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 ***********************************************************************/
44 static char *CopyRight =
45 "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa),2000 S. Kono, COW, 2002-2004 Kono, Furukawa";
46 static char *Version =
48 static char *Patchlevel =
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__)
113 #if defined(MSDOS) || defined(__OS2__)
120 #define setbinmode(fp) fsetbin(fp)
121 #else /* Microsoft C, Turbo C */
122 #define setbinmode(fp) setmode(fileno(fp), O_BINARY)
124 #else /* UNIX,OS/2 */
125 #define setbinmode(fp)
128 #ifdef _IOFBF /* SysV and MSDOS, Windows */
129 #define setvbuffer(fp, buf, size) setvbuf(fp, buf, _IOFBF, size)
131 #define setvbuffer(fp, buf, size) setbuffer(fp, buf, size)
134 /*Borland C++ 4.5 EasyWin*/
135 #if defined(__TURBOC__) && defined(_Windows) && !defined(__WIN32__) /*Easy Win */
144 /* added by satoru@isoternet.org */
147 #include <sys/stat.h>
148 #ifndef MSDOS /* UNIX, OS/2 */
152 #if defined(_MSC_VER) || defined(__MINGW32__) /* VC++, MinGW */
153 #include <sys/utime.h>
154 #elif defined(__TURBOC__) /* BCC */
156 #elif defined(LSI_C) /* LSI C */
168 /* state of output_mode and input_mode
185 /* Input Assumption */
189 #define LATIN1_INPUT 6
191 #define STRICT_MIME 8
196 #define JAPANESE_EUC 10
200 #define UTF8_INPUT 13
201 #define UTF16LE_INPUT 14
202 #define UTF16BE_INPUT 15
220 #define is_alnum(c) \
221 (('a'<=c && c<='z')||('A'<= c && c<='Z')||('0'<=c && c<='9'))
223 #define HOLD_SIZE 1024
224 #define IOBUF_SIZE 16384
226 #define DEFAULT_J 'B'
227 #define DEFAULT_R 'B'
229 #define SJ0162 0x00e1 /* 01 - 62 ku offset */
230 #define SJ6394 0x0161 /* 63 - 94 ku offset */
232 #define RANGE_NUM_MAX 18
237 #if defined( UTF8_OUTPUT_ENABLE ) || defined( UTF8_INPUT_ENABLE )
238 #define sizeof_euc_utf8 94
239 #define sizeof_euc_to_utf8_1byte 94
240 #define sizeof_euc_to_utf8_2bytes 94
241 #define sizeof_utf8_to_euc_C2 64
242 #define sizeof_utf8_to_euc_E5B8 64
243 #define sizeof_utf8_to_euc_2bytes 112
244 #define sizeof_utf8_to_euc_3bytes 112
247 /* MIME preprocessor */
250 #ifdef EASYWIN /*Easy Win */
251 extern POINT _BufferSize;
254 /* function prototype */
256 #ifdef ANSI_C_PROTOTYPE
258 #define STATIC static
270 void (*status_func)PROTO((struct input_code *, int));
271 int (*iconv_func)PROTO((int c2, int c1, int c0));
275 STATIC char *input_codename = "";
277 STATIC int noconvert PROTO((FILE *f));
278 STATIC int kanji_convert PROTO((FILE *f));
279 STATIC int h_conv PROTO((FILE *f,int c2,int c1));
280 STATIC int push_hold_buf PROTO((int c2));
281 STATIC void set_iconv PROTO((int f, int (*iconv_func)()));
282 STATIC int s_iconv PROTO((int c2,int c1,int c0));
283 STATIC int s2e_conv PROTO((int c2, int c1, int *p2, int *p1));
284 STATIC int e_iconv PROTO((int c2,int c1,int c0));
285 #ifdef UTF8_INPUT_ENABLE
286 STATIC int w2e_conv PROTO((int c2,int c1,int c0,int *p2,int *p1));
287 STATIC int w_iconv PROTO((int c2,int c1,int c0));
288 STATIC int w_iconv16 PROTO((int c2,int c1,int c0));
289 STATIC int w_iconv_common PROTO((int c1,int c0,unsigned short **pp,int psize,int *p2,int *p1));
290 STATIC int ww16_conv PROTO((int c2, int c1, int c0));
292 #ifdef UTF8_OUTPUT_ENABLE
293 STATIC int e2w_conv PROTO((int c2,int c1));
294 STATIC void w_oconv PROTO((int c2,int c1));
295 STATIC void w_oconv16 PROTO((int c2,int c1));
297 STATIC void e_oconv PROTO((int c2,int c1));
298 STATIC void e2s_conv PROTO((int c2, int c1, int *p2, int *p1));
299 STATIC void s_oconv PROTO((int c2,int c1));
300 STATIC void j_oconv PROTO((int c2,int c1));
301 STATIC void fold_conv PROTO((int c2,int c1));
302 STATIC void cr_conv PROTO((int c2,int c1));
303 STATIC void z_conv PROTO((int c2,int c1));
304 STATIC void rot_conv PROTO((int c2,int c1));
305 STATIC void hira_conv PROTO((int c2,int c1));
306 STATIC void base64_conv PROTO((int c2,int c1));
307 STATIC void iso2022jp_check_conv PROTO((int c2,int c1));
308 STATIC void no_connection PROTO((int c2,int c1));
309 STATIC int no_connection2 PROTO((int c2,int c1,int c0));
311 STATIC void code_score PROTO((struct input_code *ptr));
312 STATIC void code_status PROTO((int c));
314 STATIC void std_putc PROTO((int c));
315 STATIC int std_getc PROTO((FILE *f));
316 STATIC int std_ungetc PROTO((int c,FILE *f));
318 STATIC int broken_getc PROTO((FILE *f));
319 STATIC int broken_ungetc PROTO((int c,FILE *f));
321 STATIC int mime_begin PROTO((FILE *f));
322 STATIC int mime_getc PROTO((FILE *f));
323 STATIC int mime_ungetc PROTO((int c,FILE *f));
325 STATIC int mime_begin_strict PROTO((FILE *f));
326 STATIC int mime_getc_buf PROTO((FILE *f));
327 STATIC int mime_ungetc_buf PROTO((int c,FILE *f));
328 STATIC int mime_integrity PROTO((FILE *f,unsigned char *p));
330 STATIC int base64decode PROTO((int c));
331 STATIC void mime_putc PROTO((int c));
332 STATIC void open_mime PROTO((int c));
333 STATIC void close_mime PROTO(());
334 STATIC void usage PROTO(());
335 STATIC void version PROTO(());
336 STATIC void options PROTO((unsigned char *c));
338 STATIC void reinit PROTO(());
343 static unsigned char stdibuf[IOBUF_SIZE];
344 static unsigned char stdobuf[IOBUF_SIZE];
345 static unsigned char hold_buf[HOLD_SIZE*2];
346 static int hold_count;
348 /* MIME preprocessor fifo */
350 #define MIME_BUF_SIZE (1024) /* 2^n ring buffer */
351 #define MIME_BUF_MASK (MIME_BUF_SIZE-1)
352 #define Fifo(n) mime_buf[(n)&MIME_BUF_MASK]
353 static unsigned char mime_buf[MIME_BUF_SIZE];
354 static unsigned int mime_top = 0;
355 static unsigned int mime_last = 0; /* decoded */
356 static unsigned int mime_input = 0; /* undecoded */
359 static int unbuf_f = FALSE;
360 static int estab_f = FALSE;
361 static int nop_f = FALSE;
362 static int binmode_f = TRUE; /* binary mode */
363 static int rot_f = FALSE; /* rot14/43 mode */
364 static int hira_f = FALSE; /* hira/kata henkan */
365 static int input_f = FALSE; /* non fixed input code */
366 static int alpha_f = FALSE; /* convert JIx0208 alphbet to ASCII */
367 static int mime_f = STRICT_MIME; /* convert MIME B base64 or Q */
368 static int mimebuf_f = FALSE; /* MIME buffered input */
369 static int broken_f = FALSE; /* convert ESC-less broken JIS */
370 static int iso8859_f = FALSE; /* ISO8859 through */
371 static int mimeout_f = FALSE; /* base64 mode */
372 #if defined(MSDOS) || defined(__OS2__)
373 static int x0201_f = TRUE; /* Assume JISX0201 kana */
375 static int x0201_f = NO_X0201; /* Assume NO JISX0201 */
377 static int iso2022jp_f = FALSE; /* convert ISO-2022-JP */
378 #ifdef UTF8_OUTPUT_ENABLE
379 static int unicode_bom_f= 0; /* Output Unicode BOM */
380 static int w_oconv16_LE = 0; /* utf-16 little endian */
381 static int ms_ucs_map_f = FALSE; /* Microsoft UCS Mapping Compatible */
385 #ifdef NUMCHAR_OPTION
387 #define CLASS_MASK 0x0f000000
388 #define CLASS_UTF16 0x01000000
392 static int cap_f = FALSE;
393 static int (*i_cgetc)PROTO((FILE *)) = std_getc; /* input of cgetc */
394 static int (*i_cungetc)PROTO((int c ,FILE *f)) = std_ungetc;
395 STATIC int cap_getc PROTO((FILE *f));
396 STATIC int cap_ungetc PROTO((int c,FILE *f));
398 static int url_f = FALSE;
399 static int (*i_ugetc)PROTO((FILE *)) = std_getc; /* input of ugetc */
400 static int (*i_uungetc)PROTO((int c ,FILE *f)) = std_ungetc;
401 STATIC int url_getc PROTO((FILE *f));
402 STATIC int url_ungetc PROTO((int c,FILE *f));
404 static int numchar_f = FALSE;
405 static int (*i_ngetc)PROTO((FILE *)) = std_getc; /* input of ugetc */
406 static int (*i_nungetc)PROTO((int c ,FILE *f)) = std_ungetc;
407 STATIC int numchar_getc PROTO((FILE *f));
408 STATIC int numchar_ungetc PROTO((int c,FILE *f));
412 static int noout_f = FALSE;
413 STATIC void no_putc PROTO((int c));
414 static int debug_f = FALSE;
415 STATIC void debug PROTO((char *str));
418 static int guess_f = FALSE;
419 STATIC void print_guessed_code PROTO((char *filename));
420 STATIC void set_input_codename PROTO((char *codename));
421 static int is_inputcode_mixed = FALSE;
422 static int is_inputcode_set = FALSE;
425 static int exec_f = 0;
428 #ifdef SHIFTJIS_CP932
429 STATIC int cp932_f = TRUE;
430 #define CP932_TABLE_BEGIN (0xfa)
431 #define CP932_TABLE_END (0xfc)
433 STATIC int cp932inv_f = FALSE;
434 #define CP932INV_TABLE_BEGIN (0xed)
435 #define CP932INV_TABLE_END (0xee)
437 #endif /* SHIFTJIS_CP932 */
439 STATIC unsigned char prefix_table[256];
441 STATIC void e_status PROTO((struct input_code *, int));
442 STATIC void s_status PROTO((struct input_code *, int));
444 #ifdef UTF8_INPUT_ENABLE
445 STATIC void w_status PROTO((struct input_code *, int));
446 STATIC void w16_status PROTO((struct input_code *, int));
447 static int utf16_mode = UTF16LE_INPUT;
450 struct input_code input_code_list[] = {
451 {"EUC-JP", 0, 0, 0, {0, 0, 0}, e_status, e_iconv, 0},
452 {"Shift_JIS", 0, 0, 0, {0, 0, 0}, s_status, s_iconv, 0},
453 {"UTF-8", 0, 0, 0, {0, 0, 0}, w_status, w_iconv, 0},
454 {"UTF-16", 0, 0, 0, {0, 0, 0}, w16_status, w_iconv16, 0},
458 static int mimeout_mode = 0;
459 static int base64_count = 0;
461 /* X0208 -> ASCII converter */
464 static int f_line = 0; /* chars in line */
465 static int f_prev = 0;
466 static int fold_preserve_f = FALSE; /* preserve new lines */
467 static int fold_f = FALSE;
468 static int fold_len = 0;
471 static unsigned char kanji_intro = DEFAULT_J,
472 ascii_intro = DEFAULT_R;
476 #define FOLD_MARGIN 10
477 #define DEFAULT_FOLD 60
479 static int fold_margin = FOLD_MARGIN;
483 #ifdef DEFAULT_CODE_JIS
484 # define DEFAULT_CONV j_oconv
486 #ifdef DEFAULT_CODE_SJIS
487 # define DEFAULT_CONV s_oconv
489 #ifdef DEFAULT_CODE_EUC
490 # define DEFAULT_CONV e_oconv
492 #ifdef DEFAULT_CODE_UTF8
493 # define DEFAULT_CONV w_oconv
496 /* process default */
497 static void (*output_conv)PROTO((int c2,int c1)) = DEFAULT_CONV;
499 static void (*oconv)PROTO((int c2,int c1)) = no_connection;
500 /* s_iconv or oconv */
501 static int (*iconv)PROTO((int c2,int c1,int c0)) = no_connection2;
503 static void (*o_zconv)PROTO((int c2,int c1)) = no_connection;
504 static void (*o_fconv)PROTO((int c2,int c1)) = no_connection;
505 static void (*o_crconv)PROTO((int c2,int c1)) = no_connection;
506 static void (*o_rot_conv)PROTO((int c2,int c1)) = no_connection;
507 static void (*o_hira_conv)PROTO((int c2,int c1)) = no_connection;
508 static void (*o_base64conv)PROTO((int c2,int c1)) = no_connection;
509 static void (*o_iso2022jp_check_conv)PROTO((int c2,int c1)) = no_connection;
511 /* static redirections */
513 static void (*o_putc)PROTO((int c)) = std_putc;
515 static int (*i_getc)PROTO((FILE *f)) = std_getc; /* general input */
516 static int (*i_ungetc)PROTO((int c,FILE *f)) =std_ungetc;
518 static int (*i_bgetc)PROTO((FILE *)) = std_getc; /* input of mgetc */
519 static int (*i_bungetc)PROTO((int c ,FILE *f)) = std_ungetc;
521 static void (*o_mputc)PROTO((int c)) = std_putc ; /* output of mputc */
523 static int (*i_mgetc)PROTO((FILE *)) = std_getc; /* input of mgetc */
524 static int (*i_mungetc)PROTO((int c ,FILE *f)) = std_ungetc;
526 /* for strict mime */
527 static int (*i_mgetc_buf)PROTO((FILE *)) = std_getc; /* input of mgetc_buf */
528 static int (*i_mungetc_buf)PROTO((int c,FILE *f)) = std_ungetc;
531 static int output_mode = ASCII, /* output kanji mode */
532 input_mode = ASCII, /* input kanji mode */
533 shift_mode = FALSE; /* TRUE shift out, or X0201 */
534 static int mime_decode_mode = FALSE; /* MIME mode B base64, Q hex */
536 /* X0201 / X0208 conversion tables */
538 /* X0201 kana conversion table */
541 unsigned char cv[]= {
542 0x21,0x21,0x21,0x23,0x21,0x56,0x21,0x57,
543 0x21,0x22,0x21,0x26,0x25,0x72,0x25,0x21,
544 0x25,0x23,0x25,0x25,0x25,0x27,0x25,0x29,
545 0x25,0x63,0x25,0x65,0x25,0x67,0x25,0x43,
546 0x21,0x3c,0x25,0x22,0x25,0x24,0x25,0x26,
547 0x25,0x28,0x25,0x2a,0x25,0x2b,0x25,0x2d,
548 0x25,0x2f,0x25,0x31,0x25,0x33,0x25,0x35,
549 0x25,0x37,0x25,0x39,0x25,0x3b,0x25,0x3d,
550 0x25,0x3f,0x25,0x41,0x25,0x44,0x25,0x46,
551 0x25,0x48,0x25,0x4a,0x25,0x4b,0x25,0x4c,
552 0x25,0x4d,0x25,0x4e,0x25,0x4f,0x25,0x52,
553 0x25,0x55,0x25,0x58,0x25,0x5b,0x25,0x5e,
554 0x25,0x5f,0x25,0x60,0x25,0x61,0x25,0x62,
555 0x25,0x64,0x25,0x66,0x25,0x68,0x25,0x69,
556 0x25,0x6a,0x25,0x6b,0x25,0x6c,0x25,0x6d,
557 0x25,0x6f,0x25,0x73,0x21,0x2b,0x21,0x2c,
561 /* X0201 kana conversion table for daguten */
564 unsigned char dv[]= {
565 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
566 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
567 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
568 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
569 0x00,0x00,0x00,0x00,0x00,0x00,0x25,0x74,
570 0x00,0x00,0x00,0x00,0x25,0x2c,0x25,0x2e,
571 0x25,0x30,0x25,0x32,0x25,0x34,0x25,0x36,
572 0x25,0x38,0x25,0x3a,0x25,0x3c,0x25,0x3e,
573 0x25,0x40,0x25,0x42,0x25,0x45,0x25,0x47,
574 0x25,0x49,0x00,0x00,0x00,0x00,0x00,0x00,
575 0x00,0x00,0x00,0x00,0x25,0x50,0x25,0x53,
576 0x25,0x56,0x25,0x59,0x25,0x5c,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 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
583 /* X0201 kana conversion table for han-daguten */
586 unsigned char ev[]= {
587 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
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,
592 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
593 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
594 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
595 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
596 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
597 0x00,0x00,0x00,0x00,0x25,0x51,0x25,0x54,
598 0x25,0x57,0x25,0x5a,0x25,0x5d,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,
606 /* X0208 kigou conversion table */
607 /* 0x8140 - 0x819e */
609 unsigned char fv[] = {
611 0x00,0x00,0x00,0x00,0x2c,0x2e,0x00,0x3a,
612 0x3b,0x3f,0x21,0x00,0x00,0x27,0x60,0x00,
613 0x5e,0x00,0x5f,0x00,0x00,0x00,0x00,0x00,
614 0x00,0x00,0x00,0x00,0x00,0x2d,0x00,0x2f,
615 0x5c,0x00,0x00,0x7c,0x00,0x00,0x60,0x27,
616 0x22,0x22,0x28,0x29,0x00,0x00,0x5b,0x5d,
617 0x7b,0x7d,0x3c,0x3e,0x00,0x00,0x00,0x00,
618 0x00,0x00,0x00,0x00,0x2b,0x2d,0x00,0x00,
619 0x00,0x3d,0x00,0x3c,0x3e,0x00,0x00,0x00,
620 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
621 0x24,0x00,0x00,0x25,0x23,0x26,0x2a,0x40,
622 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00
628 static int file_out = FALSE;
630 static int overwrite = FALSE;
633 static int crmode_f = 0; /* CR, NL, CRLF */
634 #ifdef EASYWIN /*Easy Win */
635 static int end_check;
650 #ifdef EASYWIN /*Easy Win */
651 _BufferSize.y = 400;/*Set Scroll Buffer Size*/
654 for (argc--,argv++; (argc > 0) && **argv == '-'; argc--, argv++) {
655 cp = (unsigned char *)*argv;
660 if (pipe(fds) < 0 || (pid = fork()) < 0){
671 execvp(argv[1], &argv[1]);
685 if(x0201_f == WISH_TRUE)
686 x0201_f = ((!iso2022jp_f)? TRUE : NO_X0201);
688 if (binmode_f == TRUE)
690 if (freopen("","wb",stdout) == NULL)
697 setbuf(stdout, (char *) NULL);
699 setvbuffer(stdout, stdobuf, IOBUF_SIZE);
702 if (binmode_f == TRUE)
704 if (freopen("","rb",stdin) == NULL) return (-1);
708 setvbuffer(stdin, stdibuf, IOBUF_SIZE);
712 kanji_convert(stdin);
713 if (guess_f) print_guessed_code(NULL);
718 if ((fin = fopen((origfname = *argv++), "r")) == NULL) {
727 /* reopen file for stdout */
728 if (file_out == TRUE) {
731 outfname = malloc(strlen(origfname)
732 + strlen(".nkftmpXXXXXX")
738 strcpy(outfname, origfname);
742 for (i = strlen(outfname); i; --i){
743 if (outfname[i - 1] == '/'
744 || outfname[i - 1] == '\\'){
750 strcat(outfname, "ntXXXXXX");
752 fd = open(outfname, O_WRONLY | O_CREAT | O_TRUNC,
755 strcat(outfname, ".nkftmpXXXXXX");
756 fd = mkstemp(outfname);
759 || (fd_backup = dup(fileno(stdout))) < 0
760 || dup2(fd, fileno(stdout)) < 0
771 outfname = "nkf.out";
774 if(freopen(outfname, "w", stdout) == NULL) {
778 if (binmode_f == TRUE) {
780 if (freopen("","wb",stdout) == NULL)
787 if (binmode_f == TRUE)
789 if (freopen("","rb",fin) == NULL)
794 setvbuffer(fin, stdibuf, IOBUF_SIZE);
798 char *filename = NULL;
800 if (nfiles > 1) filename = origfname;
801 if (guess_f) print_guessed_code(filename);
807 #if defined(MSDOS) && !defined(__MINGW32__)
815 if (dup2(fd_backup, fileno(stdout)) < 0){
818 if (stat(origfname, &sb)) {
819 fprintf(stderr, "Can't stat %s\n", origfname);
821 /*
\e$B%Q!<%_%C%7%g%s$rI|85
\e(B */
822 if (chmod(outfname, sb.st_mode)) {
823 fprintf(stderr, "Can't set permission %s\n", outfname);
826 /*
\e$B%?%$%`%9%?%s%W$rI|85
\e(B */
827 #if defined(MSDOS) && !defined(__MINGW32__)
828 tb[0] = tb[1] = sb.st_mtime;
829 if (utime(outfname, tb)) {
830 fprintf(stderr, "Can't set timestamp %s\n", outfname);
833 tb.actime = sb.st_atime;
834 tb.modtime = sb.st_mtime;
835 if (utime(outfname, &tb)) {
836 fprintf(stderr, "Can't set timestamp %s\n", outfname);
840 if (unlink(origfname)){
844 if (rename(outfname, origfname)) {
846 fprintf(stderr, "Can't rename %s to %s\n",
847 outfname, origfname);
855 #ifdef EASYWIN /*Easy Win */
856 if (file_out == FALSE)
857 scanf("%d",&end_check);
860 #else /* for Other OS */
861 if (file_out == TRUE)
891 {"katakana-hiragana","h3"},
893 #ifdef UTF8_OUTPUT_ENABLE
898 #ifdef UTF8_INPUT_ENABLE
900 {"utf16-input", "W16"},
909 #ifdef NUMCHAR_OPTION
910 {"numchar-input", ""},
916 #ifdef SHIFTJIS_CP932
927 static int option_mode;
942 case '-': /* literal options */
943 if (!*cp) { /* ignore the rest of arguments */
947 for (i=0;i<sizeof(long_option)/sizeof(long_option[0]);i++) {
949 p = (unsigned char *)long_option[i].name;
950 for (j=0;*p && (*p != '=') && *p == cp[j];p++, j++);
958 cp = (unsigned char *)long_option[i].alias;
961 if (strcmp(long_option[i].name, "overwrite") == 0){
968 if (strcmp(long_option[i].name, "cap-input") == 0){
972 if (strcmp(long_option[i].name, "url-input") == 0){
977 #ifdef NUMCHAR_OPTION
978 if (strcmp(long_option[i].name, "numchar-input") == 0){
984 if (strcmp(long_option[i].name, "no-output") == 0){
988 if (strcmp(long_option[i].name, "debug") == 0){
993 #ifdef SHIFTJIS_CP932
994 if (strcmp(long_option[i].name, "no-cp932") == 0){
998 if (strcmp(long_option[i].name, "cp932inv") == 0){
1004 if (strcmp(long_option[i].name, "exec-in") == 0){
1008 if (strcmp(long_option[i].name, "exec-out") == 0){
1013 #ifdef UTF8_OUTPUT_ENABLE
1014 if (strcmp(long_option[i].name, "ms-ucs-map") == 0){
1015 ms_ucs_map_f = TRUE;
1019 if (strcmp(long_option[i].name, "prefix=") == 0){
1020 if (*p == '=' && ' ' < p[1] && p[1] < 128){
1021 for (i = 2; ' ' < p[i] && p[i] < 128; i++){
1022 prefix_table[p[i]] = p[1];
1029 case 'b': /* buffered mode */
1032 case 'u': /* non bufferd mode */
1035 case 't': /* transparent mode */
1038 case 'j': /* JIS output */
1040 output_conv = j_oconv;
1042 case 'e': /* AT&T EUC output */
1043 output_conv = e_oconv;
1045 case 's': /* SJIS output */
1046 output_conv = s_oconv;
1048 case 'l': /* ISO8859 Latin-1 support, no conversion */
1049 iso8859_f = TRUE; /* Only compatible with ISO-2022-JP */
1050 input_f = LATIN1_INPUT;
1052 case 'i': /* Kanji IN ESC-$-@/B */
1053 if (*cp=='@'||*cp=='B')
1054 kanji_intro = *cp++;
1056 case 'o': /* ASCII IN ESC-(-J/B */
1057 if (*cp=='J'||*cp=='B'||*cp=='H')
1058 ascii_intro = *cp++;
1065 if ('9'>= *cp && *cp>='0')
1066 hira_f |= (*cp++ -'0');
1073 #if defined(MSDOS) || defined(__OS2__)
1088 #ifdef UTF8_OUTPUT_ENABLE
1089 case 'w': /* UTF-8 output */
1090 if ('1'== cp[0] && '6'==cp[1]) {
1091 output_conv = w_oconv16; cp+=2;
1093 unicode_bom_f=2; cp++;
1096 unicode_bom_f=1; cp++;
1098 } else if (cp[0] == 'B') {
1099 unicode_bom_f=2; cp++;
1101 unicode_bom_f=1; cp++;
1104 } else if (cp[0] == '8') {
1105 output_conv = w_oconv; cp++;
1108 unicode_bom_f=1; cp++;
1111 output_conv = w_oconv;
1114 #ifdef UTF8_INPUT_ENABLE
1115 case 'W': /* UTF-8 input */
1116 if ('1'== cp[0] && '6'==cp[1]) {
1117 input_f = UTF16LE_INPUT;
1120 } else if (cp[0] == 'B') {
1122 input_f = UTF16BE_INPUT;
1125 input_f = UTF8_INPUT;
1128 /* Input code assumption */
1129 case 'J': /* JIS input */
1130 case 'E': /* AT&T EUC input */
1131 input_f = JIS_INPUT;
1133 case 'S': /* MS Kanji input */
1134 input_f = SJIS_INPUT;
1135 if (x0201_f==NO_X0201) x0201_f=TRUE;
1137 case 'Z': /* Convert X0208 alphabet to asii */
1138 /* bit:0 Convert X0208
1139 bit:1 Convert Kankaku to one space
1140 bit:2 Convert Kankaku to two spaces
1141 bit:3 Convert HTML Entity
1143 if ('9'>= *cp && *cp>='0')
1144 alpha_f |= 1<<(*cp++ -'0');
1148 case 'x': /* Convert X0201 kana to X0208 or X0201 Conversion */
1149 x0201_f = FALSE; /* No X0201->X0208 conversion */
1151 ESC-(-I in JIS, EUC, MS Kanji
1152 SI/SO in JIS, EUC, MS Kanji
1153 SSO in EUC, JIS, not in MS Kanji
1154 MS Kanji (0xa0-0xdf)
1156 ESC-(-I in JIS (0x20-0x5f)
1157 SSO in EUC (0xa0-0xdf)
1158 0xa0-0xd in MS Kanji (0xa0-0xdf)
1161 case 'X': /* Assume X0201 kana */
1162 /* Default value is NO_X0201 for EUC/MS-Kanji mix */
1165 case 'F': /* prserve new lines */
1166 fold_preserve_f = TRUE;
1167 case 'f': /* folding -f60 or -f */
1170 while('0'<= *cp && *cp <='9') { /* we don't use atoi here */
1172 fold_len += *cp++ - '0';
1174 if (!(0<fold_len && fold_len<BUFSIZ))
1175 fold_len = DEFAULT_FOLD;
1179 while('0'<= *cp && *cp <='9') { /* we don't use atoi here */
1181 fold_margin += *cp++ - '0';
1185 case 'm': /* MIME support */
1186 if (*cp=='B'||*cp=='Q') {
1187 mime_decode_mode = *cp++;
1188 mimebuf_f = FIXED_MIME;
1189 } else if (*cp=='N') {
1190 mime_f = TRUE; cp++;
1191 } else if (*cp=='S') {
1192 mime_f = STRICT_MIME; cp++;
1193 } else if (*cp=='0') {
1194 mime_f = FALSE; cp++;
1197 case 'M': /* MIME output */
1200 mimeout_f = FIXED_MIME; cp++;
1201 } else if (*cp=='Q') {
1203 mimeout_f = FIXED_MIME; cp++;
1208 case 'B': /* Broken JIS support */
1210 bit:1 allow any x on ESC-(-x or ESC-$-x
1211 bit:2 reset to ascii on NL
1213 if ('9'>= *cp && *cp>='0')
1214 broken_f |= 1<<(*cp++ -'0');
1219 case 'O':/* for Output file */
1223 case 'c':/* add cr code */
1226 case 'd':/* delete cr code */
1229 case 'I': /* ISO-2022-JP output */
1232 case 'L': /* line mode */
1233 if (*cp=='u') { /* unix */
1234 crmode_f = NL; cp++;
1235 } else if (*cp=='m') { /* mac */
1236 crmode_f = CR; cp++;
1237 } else if (*cp=='w') { /* windows */
1238 crmode_f = CRLF; cp++;
1239 } else if (*cp=='0') { /* no conversion */
1249 /* module muliple options in a string are allowed for Perl moudle */
1250 while(*cp && *cp!='-') cp++;
1254 /* bogus option but ignored */
1260 #ifdef ANSI_C_PROTOTYPE
1261 struct input_code * find_inputcode_byfunc(int (*iconv_func)(int c2,int c1,int c0))
1263 struct input_code * find_inputcode_byfunc(iconv_func)
1264 int (*iconv_func)();
1268 struct input_code *p = input_code_list;
1270 if (iconv_func == p->iconv_func){
1279 #ifdef ANSI_C_PROTOTYPE
1280 void set_iconv(int f, int (*iconv_func)(int c2,int c1,int c0))
1282 void set_iconv(f, iconv_func)
1284 int (*iconv_func)();
1288 static int (*iconv_for_check)() = 0;
1290 #ifdef INPUT_CODE_FIX
1298 #ifdef INPUT_CODE_FIX
1299 && (f == -TRUE || !input_f) /* -TRUE means "FORCE" */
1305 if (estab_f && iconv_for_check != iconv){
1306 struct input_code *p = find_inputcode_byfunc(iconv);
1308 set_input_codename(p->name);
1309 debug(input_codename);
1311 iconv_for_check = iconv;
1316 #define SCORE_L2 (1) /*
\e$BBh
\e(B2
\e$B?e=`4A;z
\e(B */
1317 #define SCORE_KANA (SCORE_L2 << 1) /*
\e$B$$$o$f$kH>3Q%+%J
\e(B */
1318 #define SCORE_DEPEND (SCORE_KANA << 1) /*
\e$B5!<o0MB8J8;z
\e(B */
1319 #ifdef SHIFTJIS_CP932
1320 #define SCORE_CP932 (SCORE_DEPEND << 1) /* CP932
\e$B$K$h$kFI$_49$(
\e(B */
1321 #define SCORE_NO_EXIST (SCORE_CP932 << 1) /*
\e$BB8:_$7$J$$J8;z
\e(B */
1323 #define SCORE_NO_EXIST (SCORE_DEPEND << 1) /*
\e$BB8:_$7$J$$J8;z
\e(B */
1325 #define SCORE_iMIME (SCORE_NO_EXIST << 1) /* MIME
\e$B$K$h$k;XDj
\e(B */
1326 #define SCORE_ERROR (SCORE_iMIME << 1) /*
\e$B%(%i!<
\e(B */
1328 #define SCORE_INIT (SCORE_iMIME)
1330 int score_table_A0[] = {
1333 0, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND,
1334 SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_NO_EXIST,
1337 int score_table_F0[] = {
1338 SCORE_L2, SCORE_L2, SCORE_L2, SCORE_L2,
1339 SCORE_L2, SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST,
1340 SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND,
1341 SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST, SCORE_ERROR,
1344 void set_code_score(ptr, score)
1345 struct input_code *ptr;
1349 ptr->score |= score;
1353 void clr_code_score(ptr, score)
1354 struct input_code *ptr;
1358 ptr->score &= ~score;
1362 void code_score(ptr)
1363 struct input_code *ptr;
1365 int c2 = ptr->buf[0];
1366 int c1 = ptr->buf[1];
1368 set_code_score(ptr, SCORE_ERROR);
1369 }else if (c2 == SSO){
1370 set_code_score(ptr, SCORE_KANA);
1371 #ifdef UTF8_OUTPUT_ENABLE
1372 }else if (!e2w_conv(c2, c1)){
1373 set_code_score(ptr, SCORE_NO_EXIST);
1375 }else if ((c2 & 0x70) == 0x20){
1376 set_code_score(ptr, score_table_A0[c2 & 0x0f]);
1377 }else if ((c2 & 0x70) == 0x70){
1378 set_code_score(ptr, score_table_F0[c2 & 0x0f]);
1379 }else if ((c2 & 0x70) >= 0x50){
1380 set_code_score(ptr, SCORE_L2);
1384 void status_disable(ptr)
1385 struct input_code *ptr;
1390 if (iconv == ptr->iconv_func) set_iconv(FALSE, 0);
1393 void status_push_ch(ptr, c)
1394 struct input_code *ptr;
1397 ptr->buf[ptr->index++] = c;
1400 void status_clear(ptr)
1401 struct input_code *ptr;
1407 void status_reset(ptr)
1408 struct input_code *ptr;
1411 ptr->score = SCORE_INIT;
1414 void status_reinit(ptr)
1415 struct input_code *ptr;
1418 ptr->_file_stat = 0;
1421 void status_check(ptr, c)
1422 struct input_code *ptr;
1425 if (c <= DEL && estab_f){
1430 void s_status(ptr, c)
1431 struct input_code *ptr;
1436 status_check(ptr, c);
1441 #ifdef NUMCHAR_OPTION
1442 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1445 }else if (0xa1 <= c && c <= 0xdf){
1446 status_push_ch(ptr, SSO);
1447 status_push_ch(ptr, c);
1450 }else if ((0x81 <= c && c < 0xa0) || (0xe0 <= c && c <= 0xef)){
1452 status_push_ch(ptr, c);
1453 #ifdef SHIFTJIS_CP932
1455 && CP932_TABLE_BEGIN <= c && c <= CP932_TABLE_END){
1457 status_push_ch(ptr, c);
1458 #endif /* SHIFTJIS_CP932 */
1460 status_disable(ptr);
1464 if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)){
1465 status_push_ch(ptr, c);
1466 s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]);
1470 status_disable(ptr);
1473 #ifdef SHIFTJIS_CP932
1475 if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)){
1476 status_push_ch(ptr, c);
1477 if (s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]) == 0){
1478 set_code_score(ptr, SCORE_CP932);
1483 status_disable(ptr);
1485 #endif /* SHIFTJIS_CP932 */
1489 void e_status(ptr, c)
1490 struct input_code *ptr;
1495 status_check(ptr, c);
1500 #ifdef NUMCHAR_OPTION
1501 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1504 }else if (SSO == c || (0xa1 <= c && c <= 0xfe)){
1506 status_push_ch(ptr, c);
1508 status_disable(ptr);
1512 if (0xa1 <= c && c <= 0xfe){
1513 status_push_ch(ptr, c);
1517 status_disable(ptr);
1523 #ifdef UTF8_INPUT_ENABLE
1524 void w16_status(ptr, c)
1525 struct input_code *ptr;
1532 if (ptr->_file_stat == 0){
1533 if (c == 0xfe || c == 0xff){
1535 status_push_ch(ptr, c);
1536 ptr->_file_stat = 1;
1538 status_disable(ptr);
1539 ptr->_file_stat = -1;
1541 }else if (ptr->_file_stat > 0){
1543 status_push_ch(ptr, c);
1544 }else if (ptr->_file_stat < 0){
1545 status_disable(ptr);
1551 status_disable(ptr);
1552 ptr->_file_stat = -1;
1554 status_push_ch(ptr, c);
1561 if (ptr->stat != c && (c == 0xfe || c == 0xff)){
1562 status_push_ch(ptr, c);
1565 status_disable(ptr);
1566 ptr->_file_stat = -1;
1572 void w_status(ptr, c)
1573 struct input_code *ptr;
1578 status_check(ptr, c);
1583 #ifdef NUMCHAR_OPTION
1584 }else if ((c & CLASS_MASK) == CLASS_UTF16){
1587 }else if (0xc0 <= c && c <= 0xdf){
1589 status_push_ch(ptr, c);
1590 }else if (0xe0 <= c && c <= 0xef){
1592 status_push_ch(ptr, c);
1594 status_disable(ptr);
1599 if (0x80 <= c && c <= 0xbf){
1600 status_push_ch(ptr, c);
1601 if (ptr->index > ptr->stat){
1602 int bom = (ptr->buf[0] == 0xef && ptr->buf[1] == 0xbb
1603 && ptr->buf[2] == 0xbf);
1604 w2e_conv(ptr->buf[0], ptr->buf[1], ptr->buf[2],
1605 &ptr->buf[0], &ptr->buf[1]);
1612 status_disable(ptr);
1623 int action_flag = 1;
1624 struct input_code *result = 0;
1625 struct input_code *p = input_code_list;
1627 (p->status_func)(p, c);
1630 }else if(p->stat == 0){
1641 if (result && !estab_f){
1642 set_iconv(TRUE, result->iconv_func);
1643 }else if (c <= DEL){
1644 struct input_code *ptr = input_code_list;
1654 #define STD_GC_BUFSIZE (256)
1655 int std_gc_buf[STD_GC_BUFSIZE];
1665 return std_gc_buf[--std_gc_ndx];
1677 if (std_gc_ndx == STD_GC_BUFSIZE){
1680 std_gc_buf[std_gc_ndx++] = c;
1700 while ((c = (*i_getc)(f)) != EOF)
1709 oconv = output_conv;
1712 /* replace continucation module, from output side */
1714 /* output redicrection */
1716 if (noout_f || guess_f){
1723 if (mimeout_f == TRUE) {
1724 o_base64conv = oconv; oconv = base64_conv;
1726 /* base64_count = 0; */
1730 o_crconv = oconv; oconv = cr_conv;
1733 o_rot_conv = oconv; oconv = rot_conv;
1736 o_iso2022jp_check_conv = oconv; oconv = iso2022jp_check_conv;
1739 o_hira_conv = oconv; oconv = hira_conv;
1742 o_fconv = oconv; oconv = fold_conv;
1745 if (alpha_f || x0201_f) {
1746 o_zconv = oconv; oconv = z_conv;
1750 i_ungetc = std_ungetc;
1751 /* input redicrection */
1754 i_cgetc = i_getc; i_getc = cap_getc;
1755 i_cungetc = i_ungetc; i_ungetc= cap_ungetc;
1758 i_ugetc = i_getc; i_getc = url_getc;
1759 i_uungetc = i_ungetc; i_ungetc= url_ungetc;
1762 #ifdef NUMCHAR_OPTION
1764 i_ngetc = i_getc; i_getc = numchar_getc;
1765 i_nungetc = i_ungetc; i_ungetc= numchar_ungetc;
1768 if (mime_f && mimebuf_f==FIXED_MIME) {
1769 i_mgetc = i_getc; i_getc = mime_getc;
1770 i_mungetc = i_ungetc; i_ungetc = mime_ungetc;
1773 i_bgetc = i_getc; i_getc = broken_getc;
1774 i_bungetc = i_ungetc; i_ungetc = broken_ungetc;
1776 if (input_f == JIS_INPUT || input_f == LATIN1_INPUT) {
1777 set_iconv(-TRUE, e_iconv);
1778 } else if (input_f == SJIS_INPUT) {
1779 set_iconv(-TRUE, s_iconv);
1780 #ifdef UTF8_INPUT_ENABLE
1781 } else if (input_f == UTF8_INPUT) {
1782 set_iconv(-TRUE, w_iconv);
1783 } else if (input_f == UTF16LE_INPUT) {
1784 set_iconv(-TRUE, w_iconv16);
1787 set_iconv(FALSE, e_iconv);
1791 struct input_code *p = input_code_list;
1799 Conversion main loop. Code detection only.
1809 module_connection();
1814 output_mode = ASCII;
1817 #define NEXT continue /* no output, get next */
1818 #define SEND ; /* output c1 and c2, get next */
1819 #define LAST break /* end of loop, go closing */
1821 while ((c1 = (*i_getc)(f)) != EOF) {
1826 /* in case of 8th bit is on */
1828 /* in case of not established yet */
1829 /* It is still ambiguious */
1830 if (h_conv(f, c2, c1)==EOF)
1836 /* in case of already established */
1838 /* ignore bogus code */
1844 /* second byte, 7 bit code */
1845 /* it might be kanji shitfted */
1846 if ((c1 == DEL) || (c1 <= SPACE)) {
1847 /* ignore bogus first code */
1855 #ifdef UTF8_INPUT_ENABLE
1864 #ifdef NUMCHAR_OPTION
1865 } else if ((c1 & CLASS_MASK) == CLASS_UTF16){
1868 } else if (c1 > DEL) {
1870 if (!estab_f && !iso8859_f) {
1871 /* not established yet */
1874 } else { /* estab_f==TRUE */
1879 } else if (SSP<=c1 && c1<0xe0 && iconv == s_iconv) {
1880 /* SJIS X0201 Case... */
1881 if(iso2022jp_f && x0201_f==NO_X0201) {
1882 (*oconv)(GETA1, GETA2);
1889 } else if (c1==SSO && iconv != s_iconv) {
1890 /* EUC X0201 Case */
1891 c1 = (*i_getc)(f); /* skip SSO */
1893 if (SSP<=c1 && c1<0xe0) {
1894 if(iso2022jp_f && x0201_f==NO_X0201) {
1895 (*oconv)(GETA1, GETA2);
1902 } else { /* bogus code, skip SSO and one byte */
1906 /* already established */
1911 } else if ((c1 > SPACE) && (c1 != DEL)) {
1912 /* in case of Roman characters */
1914 /* output 1 shifted byte */
1918 } else if (SPACE<=c1 && c1<(0xe0&0x7f) ){
1919 /* output 1 shifted byte */
1920 if(iso2022jp_f && x0201_f==NO_X0201) {
1921 (*oconv)(GETA1, GETA2);
1928 /* look like bogus code */
1931 } else if (input_mode == X0208) {
1932 /* in case of Kanji shifted */
1935 } else if (c1 == '=' && mime_f && !mime_decode_mode ) {
1936 /* Check MIME code */
1937 if ((c1 = (*i_getc)(f)) == EOF) {
1940 } else if (c1 == '?') {
1941 /* =? is mime conversion start sequence */
1942 if(mime_f == STRICT_MIME) {
1943 /* check in real detail */
1944 if (mime_begin_strict(f) == EOF)
1948 } else if (mime_begin(f) == EOF)
1958 /* normal ASCII code */
1961 } else if (c1 == SI) {
1964 } else if (c1 == SO) {
1967 } else if (c1 == ESC ) {
1968 if ((c1 = (*i_getc)(f)) == EOF) {
1969 /* (*oconv)(0, ESC); don't send bogus code */
1971 } else if (c1 == '$') {
1972 if ((c1 = (*i_getc)(f)) == EOF) {
1974 (*oconv)(0, ESC); don't send bogus code
1975 (*oconv)(0, '$'); */
1977 } else if (c1 == '@'|| c1 == 'B') {
1978 /* This is kanji introduction */
1981 set_input_codename("ISO-2022-JP");
1982 debug(input_codename);
1984 } else if (c1 == '(') {
1985 if ((c1 = (*i_getc)(f)) == EOF) {
1986 /* don't send bogus code
1992 } else if (c1 == '@'|| c1 == 'B') {
1993 /* This is kanji introduction */
1998 /* could be some special code */
2005 } else if (broken_f&0x2) {
2006 /* accept any ESC-(-x as broken code ... */
2016 } else if (c1 == '(') {
2017 if ((c1 = (*i_getc)(f)) == EOF) {
2018 /* don't send bogus code
2020 (*oconv)(0, '('); */
2024 /* This is X0201 kana introduction */
2025 input_mode = X0201; shift_mode = X0201;
2027 } else if (c1 == 'B' || c1 == 'J' || c1 == 'H') {
2028 /* This is X0208 kanji introduction */
2029 input_mode = ASCII; shift_mode = FALSE;
2031 } else if (broken_f&0x2) {
2032 input_mode = ASCII; shift_mode = FALSE;
2037 /* maintain various input_mode here */
2041 } else if ( c1 == 'N' || c1 == 'n' ){
2043 c3 = (*i_getc)(f); /* skip SS2 */
2044 if ( (SPACE<=c3 && c3 < 0x60) || (0xa0<=c3 && c3 < 0xe0)){
2059 } else if ((c1 == NL || c1 == CR) && broken_f&4) {
2060 input_mode = ASCII; set_iconv(FALSE, 0);
2066 if (input_mode == X0208)
2067 (*oconv)(c2, c1); /* this is JIS, not SJIS/EUC case */
2068 else if (input_mode)
2069 (*oconv)(input_mode, c1); /* other special case */
2070 else if ((*iconv)(c2, c1, 0) < 0){ /* can be EUC/SJIS */
2071 int c0 = (*i_getc)(f);
2074 (*iconv)(c2, c1, c0);
2080 /* goto next_word */
2084 (*iconv)(EOF, 0, 0);
2097 /** it must NOT be in the kanji shifte sequence */
2098 /** it must NOT be written in JIS7 */
2099 /** and it must be after 2 byte 8bit code */
2106 while ((c1 = (*i_getc)(f)) != EOF) {
2112 if (push_hold_buf(c1) == EOF || estab_f){
2118 struct input_code *p = input_code_list;
2119 struct input_code *result = p;
2124 if (p->score < result->score){
2129 set_iconv(FALSE, result->iconv_func);
2134 ** 1) EOF is detected, or
2135 ** 2) Code is established, or
2136 ** 3) Buffer is FULL (but last word is pushed)
2138 ** in 1) and 3) cases, we continue to use
2139 ** Kanji codes by oconv and leave estab_f unchanged.
2144 while (wc < hold_count){
2145 c2 = hold_buf[wc++];
2147 #ifdef NUMCHAR_OPTION
2148 || (c2 & CLASS_MASK) == CLASS_UTF16
2153 }else if (iconv == s_iconv && 0xa1 <= c2 && c2 <= 0xdf){
2154 (*iconv)(X0201, c2, 0);
2157 if (wc < hold_count){
2158 c1 = hold_buf[wc++];
2167 if ((*iconv)(c2, c1, 0) < 0){
2169 if (wc < hold_count){
2170 c0 = hold_buf[wc++];
2179 (*iconv)(c2, c1, c0);
2192 if (hold_count >= HOLD_SIZE*2)
2194 hold_buf[hold_count++] = c2;
2195 return ((hold_count >= HOLD_SIZE*2) ? EOF : hold_count);
2198 int s2e_conv(c2, c1, p2, p1)
2202 #ifdef SHIFTJIS_CP932
2203 if (cp932_f && CP932_TABLE_BEGIN <= c2 && c2 <= CP932_TABLE_END){
2204 extern unsigned short shiftjis_cp932[3][189];
2205 c1 = shiftjis_cp932[c2 - CP932_TABLE_BEGIN][c1 - 0x40];
2206 if (c1 == 0) return 1;
2210 #endif /* SHIFTJIS_CP932 */
2211 c2 = c2 + c2 - ((c2 <= 0x9f) ? SJ0162 : SJ6394);
2213 c1 = c1 - ((c1 > DEL) ? SPACE : 0x1f);
2230 } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) {
2233 int ret = s2e_conv(c2, c1, &c2, &c1);
2234 if (ret) return ret;
2247 } else if (c2 == SSO){
2250 } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) {
2260 #ifdef UTF8_INPUT_ENABLE
2262 w2e_conv(c2, c1, c0, p2, p1)
2266 extern unsigned short * utf8_to_euc_2bytes[];
2267 extern unsigned short ** utf8_to_euc_3bytes[];
2270 if (0xc0 <= c2 && c2 <= 0xef) {
2271 unsigned short **pp;
2274 if (c0 == 0) return -1;
2275 pp = utf8_to_euc_3bytes[c2 - 0x80];
2276 ret = w_iconv_common(c1, c0, pp, sizeof_utf8_to_euc_C2, p2, p1);
2278 ret = w_iconv_common(c2, c1, utf8_to_euc_2bytes, sizeof_utf8_to_euc_2bytes, p2, p1);
2280 #ifdef NUMCHAR_OPTION
2283 if (p1) *p1 = CLASS_UTF16 | ww16_conv(c2, c1, c0);
2288 } else if (c2 == X0201) {
2301 int ret = w2e_conv(c2, c1, c0, &c2, &c1);
2309 w16w_conv(val, p2, p1, p0)
2317 }else if (val < 0x800){
2318 *p2 = 0xc0 | (val >> 6);
2319 *p1 = 0x80 | (val & 0x3f);
2322 *p2 = 0xe0 | (val >> 12);
2323 *p1 = 0x80 | ((val >> 6) & 0x3f);
2324 *p0 = 0x80 | (val & 0x3f);
2329 ww16_conv(c2, c1, c0)
2334 val = (c2 & 0x0f) << 12;
2335 val |= (c1 & 0x3f) << 6;
2337 }else if (c2 >= 0xc0){
2338 val = (c2 & 0x1f) << 6;
2339 val |= (c1 & 0x3f) << 6;
2347 w16e_conv(val, p2, p1)
2351 extern unsigned short * utf8_to_euc_2bytes[];
2352 extern unsigned short ** utf8_to_euc_3bytes[];
2354 unsigned short **pp;
2358 w16w_conv(val, &c2, &c1, &c0);
2361 pp = utf8_to_euc_3bytes[c2 - 0x80];
2362 psize = sizeof_utf8_to_euc_C2;
2363 ret = w_iconv_common(c1, c0, pp, psize, p2, p1);
2365 pp = utf8_to_euc_2bytes;
2366 psize = sizeof_utf8_to_euc_2bytes;
2367 ret = w_iconv_common(c2, c1, pp, psize, p2, p1);
2369 #ifdef NUMCHAR_OPTION
2372 *p1 = CLASS_UTF16 | val;
2381 w_iconv16(c2, c1, c0)
2386 if (c2==0376 && c1==0377){
2387 utf16_mode = UTF16LE_INPUT;
2389 } else if (c2==0377 && c1==0376){
2390 utf16_mode = UTF16BE_INPUT;
2393 if (c2 != EOF && utf16_mode == UTF16BE_INPUT) {
2395 tmp=c1; c1=c2; c2=tmp;
2397 if ((c2==0 && c1 < 0x80) || c2==EOF) {
2401 ret = w16e_conv(((c2<<8)&0xff00) + c1, &c2, &c1);
2402 if (ret) return ret;
2408 w_iconv_common(c1, c0, pp, psize, p2, p1)
2410 unsigned short **pp;
2418 if (pp == 0) return 1;
2421 if (c1 < 0 || psize <= c1) return 1;
2423 if (p == 0) return 1;
2426 if (c0 < 0 || sizeof_utf8_to_euc_E5B8 <= c0) return 1;
2428 if (val == 0) return 1;
2431 if (c2 == SO) c2 = X0201;
2440 #ifdef UTF8_OUTPUT_ENABLE
2445 extern unsigned short euc_to_utf8_1byte[];
2446 extern unsigned short * euc_to_utf8_2bytes[];
2447 extern unsigned short * euc_to_utf8_2bytes_ms[];
2451 p = euc_to_utf8_1byte;
2454 c2 = (c2&0x7f) - 0x21;
2455 if (0<=c2 && c2<sizeof_euc_to_utf8_2bytes)
2456 p = ms_ucs_map_f ? euc_to_utf8_2bytes_ms[c2] : euc_to_utf8_2bytes[c2];
2461 c1 = (c1 & 0x7f) - 0x21;
2462 if (0<=c1 && c1<sizeof_euc_to_utf8_1byte)
2473 #ifdef NUMCHAR_OPTION
2474 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2475 w16w_conv(c1, &c2, &c1, &c0);
2479 if (c0) (*o_putc)(c0);
2488 if (unicode_bom_f==2) {
2496 output_mode = ASCII;
2498 } else if (c2 == ISO8859_1) {
2499 output_mode = ISO8859_1;
2500 (*o_putc)(c1 | 0x080);
2503 w16w_conv((unsigned short)e2w_conv(c2, c1), &c2, &c1, &c0);
2507 if (c0) (*o_putc)(c0);
2522 if (unicode_bom_f==2) {
2524 (*o_putc)((unsigned char)'\377');
2528 (*o_putc)((unsigned char)'\377');
2533 if (c2 == ISO8859_1) {
2536 #ifdef NUMCHAR_OPTION
2537 } else if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16) {
2538 c2 = (c1 >> 8) & 0xff;
2542 unsigned short val = (unsigned short)e2w_conv(c2, c1);
2543 c2 = (val >> 8) & 0xff;
2562 #ifdef NUMCHAR_OPTION
2563 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2564 w16e_conv(c1, &c2, &c1);
2570 } else if (c2 == 0) {
2571 output_mode = ASCII;
2573 } else if (c2 == X0201) {
2574 output_mode = JAPANESE_EUC;
2575 (*o_putc)(SSO); (*o_putc)(c1|0x80);
2576 } else if (c2 == ISO8859_1) {
2577 output_mode = ISO8859_1;
2578 (*o_putc)(c1 | 0x080);
2580 if ((c1<0x21 || 0x7e<c1) ||
2581 (c2<0x21 || 0x7e<c2)) {
2582 set_iconv(FALSE, 0);
2583 return; /* too late to rescue this char */
2585 output_mode = JAPANESE_EUC;
2586 (*o_putc)(c2 | 0x080);
2587 (*o_putc)(c1 | 0x080);
2592 e2s_conv(c2, c1, p2, p1)
2593 int c2, c1, *p2, *p1;
2595 if (p2) *p2 = ((c2 - 1) >> 1) + ((c2 <= 0x5e) ? 0x71 : 0xb1);
2596 if (p1) *p1 = c1 + ((c2 & 1) ? ((c1 < 0x60) ? 0x1f : 0x20) : 0x7e);
2604 #ifdef NUMCHAR_OPTION
2605 if (c2 == 0 && (c1 & CLASS_MASK) == CLASS_UTF16){
2606 w16e_conv(c1, &c2, &c1);
2612 } else if (c2 == 0) {
2613 output_mode = ASCII;
2615 } else if (c2 == X0201) {
2616 output_mode = SHIFT_JIS;
2618 } else if (c2 == ISO8859_1) {
2619 output_mode = ISO8859_1;
2620 (*o_putc)(c1 | 0x080);
2622 if ((c1<0x20 || 0x7e<c1) ||
2623 (c2<0x20 || 0x7e<c2)) {
2624 set_iconv(FALSE, 0);
2625 return; /* too late to rescue this char */
2627 output_mode = SHIFT_JIS;
2628 e2s_conv(c2, c1, &c2, &c1);
2630 #ifdef SHIFTJIS_CP932
2632 && CP932INV_TABLE_BEGIN <= c2 && c2 <= CP932INV_TABLE_END){
2633 extern unsigned short cp932inv[2][189];
2634 int c = cp932inv[c2 - CP932INV_TABLE_BEGIN][c1 - 0x40];
2640 #endif /* SHIFTJIS_CP932 */
2643 if (prefix_table[(unsigned char)c1]){
2644 (*o_putc)(prefix_table[(unsigned char)c1]);
2655 #ifdef NUMCHAR_OPTION
2656 if ((c1 & CLASS_MASK) == CLASS_UTF16){
2657 w16e_conv(c1, &c2, &c1);
2661 if (output_mode !=ASCII && output_mode!=ISO8859_1) {
2664 (*o_putc)(ascii_intro);
2665 output_mode = ASCII;
2668 } else if (c2==X0201) {
2669 if (output_mode!=X0201) {
2670 output_mode = X0201;
2676 } else if (c2==ISO8859_1) {
2677 /* iso8859 introduction, or 8th bit on */
2678 /* Can we convert in 7bit form using ESC-'-'-A ?
2680 output_mode = ISO8859_1;
2682 } else if (c2 == 0) {
2683 if (output_mode !=ASCII && output_mode!=ISO8859_1) {
2686 (*o_putc)(ascii_intro);
2687 output_mode = ASCII;
2691 if (output_mode != X0208) {
2692 output_mode = X0208;
2695 (*o_putc)(kanji_intro);
2697 if (c1<0x20 || 0x7e<c1)
2699 if (c2<0x20 || 0x7e<c2)
2711 if (base64_count>50 && !mimeout_mode && c2==0 && c1==SPACE) {
2713 } else if (base64_count>66 && mimeout_mode) {
2714 (*o_base64conv)(EOF,0);
2716 (*o_putc)('\t'); base64_count += 7;
2718 (*o_base64conv)(c2,c1);
2722 static int broken_buf[3];
2723 static int broken_counter = 0;
2724 static int broken_last = 0;
2731 if (broken_counter>0) {
2732 return broken_buf[--broken_counter];
2735 if (c=='$' && broken_last != ESC
2736 && (input_mode==ASCII || input_mode==X0201)) {
2739 if (c1=='@'|| c1=='B') {
2740 broken_buf[0]=c1; broken_buf[1]=c;
2747 } else if (c=='(' && broken_last != ESC
2748 && (input_mode==X0208 || input_mode==X0201)) { /* ) */
2751 if (c1=='J'|| c1=='B') {
2752 broken_buf[0]=c1; broken_buf[1]=c;
2770 if (broken_counter<2)
2771 broken_buf[broken_counter++]=c;
2775 static int prev_cr = 0;
2783 if (! (c2==0&&c1==NL) ) {
2789 } else if (c1=='\r') {
2791 } else if (c1=='\n') {
2792 if (crmode_f==CRLF) {
2793 (*o_crconv)(0,'\r');
2794 } else if (crmode_f==CR) {
2795 (*o_crconv)(0,'\r');
2799 } else if (c1!='\032' || crmode_f!=NL){
2805 Return value of fold_conv()
2807 \n add newline and output char
2808 \r add newline and output nothing
2811 1 (or else) normal output
2813 fold state in prev (previous character)
2815 >0x80 Japanese (X0208/X0201)
2820 This fold algorthm does not preserve heading space in a line.
2821 This is the main difference from fmt.
2824 #define char_size(c2,c1) (c2?2:1)
2833 if (c1== '\r' && !fold_preserve_f) {
2834 fold_state=0; /* ignore cr */
2835 }else if (c1== '\n'&&f_prev=='\r' && fold_preserve_f) {
2837 fold_state=0; /* ignore cr */
2838 } else if (c1== BS) {
2839 if (f_line>0) f_line--;
2841 } else if (c2==EOF && f_line != 0) { /* close open last line */
2843 } else if ((c1=='\n' && !fold_preserve_f)
2844 || ((c1=='\r'||(c1=='\n'&&f_prev!='\r'))
2845 && fold_preserve_f)) {
2847 if (fold_preserve_f) {
2851 } else if ((f_prev == c1 && !fold_preserve_f)
2852 || (f_prev == '\n' && fold_preserve_f)
2853 ) { /* duplicate newline */
2856 fold_state = '\n'; /* output two newline */
2862 if (f_prev&0x80) { /* Japanese? */
2864 fold_state = 0; /* ignore given single newline */
2865 } else if (f_prev==' ') {
2869 if (++f_line<=fold_len)
2873 fold_state = '\r'; /* fold and output nothing */
2877 } else if (c1=='\f') {
2882 fold_state = '\n'; /* output newline and clear */
2883 } else if ( (c2==0 && c1==' ')||
2884 (c2==0 && c1=='\t')||
2885 (c2=='!'&& c1=='!')) {
2886 /* X0208 kankaku or ascii space */
2887 if (f_prev == ' ') {
2888 fold_state = 0; /* remove duplicate spaces */
2891 if (++f_line<=fold_len)
2892 fold_state = ' '; /* output ASCII space only */
2894 f_prev = ' '; f_line = 0;
2895 fold_state = '\r'; /* fold and output nothing */
2899 prev0 = f_prev; /* we still need this one... , but almost done */
2901 if (c2 || c2==X0201)
2902 f_prev |= 0x80; /* this is Japanese */
2903 f_line += char_size(c2,c1);
2904 if (f_line<=fold_len) { /* normal case */
2907 if (f_line>=fold_len+fold_margin) { /* too many kinsou suspension */
2908 f_line = char_size(c2,c1);
2909 fold_state = '\n'; /* We can't wait, do fold now */
2910 } else if (c2==X0201) {
2911 /* simple kinsoku rules return 1 means no folding */
2912 if (c1==(0xde&0x7f)) fold_state = 1; /*
\e$B!+
\e(B*/
2913 else if (c1==(0xdf&0x7f)) fold_state = 1; /*
\e$B!,
\e(B*/
2914 else if (c1==(0xa4&0x7f)) fold_state = 1; /*
\e$B!#
\e(B*/
2915 else if (c1==(0xa3&0x7f)) fold_state = 1; /*
\e$B!$
\e(B*/
2916 else if (c1==(0xa1&0x7f)) fold_state = 1; /*
\e$B!W
\e(B*/
2917 else if (c1==(0xb0&0x7f)) fold_state = 1; /* - */
2918 else if (SPACE<=c1 && c1<=(0xdf&0x7f)) { /* X0201 */
2920 fold_state = '\n';/* add one new f_line before this character */
2923 fold_state = '\n';/* add one new f_line before this character */
2926 /* kinsoku point in ASCII */
2927 if ( c1==')'|| /* { [ ( */
2938 /* just after special */
2939 } else if (!is_alnum(prev0)) {
2940 f_line = char_size(c2,c1);
2942 } else if ((prev0==' ') || /* ignored new f_line */
2943 (prev0=='\n')|| /* ignored new f_line */
2944 (prev0&0x80)) { /* X0208 - ASCII */
2945 f_line = char_size(c2,c1);
2946 fold_state = '\n';/* add one new f_line before this character */
2948 fold_state = 1; /* default no fold in ASCII */
2952 if (c1=='"') fold_state = 1; /*
\e$B!"
\e(B */
2953 else if (c1=='#') fold_state = 1; /*
\e$B!#
\e(B */
2954 else if (c1=='W') fold_state = 1; /*
\e$B!W
\e(B */
2955 else if (c1=='K') fold_state = 1; /*
\e$B!K
\e(B */
2956 else if (c1=='$') fold_state = 1; /*
\e$B!$
\e(B */
2957 else if (c1=='%') fold_state = 1; /*
\e$B!%
\e(B */
2958 else if (c1=='\'') fold_state = 1; /*
\e$B!\
\e(B */
2959 else if (c1=='(') fold_state = 1; /*
\e$B!(
\e(B */
2960 else if (c1==')') fold_state = 1; /*
\e$B!)
\e(B */
2961 else if (c1=='*') fold_state = 1; /*
\e$B!*
\e(B */
2962 else if (c1=='+') fold_state = 1; /*
\e$B!+
\e(B */
2963 else if (c1==',') fold_state = 1; /*
\e$B!,
\e(B */
2964 /* default no fold in kinsoku */
2967 f_line = char_size(c2,c1);
2968 /* add one new f_line before this character */
2971 f_line = char_size(c2,c1);
2973 /* add one new f_line before this character */
2978 /* terminator process */
2979 switch(fold_state) {
2998 int z_prev2=0,z_prev1=0;
3005 /* if (c2) c1 &= 0x7f; assertion */
3007 if (x0201_f && z_prev2==X0201) { /* X0201 */
3008 if (c1==(0xde&0x7f)) { /*
\e$BByE@
\e(B */
3010 (*o_zconv)(dv[(z_prev1-SPACE)*2],dv[(z_prev1-SPACE)*2+1]);
3012 } else if (c1==(0xdf&0x7f)&&ev[(z_prev1-SPACE)*2]) { /*
\e$BH>ByE@
\e(B */
3014 (*o_zconv)(ev[(z_prev1-SPACE)*2],ev[(z_prev1-SPACE)*2+1]);
3018 (*o_zconv)(cv[(z_prev1-SPACE)*2],cv[(z_prev1-SPACE)*2+1]);
3027 if (x0201_f && c2==X0201) {
3028 if (dv[(c1-SPACE)*2]||ev[(c1-SPACE)*2]) {
3029 /* wait for
\e$BByE@
\e(B or
\e$BH>ByE@
\e(B */
3030 z_prev1 = c1; z_prev2 = c2;
3033 (*o_zconv)(cv[(c1-SPACE)*2],cv[(c1-SPACE)*2+1]);
3038 /* JISX0208 Alphabet */
3039 if (alpha_f && c2 == 0x23 ) {
3041 } else if (alpha_f && c2 == 0x21 ) {
3042 /* JISX0208 Kigou */
3047 } else if (alpha_f&0x4) {
3052 } else if (0x20<c1 && c1<0x7f && fv[c1-0x20]) {
3058 case '>': entity = ">"; break;
3059 case '<': entity = "<"; break;
3060 case '\"': entity = """; break;
3061 case '&': entity = "&"; break;
3064 while (*entity) (*o_zconv)(0, *entity++);
3074 #define rot13(c) ( \
3076 (c <= 'M') ? (c + 13): \
3077 (c <= 'Z') ? (c - 13): \
3079 (c <= 'm') ? (c + 13): \
3080 (c <= 'z') ? (c - 13): \
3084 #define rot47(c) ( \
3086 ( c <= 'O' ) ? (c + 47) : \
3087 ( c <= '~' ) ? (c - 47) : \
3095 if (c2==0 || c2==X0201 || c2==ISO8859_1) {
3101 (*o_rot_conv)(c2,c1);
3108 if ((hira_f & 1) && c2==0x25 && 0x20<c1 && c1<0x74) {
3110 } else if ((hira_f & 2) && c2==0x24 && 0x20<c1 && c1<0x74) {
3113 (*o_hira_conv)(c2,c1);
3118 iso2022jp_check_conv(c2,c1)
3121 static int range[RANGE_NUM_MAX][2] = {
3144 if(c2 >= 0x00 && c2 <= 0x20 && c1 >= 0x7f && c1 <= 0xff) {
3148 if((c2 >= 0x29 && c2 <= 0x2f) || (c2 >= 0x75 && c2 <= 0x7e)) {
3153 for (i = 0; i < RANGE_NUM_MAX; i++) {
3154 start = range[i][0];
3157 if (c >= start && c <= end) {
3162 (*o_iso2022jp_check_conv)(c2,c1);
3166 /* This converts =?ISO-2022-JP?B?HOGE HOGE?= */
3168 unsigned char *mime_pattern[] = {
3169 (unsigned char *)"\075?EUC-JP?B?",
3170 (unsigned char *)"\075?SHIFT_JIS?B?",
3171 (unsigned char *)"\075?ISO-8859-1?Q?",
3172 (unsigned char *)"\075?ISO-8859-1?B?",
3173 (unsigned char *)"\075?ISO-2022-JP?B?",
3174 (unsigned char *)"\075?ISO-2022-JP?Q?",
3175 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3176 (unsigned char *)"\075?UTF-8?B?",
3177 (unsigned char *)"\075?UTF-8?Q?",
3179 (unsigned char *)"\075?US-ASCII?Q?",
3184 /*
\e$B3:Ev$9$k%3!<%I$NM%@hEY$r>e$2$k$?$a$NL\0u
\e(B */
3185 int (*mime_priority_func[])PROTO((int c2, int c1, int c0)) = {
3186 e_iconv, s_iconv, 0, 0, 0, 0,
3187 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3193 int mime_encode[] = {
3194 JAPANESE_EUC, SHIFT_JIS,ISO8859_1, ISO8859_1, X0208, X0201,
3195 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3202 int mime_encode_method[] = {
3203 'B', 'B','Q', 'B', 'B', 'Q',
3204 #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE)
3212 #define MAXRECOVER 20
3214 /* I don't trust portablity of toupper */
3215 #define nkf_toupper(c) (('a'<=c && c<='z')?(c-('a'-'A')):c)
3216 #define nkf_isdigit(c) ('0'<=c && c<='9')
3217 #define nkf_isxdigit(c) (nkf_isdigit(c) || ('a'<=c && c<='f') || ('A'<=c && c <= 'F'))
3222 if (i_getc!=mime_getc) {
3223 i_mgetc = i_getc; i_getc = mime_getc;
3224 i_mungetc = i_ungetc; i_ungetc = mime_ungetc;
3225 if(mime_f==STRICT_MIME) {
3226 i_mgetc_buf = i_mgetc; i_mgetc = mime_getc_buf;
3227 i_mungetc_buf = i_mungetc; i_mungetc = mime_ungetc_buf;
3233 unswitch_mime_getc()
3235 if(mime_f==STRICT_MIME) {
3236 i_mgetc = i_mgetc_buf;
3237 i_mungetc = i_mungetc_buf;
3240 i_ungetc = i_mungetc;
3244 mime_begin_strict(f)
3249 unsigned char *p,*q;
3250 int r[MAXRECOVER]; /* recovery buffer, max mime pattern lenght */
3252 mime_decode_mode = FALSE;
3253 /* =? has been checked */
3255 p = mime_pattern[j];
3258 for(i=2;p[i]>' ';i++) { /* start at =? */
3259 if ( ((r[i] = c1 = (*i_getc)(f))==EOF) || nkf_toupper(c1) != p[i] ) {
3260 /* pattern fails, try next one */
3262 while ((p = mime_pattern[++j])) {
3263 for(k=2;k<i;k++) /* assume length(p) > i */
3264 if (p[k]!=q[k]) break;
3265 if (k==i && nkf_toupper(c1)==p[k]) break;
3267 if (p) continue; /* found next one, continue */
3268 /* all fails, output from recovery buffer */
3276 mime_decode_mode = p[i-2];
3278 clr_code_score(find_inputcode_byfunc(mime_priority_func[j]), SCORE_iMIME);
3280 if (mime_decode_mode=='B') {
3281 mimebuf_f = unbuf_f;
3283 /* do MIME integrity check */
3284 return mime_integrity(f,mime_pattern[j]);
3296 /* we don't keep eof of Fifo, becase it contains ?= as
3297 a terminator. It was checked in mime_integrity. */
3298 return ((mimebuf_f)?
3299 (*i_mgetc_buf)(f):Fifo(mime_input++));
3303 mime_ungetc_buf(c,f)
3308 (*i_mungetc_buf)(c,f);
3310 Fifo(--mime_input)=c;
3321 /* In NONSTRICT mode, only =? is checked. In case of failure, we */
3322 /* re-read and convert again from mime_buffer. */
3324 /* =? has been checked */
3326 Fifo(mime_last++)='='; Fifo(mime_last++)='?';
3327 for(i=2;i<MAXRECOVER;i++) { /* start at =? */
3328 /* We accept any character type even if it is breaked by new lines */
3329 c1 = (*i_getc)(f); Fifo(mime_last++)= c1 ;
3330 if (c1=='\n'||c1==' '||c1=='\r'||
3331 c1=='-'||c1=='_'||is_alnum(c1) ) continue;
3333 /* Failed. But this could be another MIME preemble */
3341 c1 = (*i_getc)(f); Fifo(mime_last++) = c1;
3342 if (!(++i<MAXRECOVER) || c1==EOF) break;
3343 if (c1=='b'||c1=='B') {
3344 mime_decode_mode = 'B';
3345 } else if (c1=='q'||c1=='Q') {
3346 mime_decode_mode = 'Q';
3350 c1 = (*i_getc)(f); Fifo(mime_last++) = c1;
3351 if (!(++i<MAXRECOVER) || c1==EOF) break;
3353 mime_decode_mode = FALSE;
3359 if (!mime_decode_mode) {
3360 /* false MIME premble, restart from mime_buffer */
3361 mime_decode_mode = 1; /* no decode, but read from the mime_buffer */
3362 /* Since we are in MIME mode until buffer becomes empty, */
3363 /* we never go into mime_begin again for a while. */
3366 /* discard mime preemble, and goto MIME mode */
3368 /* do no MIME integrity check */
3369 return c1; /* used only for checking EOF */
3384 fprintf(stderr, "%s\n", str);
3390 set_input_codename (codename)
3395 strcmp(codename, "") != 0 &&
3396 strcmp(codename, input_codename) != 0)
3398 is_inputcode_mixed = TRUE;
3400 input_codename = codename;
3401 is_inputcode_set = TRUE;
3405 print_guessed_code (filename)
3408 char *codename = "BINARY";
3409 if (!is_inputcode_mixed) {
3410 if (strcmp(input_codename, "") == 0) {
3413 codename = input_codename;
3416 if (filename != NULL) printf("%s:", filename);
3417 printf("%s\n", codename);
3424 if (nkf_isdigit(x)) return x - '0';
3425 return nkf_toupper(x) - 'A' + 10;
3430 #ifdef ANSI_C_PROTOTYPE
3431 int hex_getc(int ch, FILE *f, int (*g)(FILE *f), int (*u)(int c, FILE *f))
3434 hex_getc(ch, f, g, u)
3447 if (!nkf_isxdigit(c2)){
3452 if (!nkf_isxdigit(c3)){
3457 return (hex2bin(c2) << 4) | hex2bin(c3);
3464 return hex_getc(':', f, i_cgetc, i_cungetc);
3472 return (*i_cungetc)(c, f);
3479 return hex_getc('%', f, i_ugetc, i_uungetc);
3487 return (*i_uungetc)(c, f);
3491 #ifdef NUMCHAR_OPTION
3496 int (*g)() = i_ngetc;
3497 int (*u)() = i_nungetc;
3508 if (buf[i] == 'x' || buf[i] == 'X'){
3509 for (j = 0; j < 5; j++){
3511 if (!nkf_isxdigit(buf[i])){
3518 c |= hex2bin(buf[i]);
3521 for (j = 0; j < 6; j++){
3525 if (!nkf_isdigit(buf[i])){
3532 c += hex2bin(buf[i]);
3538 return CLASS_UTF16 | c;
3548 numchar_ungetc(c, f)
3552 return (*i_nungetc)(c, f);
3561 int c1, c2, c3, c4, cc;
3562 int t1, t2, t3, t4, mode, exit_mode;
3564 if (mime_top != mime_last) { /* Something is in FIFO */
3565 return Fifo(mime_top++);
3567 if (mime_decode_mode==1 ||mime_decode_mode==FALSE) {
3568 mime_decode_mode=FALSE;
3569 unswitch_mime_getc();
3570 return (*i_getc)(f);
3573 if (mimebuf_f == FIXED_MIME)
3574 exit_mode = mime_decode_mode;
3577 if (mime_decode_mode == 'Q') {
3578 if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF);
3580 if (c1=='_') return ' ';
3581 if (c1!='=' && c1!='?') {
3585 mime_decode_mode = exit_mode; /* prepare for quit */
3586 if (c1<=' ') return c1;
3587 if ((c2 = (*i_mgetc)(f)) == EOF) return (EOF);
3588 if (c1=='?'&&c2=='=' && mimebuf_f != FIXED_MIME) {
3589 /* end Q encoding */
3590 input_mode = exit_mode;
3591 while((c1=(*i_getc)(f))!=EOF && c1==SPACE
3592 /* && (c1==NL||c1==TAB||c1=='\r') */ ) ;
3595 if (c1=='='&&c2<' ') { /* this is soft wrap */
3596 while((c1 = (*i_mgetc)(f)) <=' ') {
3597 if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF);
3599 mime_decode_mode = 'Q'; /* still in MIME */
3600 goto restart_mime_q;
3603 mime_decode_mode = 'Q'; /* still in MIME */
3607 if ((c3 = (*i_mgetc)(f)) == EOF) return (EOF);
3608 if (c2<=' ') return c2;
3609 mime_decode_mode = 'Q'; /* still in MIME */
3610 #define hex(c) (('0'<=c&&c<='9')?(c-'0'):\
3611 ('A'<=c&&c<='F')?(c-'A'+10):('a'<=c&&c<='f')?(c-'a'+10):0)
3612 return ((hex(c2)<<4) + hex(c3));
3615 if (mime_decode_mode != 'B') {
3616 mime_decode_mode = FALSE;
3617 return (*i_mgetc)(f);
3621 /* Base64 encoding */
3623 MIME allows line break in the middle of
3624 Base64, but we are very pessimistic in decoding
3625 in unbuf mode because MIME encoded code may broken by
3626 less or editor's control sequence (such as ESC-[-K in unbuffered
3627 mode. ignore incomplete MIME.
3629 mode = mime_decode_mode;
3630 mime_decode_mode = exit_mode; /* prepare for quit */
3632 while ((c1 = (*i_mgetc)(f))<=' ') {
3637 if ((c2 = (*i_mgetc)(f))<=' ') {
3640 if (mime_f != STRICT_MIME) goto mime_c2_retry;
3641 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
3644 if ((c1 == '?') && (c2 == '=')) {
3646 while((c1=(*i_getc)(f))!=EOF && c1==SPACE
3647 /* && (c1==NL||c1==TAB||c1=='\r') */ ) ;
3651 if ((c3 = (*i_mgetc)(f))<=' ') {
3654 if (mime_f != STRICT_MIME) goto mime_c3_retry;
3655 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
3659 if ((c4 = (*i_mgetc)(f))<=' ') {
3662 if (mime_f != STRICT_MIME) goto mime_c4_retry;
3663 if (mimebuf_f!=FIXED_MIME) input_mode = ASCII;
3667 mime_decode_mode = mode; /* still in MIME sigh... */
3669 /* BASE 64 decoding */
3671 t1 = 0x3f & base64decode(c1);
3672 t2 = 0x3f & base64decode(c2);
3673 t3 = 0x3f & base64decode(c3);
3674 t4 = 0x3f & base64decode(c4);
3675 cc = ((t1 << 2) & 0x0fc) | ((t2 >> 4) & 0x03);
3677 Fifo(mime_last++) = cc;
3678 cc = ((t2 << 4) & 0x0f0) | ((t3 >> 2) & 0x0f);
3680 Fifo(mime_last++) = cc;
3681 cc = ((t3 << 6) & 0x0c0) | (t4 & 0x3f);
3683 Fifo(mime_last++) = cc;
3688 return Fifo(mime_top++);
3696 Fifo(--mime_top) = c;
3707 /* In buffered mode, read until =? or NL or buffer full
3709 mime_input = mime_top;
3710 mime_last = mime_top;
3711 while(*p) Fifo(mime_input++) = *p++;
3714 while((c=(*i_getc)(f))!=EOF) {
3715 if (((mime_input-mime_top)&MIME_BUF_MASK)==0) {
3716 break; /* buffer full */
3718 if (c=='=' && d=='?') {
3719 /* checked. skip header, start decode */
3720 Fifo(mime_input++) = c;
3721 /* mime_last_input = mime_input; */
3726 if (!( (c=='+'||c=='/'|| c=='=' || c=='?' || is_alnum(c))))
3728 /* Should we check length mod 4? */
3729 Fifo(mime_input++) = c;
3732 /* In case of Incomplete MIME, no MIME decode */
3733 Fifo(mime_input++) = c;
3734 mime_last = mime_input; /* point undecoded buffer */
3735 mime_decode_mode = 1; /* no decode on Fifo last in mime_getc */
3736 switch_mime_getc(); /* anyway we need buffered getc */
3747 i = c - 'A'; /* A..Z 0-25 */
3749 i = c - 'G' /* - 'a' + 26 */ ; /* a..z 26-51 */
3751 } else if (c > '/') {
3752 i = c - '0' + '4' /* - '0' + 52 */ ; /* 0..9 52-61 */
3753 } else if (c == '+') {
3754 i = '>' /* 62 */ ; /* + 62 */
3756 i = '?' /* 63 */ ; /* / 63 */
3761 static char basis_64[] =
3762 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
3772 p = mime_pattern[0];
3773 for(i=0;mime_encode[i];i++) {
3774 if (mode == mime_encode[i]) {
3775 p = mime_pattern[i];
3779 mimeout_mode = mime_encode_method[i];
3781 /* (*o_mputc)(' '); */