetc;
3764 int (*u)() = i_nungetc;
3775 if (buf[i] == 'x' || buf[i] == 'X'){
3776 for (j = 0; j < 5; j++){
3778 if (!nkf_isxdigit(buf[i])){
3785 c |= hex2bin(buf[i]);
3788 for (j = 0; j < 6; j++){
3792 if (!nkf_isdigit(buf[i])){
3799 c += hex2bin(buf[i]);
3805 return CLASS_UTF16 | c;
3815 numchar_ungetc(c, f)
3819 return (*i_nungetc)(c, f);
3828 int c1, c2, c3, c4, cc;
3829 int t1, t2, t3, t4, mode, exit_mode;
3833 int lwsp_size = 128;
3835 if (mime_top != mime_last) { /* Something is in FIFO */
3836 return Fifo(mime_top++);
3838 if (mime_decode_mode==1 ||mime_decode_mode==FALSE) {
3839 mime_decode_mode=FALSE;
3840 unswitch_mime_getc();
3841 return (*i_getc)(f);
3844 if (mimebuf_f == FIXED_MIME)
3845 exit_mode = mime_decode_mode;
3848 if (mime_decode_mode == 'Q') {
3849 if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF);
3851 if (c1=='_') return ' ';
3852 if (c1!='=' && c1!='?') {
3856 mime_decode_mode = exit_mode; /* prepare for quit */
3857 if (c1<=' ') return c1;
3858 if ((c2 = (*i_mgetc)(f)) == EOF) return (EOF);
3859 if (c1=='?'&&c2=='=' && mimebuf_f != FIXED_MIME) {
3860 /* end Q encoding */
3861 input_mode = exit_mode;
3863 lwsp_buf = malloc((lwsp_size+5)*sizeof(char));
3864 if (lwsp_buf==NULL) {
3865 perror("can't malloc");
3868 while ((c1=(*i_getc)(f))!=EOF) {
3873 if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) {
3881 if ((c1=(*i_getc)(f))!=EOF && c1 == NL) {
3882 if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) {
3897 lwsp_buf[lwsp_count] = c1;
3898 if (lwsp_count++>lwsp_size){
3900 lwsp_buf_new = realloc(lwsp_buf, (lwsp_size+5)*sizeof(char));
3901 if (lwsp_buf_new==NULL) {
3904 perror("can't realloc");
3907 lwsp_buf = lwsp_buf_new;
3913 if (lwsp_count > 0) {
3914 if (c1=='=' && (lwsp_buf[lwsp_count-1]==SPACE||lwsp_buf[lwsp_count-1]==TAB)) {
3918 for(lwsp_count--;lwsp_count>0;lwsp_count--)
3919 i_ungetc(lwsp_buf[lwsp_count],f);
3927 if (c1=='='&&c2<' ') { /* this is soft wrap */
3928 while((c1 = (*i_mgetc)(f)) <=' ') {
3929 if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF);
3931 mime_decode_mode = 'Q'; /* still in MIME */
3932 goto restart_mime_q;
3935 mime_decode_mode = 'Q'; /* still in MIME */
3939 if ((c3 = (*i_mgetc)(f)) == EOF) return (EOF);
3940 if (c2<=' ') return c2;
3941 mime_decode_mode = 'Q'; /* still in MIME */
3942 #define hex(c) (('0'<=c&&c<='9')?(c-'0'):\
3943 ('A'<=c&&c<='F')?(c-'A'+10):('a'<=c&&c<='f')?(c-'a'+10):0)
3944 return ((hex(c2)<<4) + hex(c3));
3947 if (mime_decode_mode != 'B') {
3948 mime_decode_mode = FALSE;
3949 return (*i_mgetc)(f);
3953 /* Base64 encoding */
3955 MIME allows line break in the middle of
3956 Base64, but we are very pessimistic in decoding
3957 in unbuf mode because MIME encoded code may broken by
3958 less or editor's control sequence (such as ESC-[-K in unbuffered
3959 mode. ignore incomplete MIME.
3961 mode = mime_decode_mode;
3962 mime_decode_mode = exit_mode; /* prepare for quit */
3964 while ((c1 = (*i_mgetc)(f))<=' ') {
3969 if ((c2 = (*i_mgetc)(f))<=' ') {
3972 if (mime_f != STRICT_MIME) goto mime_c2_retry;
3973 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
3976 if ((c1 == '?') && (c2 == '=')) {
3979 lwsp_buf = malloc((lwsp_size+5)*sizeof(char));
3980 if (lwsp_buf==NULL) {
3981 perror("can't malloc");
3984 while ((c1=(*i_getc)(f))!=EOF) {
3989 if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) {
3997 if ((c1=(*i_getc)(f))!=EOF) {
4001 } else if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) {
4016 lwsp_buf[lwsp_count] = c1;
4017 if (lwsp_count++>lwsp_size){
4019 lwsp_buf_new = realloc(lwsp_buf, (lwsp_size+5)*sizeof(char));
4020 if (lwsp_buf_new==NULL) {
4023 perror("can't realloc");
4026 lwsp_buf = lwsp_buf_new;
4032 if (lwsp_count > 0) {
4033 if (c1=='=' && (lwsp_buf[lwsp_count-1]==SPACE||lwsp_buf[lwsp_count-1]==TAB)) {
4037 for(lwsp_count--;lwsp_count>0;lwsp_count--)
4038 i_ungetc(lwsp_buf[lwsp_count],f);
4047 if ((c3 = (*i_mgetc)(f))<=' ') {
4050 if (mime_f != STRICT_MIME) goto mime_c3_retry;
4051 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
4055 if ((c4 = (*i_mgetc)(f))<=' ') {
4058 if (mime_f != STRICT_MIME) goto mime_c4_retry;
4059 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
4063 mime_decode_mode = mode; /* still in MIME sigh... */
4065 /* BASE 64 decoding */
4067 t1 = 0x3f & base64decode(c1);
4068 t2 = 0x3f & base64decode(c2);
4069 t3 = 0x3f & base64decode(c3);
4070 t4 = 0x3f & base64decode(c4);
4071 cc = ((t1 << 2) & 0x0fc) | ((t2 >> 4) & 0x03);
4073 Fifo(mime_last++) = cc;
4074 cc = ((t2 << 4) & 0x0f0) | ((t3 >> 2) & 0x0f);
4076 Fifo(mime_last++) = cc;
4077 cc = ((t3 << 6) & 0x0c0) | (t4 & 0x3f);
4079 Fifo(mime_last++) = cc;
4084 return Fifo(mime_top++);
4092 Fifo(--mime_top) = c;
4103 /* In buffered mode, read until =? or NL or buffer full
4105 mime_input = mime_top;
4106 mime_last = mime_top;
4107 while(*p) Fifo(mime_input++) = *p++;
4110 while((c=(*i_getc)(f))!=EOF) {
4111 if (((mime_input-mime_top)&MIME_BUF_MASK)==0) {
4112 break; /* buffer full */
4114 if (c=='=' && d=='?') {
4115 /* checked. skip header, start decode */
4116 Fifo(mime_input++) = c;
4117 /* mime_last_input = mime_input; */
4122 if (!( (c=='+'||c=='/'|| c=='=' || c=='?' || is_alnum(c))))
4124 /* Should we check length mod 4? */
4125 Fifo(mime_input++) = c;
4128 /* In case of Incomplete MIME, no MIME decode */
4129 Fifo(mime_input++) = c;
4130 mime_last = mime_input; /* point undecoded buffer */
4131 mime_decode_mode = 1; /* no decode on Fifo last in mime_getc */
4132 switch_mime_getc(); /* anyway we need buffered getc */
4143 i = c - 'A'; /* A..Z 0-25 */
4145 i = c - 'G' /* - 'a' + 26 */ ; /* a..z 26-51 */
4147 } else if (c > '/') {
4148 i = c - '0' + '4' /* - '0' + 52 */ ; /* 0..9 52-61 */
4149 } else if (c == '+') {
4150 i = '>' /* 62 */ ; /* + 62 */
4152 i = '?' /* 63 */ ; /* / 63 */
4157 static char basis_64[] =
4158 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
4161 #define MIMEOUT_BUF_LENGTH (60)
4162 char mimeout_buf[MIMEOUT_BUF_LENGTH+1];
4163 int mimeout_buf_count = 0;
4164 int mimeout_preserve_space = 0;
4165 #define itoh4(c) (c>=10?c+'A'-10:c+'0')
4174 p = mime_pattern[0];
4175 for(i=0;mime_encode[i];i++) {
4176 if (mode == mime_encode[i]) {
4177 p = mime_pattern[i];
4181 mimeout_mode = mime_encode_method[i];
4184 if (base64_count>45) {
4188 if (!mimeout_preserve_space && mimeout_buf_count>0
4189 && (mimeout_buf[i]==SPACE || mimeout_buf[i]==TAB
4190 || mimeout_buf[i]==CR || mimeout_buf[i]==NL )) {
4194 if (!mimeout_preserve_space) {
4195 for (;i<mimeout_buf_count;i++) {
4196 if (mimeout_buf[i]==SPACE || mimeout_buf[i]==TAB
4197 || mimeout_buf[i]==CR || mimeout_buf[i]==NL ) {
4198 (*o_mputc)(mimeout_buf[i]);
4205 mimeout_preserve_space = FALSE;
4211 j = mimeout_buf_count;
4212 mimeout_buf_count = 0;
4214 mime_putc(mimeout_buf[i]);
4230 switch(mimeout_mode) {
4235 (*o_mputc)(basis_64[((b64c & 0x3)<< 4)]);
4241 (*o_mputc)(basis_64[((b64c & 0xF) << 2)]);
4247 if (mimeout_f!=FIXED_MIME) {
4249 } else if (mimeout_mode != 'Q')
4258 switch(mimeout_mode) {
4262 (*o_mputc)(itoh4(((c>>4)&0xf)));
4263 (*o_mputc)(itoh4((c&0xf)));
4272 (*o_mputc)(basis_64[c>>2]);
4277 (*o_mputc)(basis_64[((b64c & 0x3)<< 4) | ((c & 0xF0) >> 4)]);
4283 (*o_mputc)(basis_64[((b64c & 0xF) << 2) | ((c & 0xC0) >>6)]);
4284 (*o_mputc)(basis_64[c & 0x3F]);
4298 if (mimeout_f==FIXED_MIME && base64_count>50) {
4302 } else if (c==CR||c==NL) {
4305 if (c!=EOF && mimeout_f!=FIXED_MIME) {
4306 if ( c<=DEL &&(output_mode==ASCII ||output_mode == ISO8859_1 ) ) {
4307 if (mimeout_mode=='Q') {
4316 } else if (mimeout_mode) {
4317 if (base64_count>63) {
4322 mimeout_preserve_space = TRUE;
4324 if (c==SPACE || c==TAB || c==CR || c==NL) {
4325 for (i=0;i<mimeout_buf_count;i++) {
4326 if (SPACE<mimeout_buf[i] && mimeout_buf[i]<DEL) {
4328 for (i=0;i<mimeout_buf_count;i++) {
4329 (*o_mputc)(mimeout_buf[i]);
4332 mimeout_buf_count = 0;
4335 mimeout_buf[mimeout_buf_count++] = c;
4336 if (mimeout_buf_count>MIMEOUT_BUF_LENGTH) {
4339 for (i=0;i<mimeout_buf_count;i++) {
4340 (*o_mputc)(mimeout_buf[i]);
4346 if (mimeout_buf_count>0 && SPACE<c) {
4347 mimeout_buf[mimeout_buf_count++] = c;
4348 if (mimeout_buf_count>MIMEOUT_BUF_LENGTH) {
4353 } else if (!mimeout_mode) {
4354 if (c==SPACE || c==TAB || c==CR || c==NL) {
4355 if ((c==CR || c==NL)
4356 &&(mimeout_buf[mimeout_buf_count-1]==SPACE
4357 || mimeout_buf[mimeout_buf_count-1]==TAB)) {
4358 mimeout_buf_count--;
4360 for (i=0;i<mimeout_buf_count;i++) {
4361 (*o_mputc)(mimeout_buf[i]);
4364 mimeout_buf_count = 0;
4366 mimeout_buf[mimeout_buf_count++] = c;
4367 if (mimeout_buf_count>75) {
4368 open_mime(output_mode);
4372 } else if (!mimeout_mode) {
4373 if (mimeout_buf_count>0 && mimeout_buf[mimeout_buf_count-1]==SPACE) {
4374 for (i=0;i<mimeout_buf_count-1;i++) {
4375 (*o_mputc)(mimeout_buf[i]);
4378 mimeout_buf[0] = SPACE;
4379 mimeout_buf_count = 1;
4381 open_mime(output_mode);
4383 } else { /* c==EOF */
4384 j = mimeout_buf_count;
4387 if (mimeout_buf[i]==SPACE || mimeout_buf[i]==TAB
4388 || mimeout_buf[i]==CR || mimeout_buf[i]==NL)
4390 (*mime_putc)(mimeout_buf[i]);
4394 (*o_mputc)(mimeout_buf[i]);
4400 if (mimeout_buf_count>0) {
4401 j = mimeout_buf_count;
4402 mimeout_buf_count = 0;
4404 mimeout_addchar(mimeout_buf[i]);
4416 struct input_code *p = input_code_list;
4429 mime_f = STRICT_MIME;
4434 #if defined(MSDOS) || defined(__OS2__)
4439 iso2022jp_f = FALSE;
4440 #ifdef UTF8_OUTPUT_ENABLE
4443 ms_ucs_map_f = FALSE;
4455 is_inputcode_mixed = FALSE;
4456 is_inputcode_set = FALSE;
4460 #ifdef SHIFTJIS_CP932
4466 for (i = 0; i < 256; i++){
4467 prefix_table[i] = 0;
4470 #ifdef UTF8_INPUT_ENABLE
4471 utf16_mode = UTF16LE_INPUT;
4473 mimeout_buf_count = 0;
4478 fold_preserve_f = FALSE;
4481 kanji_intro = DEFAULT_J;
4482 ascii_intro = DEFAULT_R;
4483 fold_margin = FOLD_MARGIN;
4484 output_conv = DEFAULT_CONV;
4485 oconv = DEFAULT_CONV;
4486 o_zconv = no_connection;
4487 o_fconv = no_connection;
4488 o_crconv = no_connection;
4489 o_rot_conv = no_connection;
4490 o_hira_conv = no_connection;
4491 o_base64conv = no_connection;
4492 o_iso2022jp_check_conv = no_connection;
4495 i_ungetc = std_ungetc;
4497 i_bungetc = std_ungetc;
4500 i_mungetc = std_ungetc;
4501 i_mgetc_buf = std_getc;
4502 i_mungetc_buf = std_ungetc;
4503 output_mode = ASCII;
4506 mime_decode_mode = FALSE;
4512 z_prev2=0,z_prev1=0;
4518 no_connection(c2,c1)
4521 no_connection2(c2,c1,0);
4525 no_connection2(c2,c1,c0)
4528 fprintf(stderr,"nkf internal module connection failure.\n");
4536 fprintf(stderr,"USAGE: nkf(nkf32,wnkf,nkf2) -[flags] [in file] .. [out file for -O flag]\n");
4537 fprintf(stderr,"Flags:\n");
4538 fprintf(stderr,"b,u Output is buffered (DEFAULT),Output is unbuffered\n");
4539 #ifdef DEFAULT_CODE_SJIS
4540 fprintf(stderr,"j,s,e,w Outout code is JIS 7 bit, Shift JIS (DEFAULT), AT&T JIS (EUC), UTF-8\n");
4542 #ifdef DEFAULT_CODE_JIS
4543 fprintf(stderr,"j,s,e,w Outout code is JIS 7 bit (DEFAULT), Shift JIS, AT&T JIS (EUC), UTF-8\n");
4545 #ifdef DEFAULT_CODE_EUC
4546 fprintf(stderr,"j,s,e,w Outout code is JIS 7 bit, Shift JIS, AT&T JIS (EUC) (DEFAULT), UTF-8\n");
4548 #ifdef DEFAULT_CODE_UTF8
4549 fprintf(stderr,"j,s,e,w Outout code is JIS 7 bit, Shift JIS, AT&T JIS (EUC), UTF-8 (DEFAULT)\n");
4551 #ifdef UTF8_OUTPUT_ENABLE
4552 fprintf(stderr," After 'w' you can add more options. (80?|16((B|L)0?)?) \n");
4554 fprintf(stderr,"J,S,E,W Input assumption is JIS 7 bit , Shift JIS, AT&T JIS (EUC), UTF-8\n");
4555 #ifdef UTF8_INPUT_ENABLE
4556 fprintf(stderr," After 'W' you can add more options. (8|16(B|L)?) \n");
4558 fprintf(stderr,"t no conversion\n");
4559 fprintf(stderr,"i_/o_ Output sequence to designate JIS-kanji/ASCII (DEFAULT B)\n");
4560 fprintf(stderr,"r {de/en}crypt ROT13/47\n");
4561 fprintf(stderr,"h 1 hirakana->katakana, 2 katakana->hirakana,3 both\n");
4562 fprintf(stderr,"v Show this usage. V: show version\n");
4563 fprintf(stderr,"m[BQN0] MIME decode [B:base64,Q:quoted,N:non-strict,0:no decode]\n");
4564 fprintf(stderr,"M[BQ] MIME encode [B:base64 Q:quoted]\n");
4565 fprintf(stderr,"l ISO8859-1 (Latin-1) support\n");
4566 fprintf(stderr,"f/F Folding: -f60 or -f or -f60-10 (fold margin 10) F preserve nl\n");
4567 fprintf(stderr,"Z[0-3] Convert X0208 alphabet to ASCII 1: Kankaku to space,2: 2 spaces,\n");
4568 fprintf(stderr," 3: Convert HTML Entity\n");
4569 fprintf(stderr,"X,x Assume X0201 kana in MS-Kanji, -x preserves X0201\n");
4570 fprintf(stderr,"B[0-2] Broken input 0: missing ESC,1: any X on ESC-[($]-X,2: ASCII on NL\n");
4572 fprintf(stderr,"T Text mode output\n");
4574 fprintf(stderr,"O Output to File (DEFAULT 'nkf.out')\n");
4575 fprintf(stderr,"d,c Delete \\r in line feed and \\032, Add \\r in line feed\n");
4576 fprintf(stderr,"I Convert non ISO-2022-JP charactor to GETA\n");
4577 fprintf(stderr,"-L[uwm] line mode u:LF w:CRLF m:CR (DEFAULT noconversion)\n");
4578 fprintf(stderr,"long name options\n");
4579 fprintf(stderr," --fj,--unix,--mac,--windows convert for the system\n");
4580 fprintf(stderr," --jis,--euc,--sjis,--utf8,--utf16,--mime,--base64 convert for the code\n");
4581 fprintf(stderr," --hiragana, --katakana Hiragana/Katakana Conversion\n");
4582 fprintf(stderr," --cp932, --no-cp932 CP932 compatible\n");
4584 fprintf(stderr," --cap-input, --url-input Convert hex after ':' or '%%'\n");
4586 #ifdef NUMCHAR_OPTION
4587 fprintf(stderr," --numchar-input Convert Unicode Character Reference\n");
4589 #ifdef UTF8_OUTPUT_ENABLE
4590 fprintf(stderr," --ms-ucs-map Microsoft UCS Mapping Compatible\n");
4593 fprintf(stderr," --overwrite Overwrite original listed files by filtered result\n");
4595 fprintf(stderr," -g, --guess Guess the input code\n");
4596 fprintf(stderr," --help,--version\n");
4603 fprintf(stderr,"Network Kanji Filter Version %s (%s) "
4604 #if defined(MSDOS) && !defined(__WIN32__) && !defined(__WIN16__)
4607 #if defined(MSDOS) && defined(__WIN16__)
4610 #if defined(MSDOS) && defined(__WIN32__)
4616 ,NKF_VERSION,NKF_RELEASE_DATE);
4617 fprintf(stderr,"\n%s\n",CopyRight);
4622 **
\e$B%Q%C%A@):n<T
\e(B
4623 ** void@merope.pleiades.or.jp (Kusakabe Youichi)
4624 ** NIDE Naoyuki <nide@ics.nara-wu.ac.jp>
4625 ** ohta@src.ricoh.co.jp (Junn Ohta)
4626 ** inouet@strl.nhk.or.jp (Tomoyuki Inoue)
4627 ** kiri@pulser.win.or.jp (Tetsuaki Kiriyama)
4628 ** Kimihiko Sato <sato@sail.t.u-tokyo.ac.jp>
4629 ** a_kuroe@kuroe.aoba.yokohama.jp (Akihiko Kuroe)
4630 ** kono@ie.u-ryukyu.ac.jp (Shinji Kono)
4631 ** GHG00637@nifty-serve.or.jp (COW)