/* * @(#)$Header: /mm2/home/cvs/bc-src/tgif/nkf.c,v 1.1 2004/06/18 23:19:09 william Exp $ */ /** Network Kanji Filter. (PDS Version) ************************************************************************ ** Copyright (C) 1987, Fujitsu LTD. (Itaru ICHIKAWA) ** 連絡先： （株）富士通研究所　ソフト３研　市川　至 ** （E-Mail Address: ichikawa@flab.fujitsu.co.jp） ** 営利を目的としない限り、このソースのいかなる ** 複写，改変，修正も許諾します。その際には、この部分を残すこと。 ** このプログラムについては特に何の保証もしない、悪しからず。 ** Everyone is permitted to do anything on this program ** including copying, modifying, improving ** as long as you don't try to make money off it, ** or pretend that you wrote it. ** i.e., the above copyright notice has to appear in all copies. ** THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE. ***********************************************************************/ /*********************************************************************** ** 1996/03/10 modified for Kconv - by Ikuo Nakagawa ***********************************************************************/ #ifndef _NO_NKF #define _INCLUDE_FROM_NKF_C_ #include "tgifdefs.h" /* for Kconv: _AUTO, _EUC, _SJIS, _JIS */ #include "nkf.e" #include "setup.e" #include "util.e" int useNKF=FALSE; #ifdef notdef static char *CopyRight = "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa)"; static char *Version = "1.5"; static char *Patchlevel = "1/9603/Ikuo Nakagawa"; #endif /* defined(notdef) */ /* ** ** ** ** USAGE: nkf [flags] [file] ** ** Flags: ** b Output is bufferred (DEFAULT) ** u Output is unbufferred ** ** t no operation ** ** j Outout code is JIS 7 bit (DEFAULT) ** s Output code is MS Kanji ** e Output code is AT&T JIS ** l Output code is JIS 7bit and ISO8859-1 Latin-1 ** ** m MIME conversion for ISO-2022-JP ** i_ Output sequence to designate JIS-kanji (DEFAULT_J) ** o_ Output sequence to designate single-byte roman characters (DEFAULT_R) ** ** r {de/en}crypt ROT13/47 ** ** v display Version ** ** T Text mode output (for MS-DOS) ** ** x Do not convert X0201 kana into X0208 ** Z Convert X0208 alphabet to ASCII ** ** f60 fold option ** ** m MIME decode ** B try to fix broken JIS, missing Escape ** **/ #if (defined(__TURBOC__) || defined(LSI_C)) && !defined(MSDOS) #define MSDOS #endif #include #if 0 #include #endif /* notdef */ #ifdef MSDOS #include #include #include #endif #ifdef MSDOS #ifdef LSI_C #define setbinmode(fp) fsetbin(fp) #else /* Microsoft C, Turbo C */ #define setbinmode(fp) setmode(fileno(fp), O_BINARY) #endif #else /* UNIX */ #define setbinmode(fp) #endif #ifdef _IOFBF /* SysV and MSDOS */ #define setvbuffer(fp, buf, size) setvbuf(fp, buf, _IOFBF, size) #else /* BSD */ #define setvbuffer(fp, buf, size) setbuffer(fp, buf, size) #endif #define FALSE 0 #define TRUE 1 /* state of output_mode and input_mode */ #define ASCII 0 #define X0208 1 #define X0201 2 #define NO_X0201 3 #define JIS_INPUT 4 #define SJIS_INPUT 5 #define LATIN1_INPUT 6 #define ESC 0x1b #define SP 0x20 #define AT 0x40 #define SSP 0xa0 #define DEL 0x7f #define SI 0x0f #define SO 0x0e #define SSO 0x8e #define HOLD_SIZE 32 #define IOBUF_SIZE 16384 #define DEFAULT_J 'B' #define DEFAULT_R 'B' #define SJ0162 0x00e1 /* 01 - 62 ku offset */ #define SJ6394 0x0161 /* 63 - 94 ku offset */ /* MIME preprocessor */ #define _GETC() (*inptr ? (int)(*inptr++) : EOF) #define _UNGETC(c) (*--inptr = (c)) #define PUTCHAR(c) (outlen + 1 < outsiz ? \ ((outptr[outlen++] = (c)), (outptr[outlen] = '\0')) : EOF) #define GETC() ((!mime_mode)?_GETC():mime_getc()) #define UNGETC(c) ((!mime_mode)?_UNGETC(c):mime_ungetc(c)) /* buffers */ #ifdef notdef static char stdibuf[IOBUF_SIZE]; static char stdobuf[IOBUF_SIZE]; #endif /* notdef */ static unsigned char hold_buf[HOLD_SIZE*2]; static int hold_count; static unsigned char *inptr; static char *outptr; static int outsiz; static int outlen; /* MIME preprocessor fifo */ #define MIME_BUF_SIZE 4 /* 2^n ring buffer */ #define MIME_BUF_MASK (MIME_BUF_SIZE-1) #define Fifo(n) mime_buf[(n)&MIME_BUF_MASK] static unsigned char mime_buf[MIME_BUF_SIZE]; static unsigned int mime_top = 0; static unsigned int mime_last = 0; /* flags */ static int unbuf_f = FALSE; static int estab_f = FALSE; #ifndef KCONV #ifdef notdef static int nop_f = FALSE; static int binmode_f = TRUE; /* binary mode */ #endif /* notdef */ #endif static int rot_f = FALSE; /* rot14/43 mode */ static int input_f = FALSE; /* non fixed input code */ static int alpha_f = FALSE; /* convert JIx0208 alphbet to ASCII */ static int mime_f = FALSE; /* convert MIME base64 */ static int broken_f = FALSE; /* convert ESC-less broken JIS */ static int iso8859_f = FALSE; /* ISO8859 through */ #ifdef MSDOS static int x0201_f = TRUE; /* Assume JISX0201 kana */ #else static int x0201_f = NO_X0201; /* Assume NO JISX0201 */ #endif /* X0208 -> ASCII converter */ static int c1_return; static int pre_convert(); /* rot13 or JISX0201 -> JISX0208 */ /* fold parameter */ static int line = 0; /* chars in line */ static int prev = 0; static int fold_f = FALSE; static int fold_len = 0; /* options */ static char kanji_intro = DEFAULT_J, ascii_intro = DEFAULT_R; /* Folding */ static int fold(); #define FOLD_MARGIN 10 #define DEFAULT_FOLD 60 /* converters */ static int s_iconv (); static int e_oconv (); static int j_oconv (); static int s_oconv (); static int h_conv (); static int push_hold_buf (); static int (*iconv) (); /* s_iconv or oconv */ static int (*oconv) (); /* [ejs]_oconv */ /* Global states */ static int output_mode = ASCII, /* output kanji mode */ input_mode = ASCII, /* input kanji mode */ shift_mode = FALSE; /* TRUE shift out, or X0201 */ static int mime_mode = FALSE; /* MIME mode B base64, Q hex */ /* X0201 / X0208 conversion tables - we defined here */ /* 90-9F A0-DF */ static unsigned char cv[]= { 0x21,0x21,0x21,0x23,0x21,0x56,0x21,0x57, 0x21,0x22,0x21,0x26,0x25,0x72,0x25,0x21, 0x25,0x23,0x25,0x25,0x25,0x27,0x25,0x29, 0x25,0x63,0x25,0x65,0x25,0x67,0x25,0x43, 0x21,0x3c,0x25,0x22,0x25,0x24,0x25,0x26, 0x25,0x28,0x25,0x2a,0x25,0x2b,0x25,0x2d, 0x25,0x2f,0x25,0x31,0x25,0x33,0x25,0x35, 0x25,0x37,0x25,0x39,0x25,0x3b,0x25,0x3d, 0x25,0x3f,0x25,0x41,0x25,0x44,0x25,0x46, 0x25,0x48,0x25,0x4a,0x25,0x4b,0x25,0x4c, 0x25,0x4d,0x25,0x4e,0x25,0x4f,0x25,0x52, 0x25,0x55,0x25,0x58,0x25,0x5b,0x25,0x5e, 0x25,0x5f,0x25,0x60,0x25,0x61,0x25,0x62, 0x25,0x64,0x25,0x66,0x25,0x68,0x25,0x69, 0x25,0x6a,0x25,0x6b,0x25,0x6c,0x25,0x6d, 0x25,0x6f,0x25,0x73,0x21,0x2b,0x21,0x2c, 0x00,0x00}; /* X0201 kana conversion table for daguten */ /* 90-9F A0-DF */ static unsigned char dv[]= { 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x25,0x2c,0x25,0x2e, 0x25,0x30,0x25,0x32,0x25,0x34,0x25,0x36, 0x25,0x38,0x25,0x3a,0x25,0x3c,0x25,0x3e, 0x25,0x40,0x25,0x42,0x25,0x45,0x25,0x47, 0x25,0x49,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x25,0x50,0x25,0x53, 0x25,0x56,0x25,0x59,0x25,0x5c,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00}; /* X0201 kana conversion table for han-daguten */ /* 90-9F A0-DF */ static unsigned char ev[]= { 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x25,0x51,0x25,0x54, 0x25,0x57,0x25,0x5a,0x25,0x5d,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x00,0x00}; /* X0208 kigou conversion table */ /* 0x8140 - 0x819e */ static unsigned char fv[] = { 0x00,0x00,0x00,0x00,0x2c,0x2e,0x00,0x3a, 0x3b,0x3f,0x21,0x00,0x00,0x27,0x60,0x00, 0x5e,0x00,0x5f,0x00,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x2d,0x00,0x2f, 0x5c,0x7e,0x00,0x7c,0x00,0x00,0x60,0x27, 0x22,0x22,0x28,0x29,0x00,0x00,0x5b,0x5d, 0x7b,0x7d,0x3c,0x3e,0x00,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x2b,0x2d,0x00,0x00, 0x00,0x3d,0x00,0x3c,0x3e,0x00,0x00,0x00, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0x24,0x00,0x00,0x25,0x23,0x26,0x2a,0x40, 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00 } ; /* mime related */ static int base64decode(); static int mime_getc(); static int mime_ungetc(); static int mime_begin(); #ifdef notdef main (argc, argv) int argc; char **argv; { register FILE *fin; register int c; register char *cp; oconv = j_oconv; /* DEFAULT OUTPUT */ for (argc--,argv++; (argc > 0) && **argv == '-'; argc--, argv++) { cp = *argv; while (*cp) { switch (*cp++) { case 'b': /* buffered mode */ unbuf_f = FALSE; continue; case 'u': /* non bufferd mode */ unbuf_f = TRUE; continue; case 't': /* transparent mode */ nop_f = TRUE; continue; case 'j': /* JIS output */ case 'n': oconv = j_oconv; continue; case 'e': /* AT&T EUC output */ oconv = e_oconv; continue; case 's': /* SJIS output */ oconv = s_oconv; continue; case 'l': /* ISO8859 Latin-1 support, no conversion */ iso8859_f = TRUE; /* Only compatible with ISO-2022-JP */ input_f = LATIN1_INPUT; continue; case 'i': /* Kanji IN ESC-$-@/B */ if(*cp) kanji_intro = *cp++; continue; case 'o': /* ASCII IN ESC-(-J/B */ if(*cp) ascii_intro = *cp++; continue; case 'r': rot_f = TRUE; continue; #ifdef MSDOS case 'T': binmode_f = FALSE; continue; #endif case 'v': usage(); exit(1); break; /* Input code assumption */ case 'J': /* JIS input */ case 'E': /* AT&T EUC input */ input_f = JIS_INPUT; continue; case 'S': /* MS Kanji input */ input_f = SJIS_INPUT; if(x0201_f==NO_X0201) x0201_f=TRUE; continue; case 'Z': /* Convert X0208 alphabet to asii */ alpha_f = TRUE; continue; case 'x': /* Convert X0201 kana to X0208 or X0201 Conversion */ x0201_f = FALSE; /* No X0201->X0208 conversion */ /* accept X0201 ESC-(-I in JIS, EUC, MS Kanji SI/SO in JIS, EUC, MS Kanji SSO in EUC, JIS, not in MS Kanji MS Kanji (0xa0-0xdf) output X0201 ESC-(-I in JIS (0x20-0x5f) SSO in EUC (0xa0-0xdf) 0xa0-0xd in MS Kanji (0xa0-0xdf) */ continue; case 'X': /* Assume X0201 kana */ /* Default value is NO_X0201 for EUC/MS-Kanji mix */ x0201_f = TRUE; continue; case 'f': /* folding -f60 or -f */ fold_f = TRUE; fold_len = atoi(cp); if(!(0 DEL) { /* in case of 8th bit is on */ if (!estab_f) { /* in case of not established yet */ if (c1 > SSP) { /* It is still ambiguious */ h_conv (c2, c1); c2 = 0; NEXT; } else if (c1 < AT) { /* ignore bogus code */ c2 = 0; NEXT; } else { /* established */ /* it seems to be MS Kanji */ estab_f = TRUE; iconv = s_iconv; SEND; } } else /* in case of already established */ if (c1 < AT) { /* ignore bogus code */ c2 = 0; NEXT; } else SEND; } else /* 7 bit code */ /* it might be kanji shitfted */ if ((c1 == DEL) || (c1 <= SP)) { /* ignore bogus first code */ c2 = 0; NEXT; } else SEND; } else { /* first byte */ if (c1 > DEL) { /* 8 bit code */ if (!estab_f && !iso8859_f) { /* not established yet */ if (c1 < SSP) { /* it seems to be MS Kanji */ estab_f = TRUE; iconv = s_iconv; } else if (c1 < 0xe0) { /* it seems to be EUC */ estab_f = TRUE; iconv = oconv; } else { /* still ambiguious */ } c2 = c1; NEXT; } else { /* estab_f==TRUE */ if(iso8859_f) { SEND; } else if(SSP<=c1 && c1<0xe0 && iconv == s_iconv) { /* SJIS X0201 Case... */ /* This is too arrogant, but ... */ if(x0201_f==NO_X0201) { iconv = oconv; c2 = c1; NEXT; } else if(x0201_f) { if(dv[(c1-SSP)*2]||ev[(c1-SSP)*2]) { /* look ahead for X0201/X0208conversion */ if ((c2 = GETC()) == EOF) { (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); LAST; } else if (c2==(0xde)) { /* 濁点 */ (*oconv)(dv[(c1-SSP)*2],dv[(c1-SSP)*2+1]); c2=0; NEXT; } else if (c2==(0xdf)&&ev[(c1-SSP)*2]) { /* 半濁点 */ (*oconv)(ev[(c1-SSP)*2],ev[(c1-SSP)*2+1]); c2=0; NEXT; } UNGETC(c2); c2 = 0; } (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); NEXT; } else SEND; } else if(c1==SSO && iconv != s_iconv) { /* EUC X0201 Case */ /* This is too arrogant if(x0201_f == NO_X0201) { estab_f = FALSE; c2 = 0; NEXT; } */ c1 = GETC(); /* skip SSO */ euc_1byte_check: if(x0201_f && SSP<=c1 && c1<0xe0) { if(dv[(c1-SSP)*2]||ev[(c1-SSP)*2]) { if ((c2 = GETC()) == EOF) { (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); LAST; } /* forward lookup 濁点/半濁点 */ if (c2 != SSO) { UNGETC(c2); c2 = 0; (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); NEXT; } else if ((c2 = GETC()) == EOF) { (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); (*oconv)(0,SSO); LAST; } else if (c2==(0xde)) { /* 濁点 */ (*oconv)(dv[(c1-SSP)*2],dv[(c1-SSP)*2+1]); c2=0; NEXT; } else if (c2==(0xdf)&&ev[(c1-SSP)*2]) { /* 半濁点 */ (*oconv)(ev[(c1-SSP)*2],ev[(c1-SSP)*2+1]); c2=0; NEXT; } else { (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); /* we have to check this c2 */ /* and no way to push back SSO */ c1 = c2; c2 = 0; goto euc_1byte_check; } } (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); NEXT; } else SEND; } else if(c1 < SSP && iconv != s_iconv) { /* strange code in EUC */ iconv = s_iconv; /* try SJIS */ c2 = c1; NEXT; } else { /* already established */ c2 = c1; NEXT; } } } else if ((c1 > SP) && (c1 != DEL)) { /* in case of Roman characters */ if (shift_mode) { c1 |= 0x80; /* output 1 shifted byte */ if(x0201_f && (!iso8859_f||input_mode==X0201) && SSP<=c1 && c1<0xe0 ) { if(dv[(c1-SSP)*2]||ev[(c1-SSP)*2]) { if ((c2 = GETC()) == EOF) { (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); LAST; } else if (c2==(0xde&0x7f)) { /* 濁点 */ (*oconv)(dv[(c1-SSP)*2],dv[(c1-SSP)*2+1]); c2=0; NEXT; } else if (c2==(0xdf&0x7f)&&ev[(c1-SSP)*2]) { /* 半濁点 */ (*oconv)(ev[(c1-SSP)*2],ev[(c1-SSP)*2+1]); c2=0; NEXT; } UNGETC(c2); c2 = 0; } (*oconv)(cv[(c1-SSP)*2],cv[(c1-SSP)*2+1]); NEXT; } else SEND; } else if(c1 == '(' && broken_f && input_mode == X0208 && !mime_mode ) { /* Try to recover missing escape */ if ((c1 = GETC()) == EOF) { (*oconv) (0, '('); LAST; } else { if (c1 == 'B' || c1 == 'J' || c1 == 'H') { input_mode = ASCII; shift_mode = FALSE; NEXT; } else { (*oconv) (0, '('); /* do not modify various input_mode */ /* It can be vt100 sequence */ SEND; } } } else if (input_mode == X0208) { /* in case of Kanji shifted */ c2 = c1; NEXT; /* goto next_byte */ } else if(mime_f && !mime_mode && c1 == '=') { if ((c1 = _GETC()) == EOF) { (*oconv) (0, '='); LAST; } else if (c1 == '?') { /* =? is mime conversion start sequence */ if(mime_begin() == EOF) /* check in detail */ LAST; else NEXT; } else { (*oconv) (0, '='); _UNGETC(c1); NEXT; } } else if(c1 == '$' && broken_f && !mime_mode) { /* try to recover missing escape */ if ((c1 = GETC()) == EOF) { (*oconv) (0, '$'); LAST; } else if(c1 == '@'|| c1 == 'B') { /* in case of Kanji in ESC sequence */ input_mode = X0208; shift_mode = FALSE; NEXT; } else { /* sorry */ (*oconv) (0, '$'); (*oconv) (0, c1); NEXT; } } else SEND; } else if (c1 == SI) { shift_mode = FALSE; NEXT; } else if (c1 == SO) { shift_mode = TRUE; NEXT; } else if (c1 == ESC ) { if ((c1 = GETC()) == EOF) { (*oconv) (0, ESC); LAST; } else if (c1 == '$') { if ((c1 = GETC()) == EOF) { (*oconv) (0, ESC); (*oconv) (0, '$'); LAST; } else if(c1 == '@'|| c1 == 'B') { /* This is kanji introduction */ input_mode = X0208; shift_mode = FALSE; NEXT; } else { (*oconv) (0, ESC); (*oconv) (0, '$'); (*oconv) (0, c1); NEXT; } } else if (c1 == '(') { if ((c1 = GETC()) == EOF) { (*oconv) (0, ESC); (*oconv) (0, '('); LAST; } else { if (c1 == 'I') { /* This is X0201 kana introduction */ input_mode = X0201; shift_mode = X0201; NEXT; } else if (c1 == 'B' || c1 == 'J' || c1 == 'H') { /* This is X0208 kanji introduction */ input_mode = ASCII; shift_mode = FALSE; NEXT; } else { (*oconv) (0, ESC); (*oconv) (0, '('); /* maintain various input_mode here */ SEND; } } } else { /* lonely ESC */ (*oconv) (0, ESC); SEND; } } else SEND; } /* send: */ if (input_mode == X0208) (*oconv) (c2, c1); /* this is JIS, not SJIS/EUC case */ else (*iconv) (c2, c1); /* can be EUC/SJIS */ c2 = 0; continue; /* goto next_word */ } /* epilogue */ (*iconv) (EOF, 0); return outlen; } static int h_conv (c2, c1) register int c1, c2; { register int wc; /** it must NOT be in the kanji shifte sequence */ /** it must NOT be written in JIS7 */ /** and it must be after 2 byte 8bit code */ hold_count = 0; push_hold_buf (c2, c1); c2 = 0; while ((c1 = GETC()) != EOF) { if (c2) { /* second byte */ if (!estab_f) { /* not established */ if (c1 > SSP) { /* it is still ambiguious yet */ SEND; } else if (c1 < AT) { /* ignore bogus first byte */ c2 = 0; SEND; } else { /* now established */ /* it seems to be MS Kanji */ estab_f = TRUE; iconv = s_iconv; SEND; } } else SEND; } else { /* First byte */ if (c1 > DEL) { /* 8th bit is on */ if (c1 < SSP) { /* it seems to be MS Kanji */ estab_f = TRUE; iconv = s_iconv; } else if (c1 < 0xe0) { /* it seems to be EUC */ estab_f = TRUE; iconv = oconv; } else { /* still ambiguious */ } c2 = c1; NEXT; } else /* 7 bit code , then send without any process */ SEND; } /* send: */ if ((push_hold_buf (c2, c1) == EOF) || estab_f) break; c2 = 0; continue; } /** now, ** 1) EOF is detected, or ** 2) Code is established, or ** 3) Buffer is FULL (but last word is pushed) ** ** in 1) and 3) cases, we continue to use ** Kanji codes by oconv and leave estab_f unchanged. **/ for (wc = 0; wc < hold_count; wc += 2) { c2 = hold_buf[wc]; c1 = hold_buf[wc+1]; (*iconv) (c2, c1); } return 1; } static int push_hold_buf (c2, c1) int c2, c1; { if (hold_count >= HOLD_SIZE*2) return (EOF); hold_buf[hold_count++] = c2; hold_buf[hold_count++] = c1; return ((hold_count >= HOLD_SIZE*2) ? EOF : hold_count); } static int s_iconv (c2, c1) register int c2, c1; { if ((c2 == EOF) || (c2 == 0)) { /* NOP */ } else { c2 = c2 + c2 - ((c2 <= 0x9f) ? SJ0162 : SJ6394); if (c1 < 0x9f) c1 = c1 - ((c1 > DEL) ? SP : 0x1f); else { c1 = c1 - 0x7e; c2++; } } (*oconv) (c2, c1); return 1; } static int e_oconv (c2, c1) register int c2, c1; { c2 = pre_convert(c1,c2); c1 = c1_return; if(fold_f) { switch(fold(c2,c1)) { case '\n': PUTCHAR('\n'); break; case 0: return 1; case '\r': c1 = '\n'; c2 = 0; break; case '\t': case ' ': c1 = ' '; c2 = 0; break; } } if (c2 == EOF) return 1; else if (c2 == 0 && (c1&0x80)) { PUTCHAR(SSO); PUTCHAR(c1); } else if (c2 == 0) { PUTCHAR(c1); } else { if((c1<0x20 || 0x7e> 1) + ((c2 <= 0x5e) ? 0x71 : 0xb1))); PUTCHAR((c1 + ((c2 & 1) ? ((c1 < 0x60) ? 0x1f : 0x20) : 0x7e))); } return 1; } static int j_oconv (c2, c1) register int c2, c1; { c2 = pre_convert(c1,c2); c1 = c1_return; if(fold_f) { switch(fold(c2,c1)) { case '\n': if (output_mode) { PUTCHAR(ESC); PUTCHAR('('); PUTCHAR(ascii_intro); } PUTCHAR('\n'); output_mode = ASCII; break; case '\r': c1 = '\n'; c2 = 0; break; case '\t': case ' ': c1 = ' '; c2 = 0; break; case 0: return 1; } } if (c2 == EOF) { if (output_mode) { PUTCHAR(ESC); PUTCHAR('('); PUTCHAR(ascii_intro); } } else if (c2 == 0 && (c1 & 0x80)) { if(input_mode==X0201 || !iso8859_f) { if(output_mode!=X0201) { PUTCHAR(ESC); PUTCHAR('('); PUTCHAR('I'); output_mode = X0201; } c1 &= 0x7f; } else { /* iso8859 introduction, or 8th bit on */ /* Can we convert in 7bit form using ESC-'-'-A ? Is this popular? */ } PUTCHAR(c1); } else if (c2 == 0) { if (output_mode) { PUTCHAR(ESC); PUTCHAR('('); PUTCHAR(ascii_intro); output_mode = ASCII; } PUTCHAR(c1); } else { if (output_mode != X0208) { PUTCHAR(ESC); PUTCHAR('$'); PUTCHAR(kanji_intro); output_mode = X0208; } if(c1<0x20 || 0x7e0x80 Japanese (X0208/X0201) <0x80 ASCII \n new line ' ' space This fold algorthm does not conserve heading space in a line. This is the main difference from fmt. */ static int fold(c2,c1) register int c2,c1; { int prev0; if(c1=='\r') return 0; /* ignore cr */ if(c1== 8) { if(line>0) line--; return 1; } if(c2==EOF && line != 0) /* close open last line */ return '\n'; /* new line */ if(c1=='\n') { if(prev == c1) { /* duplicate newline */ if(line) { line = 0; return '\n'; /* output two newline */ } else { line = 0; return 1; } } else { if(prev&0x80) { /* Japanese? */ prev = c1; return 0; /* ignore given single newline */ } else if(prev==' ') { return 0; } else { prev = c1; if(++line<=fold_len) return ' '; else { line = 0; return '\r'; /* fold and output nothing */ } } } } if(c1=='\f') { prev = '\n'; if(line==0) return 1; line = 0; return '\n'; /* output newline and clear */ } /* X0208 kankaku or ascii space */ if((c2==0&&c1==' ')|| (c2==0&&c1=='\t')|| (c2=='!'&& c1=='!')) { if(prev == ' ') { return 0; /* remove duplicate spaces */ } prev = ' '; if(++line<=fold_len) return ' '; /* output ASCII space only */ else { prev = ' '; line = 0; return '\r'; /* fold and output nothing */ } } prev0 = prev; /* we still need this one... , but almost done */ prev = c1; if (c2 || (SSP<=c1 && c1<=0xdf)) prev |= 0x80; /* this is Japanese */ line += (c2==0)?1:2; if(line<=fold_len) { /* normal case */ return 1; } if(line>=fold_len+FOLD_MARGIN) { /* too many kinsou suspension */ line = (c2==0)?1:2; return '\n'; /* We can't wait, do fold now */ } /* simple kinsoku rules return 1 means no folding */ if(c2==0) { if(c1==0xde) return 1; /* ゛*/ if(c1==0xdf) return 1; /* ゜*/ if(c1==0xa4) return 1; /* 。*/ if(c1==0xa3) return 1; /* ，*/ if(c1==0xa1) return 1; /* 」*/ if(c1==0xb0) return 1; /* - */ if(SSP<=c1 && c1<=0xdf) { /* X0201 */ line = 1; return '\n';/* add one new line before this character */ } /* fold point in ASCII { [ ( */ if(( c1!=')'&& c1!=']'&& c1!='}'&& c1!='.'&& c1!=','&& c1!='!'&& c1!='?'&& c1!='/'&& c1!=':'&& c1!=';')&& ((prev0=='\n')|| (prev0==' ')|| /* ignored new line */ (prev0&0x80)) /* X0208 - ASCII */ ) { line = 1; return '\n';/* add one new line before this character */ } return 1; /* default no fold in ASCII */ } else { if(c2=='!') { if(c1=='"') return 1; /* 、 */ if(c1=='#') return 1; /* 。 */ if(c1=='$') return 1; /* ， */ if(c1=='%') return 1; /* ． */ if(c1=='\'') return 1; /* ＋ */ if(c1=='(') return 1; /* ； */ if(c1==')') return 1; /* ？ */ if(c1=='*') return 1; /* ！ */ if(c1=='+') return 1; /* ゛ */ if(c1==',') return 1; /* ゜ */ } line = 2; return '\n'; /* add one new line before this character */ } } static int pre_convert(c1,c2) register int c1,c2; { if(c2) c1 &= 0x7f; c1_return = c1; if(c2==EOF) return c2; c2 &= 0x7f; if(rot_f) { if(c2) { c1 = rot47(c1); c2 = rot47(c2); } else { if (!(c1 & 0x80)) c1 = rot13(c1); } c1_return = c1; } /* JISX0208 Alphabet */ if (alpha_f && c2 == 0x23 ) return 0; /* JISX0208 Kigou */ if (alpha_f && c2 == 0x21 ) { if(0x20' ';i++) { if( ((c1 = _GETC())==EOF) || c1 != p[i] ) { q = p; while ((p=mime_pattern[++j]) != NULL) { for(k=2;k i */ if(p[k]!=q[k]) break; if(k==i && c1==p[k]) break; } if(p) continue; _UNGETC(c1); for(j=0;j> 4) & 0x03); if (c2 != '=') { Fifo(mime_last++) = cc; cc = ((t2 << 4) & 0x0f0) | ((t3 >> 2) & 0x0f); if (c3 != '=') { Fifo(mime_last++) = cc; cc = ((t3 << 6) & 0x0c0) | (t4 & 0x3f); if (c4 != '=') Fifo(mime_last++) = cc; } } else { return c1; } return Fifo(mime_top++); } static int mime_ungetc(c) unsigned int c; { Fifo(mime_last++) = c; return c; } static int base64decode(c) int c; { int i; if (c > '@') { if (c < '[') i = c - 'A'; /* A..Z 0-25 */ else i = c - 'G' /* - 'a' + 26 */ ; /* a..z 26-51 */ } else if (c > '/') i = c - '0' + '4' /* - '0' + 52 */ ; /* 0..9 52-61 */ else if (c == '+') i = '>' /* 62 */ ; /* + 62 */ else i = '?' /* 63 */ ; /* / 63 */ return (i); } #ifdef notdef int usage() { fprintf(stderr,"USAGE: nkf [flags] [file]\n"); fprintf(stderr,"Flags:\n"); fprintf(stderr,"b Output is bufferred (DEFAULT)\n"); fprintf(stderr,"u Output is unbufferred\n"); fprintf(stderr,"t no operation\n"); fprintf(stderr,"j Outout code is JIS 7 bit (DEFAULT)\n"); fprintf(stderr,"s Output code is Shift JIS\n"); fprintf(stderr,"e Output code is AT&T JIS (EUC)\n"); fprintf(stderr,"i_ Output sequence to designate JIS-kanji (DEFAULT B)\n"); fprintf(stderr,"o_ Output sequence to designate ASCII (DEFAULT B)\n"); fprintf(stderr,"r {de/en}crypt ROT13/47\n"); fprintf(stderr,"v Show this usage\n"); fprintf(stderr,"m MIME ISO-2022-JP decode. ISO8859-1 needs -l\n"); fprintf(stderr,"l ISO8859-1 (Latin-1) support\n"); fprintf(stderr,"f Folding: -f60 or -f \n"); fprintf(stderr,"Z Convert X0208 alphabet to ASCII\n"); fprintf(stderr,"X,x Assume X0201 kana in MS-Kanji, -x preserves X0201\n"); fprintf(stderr,"J Assume JIS 7 bit input\n"); fprintf(stderr,"E Assume AT&T JIS (EUC) input\n"); fprintf(stderr,"S Assume MS-Kanji input\n"); fprintf(stderr,"B Assume Broken JIS; ESC missing input\n"); #ifdef MSDOS fprintf(stderr,"T Text mode output\n"); #endif fprintf(stderr,"Network Kanji Filter Version %s (%s)\n",Version,Patchlevel); fprintf(stderr,"%s\n",CopyRight); return 0; } #endif /* defined(notdef) */ #if defined(KCONV_TEST) #include main(int argc, char **argv) { char b_old[1024]; char b_new[1024]; int code = _JIS; if (argc > 1) { if (!strcmp(argv[1], "-j")) { code = _JIS; } else if (!strcmp(argv[1], "-e")) { code = _EUC; } else if (!strcmp(argv[1], "-s")) { code = _SJIS; } } while (fgets(b_old, sizeof(b_old), stdin)) { bzero(b_new, sizeof(b_new)); do_kconv(b_old, b_new, sizeof(b_new), code, _AUTO); fputs(b_new, stdout); } } #endif /* defined(KCONV_TEST) */ /** ** パッチ制作者 ** void@merope.pleiades.or.jp (Kusakabe Youichi) ** NIDE Naoyuki ** ohta@src.ricoh.co.jp (Junn Ohta) ** inouet@strl.nhk.or.jp (Tomoyuki Inoue) ** kiri@pulser.win.or.jp (Tetsuaki Kiriyama) ** Kimihiko Sato ** a_kuroe@kuroe.aoba.yokohama.jp (Akihiko Kuroe) ** kono@csl.sony.co.jp (Shinji Kono) ** ikuo@isl.intec.co.jp (Ikuo Nakagawa) ** ** 最終更新日 ** 1996.03.10 **/ void CleanUpNKF() { } int InitNKF() { char *psz=NULL; useNKF = FALSE; if ((psz=XGetDefault(mainDisplay, TOOL_NAME, "UseNKF")) != NULL && UtilStrICmp(psz, "true") == 0) { useNKF = TRUE; } return TRUE; } #endif /* ~_NO_NKF */ /* end */