X-Git-Url: http://git.sourceforge.jp/view?p=nkf%2Fnkf.git;a=blobdiff_plain;f=nkf.c;h=505d19290fc58126e7d98b611bbe287ab318b356;hp=5547be34f39edfc442aaf5cd17ae7a8efbe25f2f;hb=477df6ae857330396b089025d986f949f27bc5ac;hpb=958de30bc09aef38f2a44b5da0dbb1bb3c79e7d3 diff --git a/nkf.c b/nkf.c index 5547be3..505d192 100644 --- a/nkf.c +++ b/nkf.c @@ -1,7 +1,7 @@ /** 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 +** 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） ** Copyright (C) 1996,1998 ** Copyright (C) 2002 @@ -17,85 +17,66 @@ ** バイナリの配布の際にはversion messageを保存することを条件とします。 ** 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 +** Everyone is permitted to do anything on this program ** including copying, modifying, improving, ** as long as you don't try to pretend that you wrote it. -** i.e., the above copyright notice has to appear in all copies. +** i.e., the above copyright notice has to appear in all copies. ** Binary distribution requires original version messages. ** You don't have to ask before copying, redistribution or publishing. ** THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE. ***********************************************************************/ /*********************************************************************** -** UTF-8 サポートについて -** 従来の nkf と入れかえてそのまま使えるようになっています -** nkf -e などとして起動すると、自動判別で UTF-8 と判定されれば、 -** そのまま euc-jp に変換されます -** -** まだバグがある可能性が高いです。 -** (特に自動判別、コード混在、エラー処理系) -** -** 何か問題を見つけたら、 -** E-Mail: furukawa@tcp-ip.or.jp -** まで御連絡をお願いします。 + * 現在、nkf は SorceForge にてメンテナンスが続けられています。 + * http://sourceforge.jp/projects/nkf/ ***********************************************************************/ -/* $Id: nkf.c,v 1.130 2007/08/31 14:06:08 naruse Exp $ */ +/* $Id: nkf.c,v 1.159 2007/12/23 07:55:20 naruse Exp $ */ #define NKF_VERSION "2.0.8" -#define NKF_RELEASE_DATE "2007-08-31" -#include "config.h" -#include "utf8tbl.h" - +#define NKF_RELEASE_DATE "2007-12-22" #define COPY_RIGHT \ "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa),2000 S. Kono, COW\n" \ "Copyright (C) 2002-2007 Kono, Furukawa, Naruse, mastodon" +#include "config.h" +#include "utf8tbl.h" -/* -** -** -** -** USAGE: nkf [flags] [file] -** -** Flags: -** b Output is buffered (DEFAULT) -** u Output is unbuffered -** -** t no operation -** -** j Output code is JIS 7 bit (DEFAULT SELECT) -** s Output code is MS Kanji (DEFAULT SELECT) -** e Output code is AT&T JIS (DEFAULT SELECT) -** w Output code is AT&T JIS (DEFAULT SELECT) -** l Output code is JIS 7bit and ISO8859-1 Latin-1 -** -** m MIME conversion for ISO-2022-JP -** I Convert non ISO-2022-JP charactor to GETA by Pekoe -** i_ Output sequence to designate JIS-kanji (DEFAULT_J) -** o_ Output sequence to designate single-byte roman characters (DEFAULT_R) -** M MIME output conversion -** -** 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 -** B[1-9] broken level -** -** O Output to 'nkf.out' file or last file name -** d Delete \r in line feed -** c Add \r in line feed -** -- other long option -** -- ignore following option (don't use with -O ) -** -**/ +#if defined(DEFAULT_CODE_JIS) +#elif defined(DEFAULT_CODE_SJIS) +#elif defined(DEFAULT_CODE_EUC) +#elif defined(DEFAULT_CODE_UTF8) +#else +#define DEFAULT_CODE_JIS 1 +#endif + +#ifndef MIME_DECODE_DEFAULT +#define MIME_DECODE_DEFAULT STRICT_MIME +#endif +#ifndef X0201_DEFAULT +#define X0201_DEFAULT TRUE +#endif + +#if DEFAULT_NEWLINE == 0x0D0A +#define PUT_NEWLINE(func) do {\ + func(0x0D);\ + func(0x0A);\ +} while (0) +#define OCONV_NEWLINE(func) do {\ + func(0, 0x0D);\ + func(0, 0x0A);\ +} while (0) +#elif DEFAULT_NEWLINE == 0x0D +#define PUT_NEWLINE(func) func(0x0D) +#define OCONV_NEWLINE(func) func(0, 0x0D) +#else +#define DEFAULT_NEWLINE 0x0A +#define PUT_NEWLINE(func) func(0x0A) +#define OCONV_NEWLINE(func) func(0, 0x0A) +#endif +#ifdef HELP_OUTPUT_STDERR +#define HELP_OUTPUT stderr +#else +#define HELP_OUTPUT stdout +#endif #if (defined(__TURBOC__) || defined(_MSC_VER) || defined(LSI_C) || defined(__MINGW32__) || defined(__EMX__) || defined(__MSDOS__) || defined(__WINDOWS__) || defined(__DOS__) || defined(__OS2__)) && !defined(MSDOS) #define MSDOS @@ -196,45 +177,32 @@ void djgpp_setbinmode(FILE *fp) #define FALSE 0 #define TRUE 1 -/* state of output_mode and input_mode +/* state of output_mode and input_mode c2 0 means ASCII - X0201 - ISO8859_1 - X0208 + JIS_X_0201 + ISO_8859_1 + JIS_X_0208 EOF all termination c1 32bit data */ -#define ASCII 0 -#define X0208 1 -#define X0201 2 -#define ISO8859_1 8 -#define NO_X0201 3 -#define X0212 0x2844 -#define X0213_1 0x284F -#define X0213_2 0x2850 - /* Input Assumption */ #define JIS_INPUT 4 #define EUC_INPUT 16 #define SJIS_INPUT 5 #define LATIN1_INPUT 6 +#define UTF8_INPUT 13 +#define UTF16_INPUT 1015 +#define UTF32_INPUT 1017 + #define FIXED_MIME 7 #define STRICT_MIME 8 /* MIME ENCODE */ -#define ISO2022JP 9 -#define JAPANESE_EUC 10 -#define SHIFT_JIS 11 - -#define UTF8 12 -#define UTF8_INPUT 13 -#define UTF16_INPUT 1015 -#define UTF32_INPUT 1017 /* byte order */ @@ -243,16 +211,14 @@ void djgpp_setbinmode(FILE *fp) #define ENDIAN_2143 2143 #define ENDIAN_3412 3412 -#define WISH_TRUE 15 - /* ASCII CODE */ #define BS 0x08 #define TAB 0x09 -#define NL 0x0a +#define LF 0x0a #define CR 0x0d #define ESC 0x1b -#define SPACE 0x20 +#define SP 0x20 #define AT 0x40 #define SSP 0xa0 #define DEL 0x7f @@ -260,6 +226,181 @@ void djgpp_setbinmode(FILE *fp) #define SO 0x0e #define SSO 0x8e #define SS3 0x8f +#define CRLF 0x0D0A + + +/* encodings */ + +enum nkf_encodings { + ASCII, + ISO_8859_1, + ISO_2022_JP, + CP50220, + CP50221, + CP50222, + ISO_2022_JP_1, + ISO_2022_JP_3, + SHIFT_JIS, + WINDOWS_31J, + CP10001, + EUC_JP, + CP51932, + EUCJP_MS, + EUCJP_ASCII, + SHIFT_JISX0213, + SHIFT_JIS_2004, + EUC_JISX0213, + EUC_JIS_2004, + UTF_8, + UTF_8N, + UTF_8_BOM, + UTF8_MAC, + UTF_16, + UTF_16BE, + UTF_16BE_BOM, + UTF_16LE, + UTF_16LE_BOM, + UTF_32, + UTF_32BE, + UTF_32BE_BOM, + UTF_32LE, + UTF_32LE_BOM, + JIS_X_0201=0x1000, + JIS_X_0208, + JIS_X_0212, + JIS_X_0213_1, + JIS_X_0213_2, + BINARY +}; + +nkf_char s_iconv(nkf_char c2, nkf_char c1, nkf_char c0); +nkf_char e_iconv(nkf_char c2, nkf_char c1, nkf_char c0); +nkf_char w_iconv(nkf_char c2, nkf_char c1, nkf_char c0); +nkf_char w_iconv16(nkf_char c2, nkf_char c1, nkf_char c0); +nkf_char w_iconv32(nkf_char c2, nkf_char c1, nkf_char c0); +void j_oconv(nkf_char c2, nkf_char c1); +void s_oconv(nkf_char c2, nkf_char c1); +void e_oconv(nkf_char c2, nkf_char c1); +void w_oconv(nkf_char c2, nkf_char c1); +void w_oconv16(nkf_char c2, nkf_char c1); +void w_oconv32(nkf_char c2, nkf_char c1); + +typedef struct { + char *name; + nkf_char (*iconv_func)(nkf_char c2, nkf_char c1, nkf_char c0); + void (*oconv_func)(nkf_char c2, nkf_char c1); +} nkf_native_encoding; + +nkf_native_encoding NkfEncodingASCII = { "US_ASCII", e_iconv, e_oconv }; +nkf_native_encoding NkfEncodingISO_2022_JP = { "ISO-2022-JP", e_iconv, j_oconv }; +nkf_native_encoding NkfEncodingShift_JIS = { "Shift_JIS", s_iconv, s_oconv }; +nkf_native_encoding NkfEncodingEUC_JP = { "EUC-JP", e_iconv, e_oconv }; +nkf_native_encoding NkfEncodingUTF_8 = { "UTF-8", w_iconv, w_oconv }; +nkf_native_encoding NkfEncodingUTF_16 = { "UTF-16", w_iconv16, w_oconv16 }; +nkf_native_encoding NkfEncodingUTF_32 = { "UTF-32", w_iconv32, w_oconv32 }; + +typedef struct { + int id; + char *name; + nkf_native_encoding *based_encoding; +} nkf_encoding; +nkf_encoding nkf_encoding_table[] = { + {ASCII, "ASCII", &NkfEncodingASCII}, + {ISO_8859_1, "ISO-8859-1", &NkfEncodingASCII}, + {ISO_2022_JP, "ISO-2022-JP", &NkfEncodingASCII}, + {CP50220, "CP50220", &NkfEncodingISO_2022_JP}, + {CP50221, "CP50221", &NkfEncodingISO_2022_JP}, + {CP50222, "CP50222", &NkfEncodingISO_2022_JP}, + {ISO_2022_JP_1, "ISO-2022-JP-1", &NkfEncodingISO_2022_JP}, + {ISO_2022_JP_3, "ISO-2022-JP-3", &NkfEncodingISO_2022_JP}, + {SHIFT_JIS, "Shift_JIS", &NkfEncodingShift_JIS}, + {WINDOWS_31J, "WINDOWS-31J", &NkfEncodingShift_JIS}, + {CP10001, "CP10001", &NkfEncodingShift_JIS}, + {EUC_JP, "EUC-JP", &NkfEncodingEUC_JP}, + {CP51932, "CP51932", &NkfEncodingEUC_JP}, + {EUCJP_MS, "eucJP-MS", &NkfEncodingEUC_JP}, + {EUCJP_ASCII, "eucJP-ASCII", &NkfEncodingEUC_JP}, + {SHIFT_JISX0213, "Shift_JISX0213", &NkfEncodingShift_JIS}, + {SHIFT_JIS_2004, "Shift_JIS-2004", &NkfEncodingShift_JIS}, + {EUC_JISX0213, "EUC-JISX0213", &NkfEncodingEUC_JP}, + {EUC_JIS_2004, "EUC-JIS-2004", &NkfEncodingEUC_JP}, + {UTF_8, "UTF-8", &NkfEncodingUTF_8}, + {UTF_8N, "UTF-8N", &NkfEncodingUTF_8}, + {UTF_8_BOM, "UTF-8-BOM", &NkfEncodingUTF_8}, + {UTF8_MAC, "UTF8-MAC", &NkfEncodingUTF_8}, + {UTF_16, "UTF-16", &NkfEncodingUTF_16}, + {UTF_16BE, "UTF-16BE", &NkfEncodingUTF_16}, + {UTF_16BE_BOM, "UTF-16BE-BOM", &NkfEncodingUTF_16}, + {UTF_16LE, "UTF-16LE", &NkfEncodingUTF_16}, + {UTF_16LE_BOM, "UTF-16LE-BOM", &NkfEncodingUTF_16}, + {UTF_32, "UTF-32", &NkfEncodingUTF_32}, + {UTF_32BE, "UTF-32BE", &NkfEncodingUTF_32}, + {UTF_32BE_BOM, "UTF-32BE-BOM", &NkfEncodingUTF_32}, + {UTF_32LE, "UTF-32LE", &NkfEncodingUTF_32}, + {UTF_32LE_BOM, "UTF-32LE-BOM", &NkfEncodingUTF_32}, + {BINARY, "BINARY", &NkfEncodingASCII}, + {-1, NULL, NULL} +}; +#define NKF_ENCODING_TABLE_SIZE 34 +struct { + const char *name; + const int id; +} encoding_name_to_id_table[] = { + {"ASCII", ASCII}, + {"ISO-2022-JP", ISO_2022_JP}, + {"X-ISO2022JP-CP932", CP50220}, + {"CP50220", CP50220}, + {"CP50221", CP50221}, + {"CP50222", CP50222}, + {"ISO-2022-JP-1", ISO_2022_JP_1}, + {"ISO-2022-JP-3", ISO_2022_JP_3}, + {"SHIFT_JIS", SHIFT_JIS}, + {"SJIS", SHIFT_JIS}, + {"WINDOWS-31J", WINDOWS_31J}, + {"CSWINDOWS31J", WINDOWS_31J}, + {"CP932", WINDOWS_31J}, + {"MS932", WINDOWS_31J}, + {"CP10001", CP10001}, + {"EUCJP", EUC_JP}, + {"EUC-JP", EUC_JP}, + {"CP51932", CP51932}, + {"EUC-JP-MS", EUCJP_MS}, + {"EUCJP-MS", EUCJP_MS}, + {"EUCJPMS", EUCJP_MS}, + {"EUC-JP-ASCII", EUCJP_ASCII}, + {"EUCJP-ASCII", EUCJP_ASCII}, + {"SHIFT_JISX0213", SHIFT_JISX0213}, + {"SHIFT_JIS-2004", SHIFT_JIS_2004}, + {"EUC-JISX0213", EUC_JISX0213}, + {"EUC-JIS-2004", EUC_JIS_2004}, + {"UTF-8", UTF_8}, + {"UTF-8N", UTF_8N}, + {"UTF-8-BOM", UTF_8_BOM}, + {"UTF8-MAC", UTF8_MAC}, + {"UTF-8-MAC", UTF8_MAC}, + {"UTF-16", UTF_16}, + {"UTF-16BE", UTF_16BE}, + {"UTF-16BE-BOM", UTF_16BE_BOM}, + {"UTF-16LE", UTF_16LE}, + {"UTF-16LE-BOM", UTF_16LE_BOM}, + {"UTF-32", UTF_32}, + {"UTF-32BE", UTF_32BE}, + {"UTF-32BE-BOM", UTF_32BE_BOM}, + {"UTF-32LE", UTF_32LE}, + {"UTF-32LE-BOM", UTF_32LE_BOM}, + {"BINARY", BINARY}, + {NULL, -1} +}; +#if defined(DEFAULT_CODE_JIS) +#define DEFAULT_ENCODING ISO_2022_JP +#elif defined(DEFAULT_CODE_SJIS) +#define DEFAULT_ENCODING SHIFT_JIS +#elif defined(DEFAULT_CODE_EUC) +#define DEFAULT_ENCODING EUC_JP +#elif defined(DEFAULT_CODE_UTF8) +#define DEFAULT_ENCODING UTF_8 +#endif + #define is_alnum(c) \ (('a'<=c && c<='z')||('A'<= c && c<='Z')||('0'<=c && c<='9')) @@ -269,16 +410,20 @@ void djgpp_setbinmode(FILE *fp) #define nkf_isoctal(c) ('0'<=c && c<='7') #define nkf_isdigit(c) ('0'<=c && c<='9') #define nkf_isxdigit(c) (nkf_isdigit(c) || ('a'<=c && c<='f') || ('A'<=c && c <= 'F')) -#define nkf_isblank(c) (c == SPACE || c == TAB) -#define nkf_isspace(c) (nkf_isblank(c) || c == CR || c == NL) +#define nkf_isblank(c) (c == SP || c == TAB) +#define nkf_isspace(c) (nkf_isblank(c) || c == CR || c == LF) #define nkf_isalpha(c) (('a' <= c && c <= 'z') || ('A' <= c && c <= 'Z')) #define nkf_isalnum(c) (nkf_isdigit(c) || nkf_isalpha(c)) -#define nkf_isprint(c) (' '<=c && c<='~') +#define nkf_isprint(c) (SP<=c && c<='~') #define nkf_isgraph(c) ('!'<=c && c<='~') #define hex2bin(c) (('0'<=c&&c<='9') ? (c-'0') : \ ('A'<=c&&c<='F') ? (c-'A'+10) : \ - ('a'<=c&&c<='f') ? (c-'a'+10) : 0 ) + ('a'<=c&&c<='f') ? (c-'a'+10) : 0) +#define bin2hex(c) ("0123456789ABCDEF"[c&15]) #define is_eucg3(c2) (((unsigned short)c2 >> 8) == SS3) +#define nkf_noescape_mime(c) ((c == CR) || (c == LF) || \ + ((c > SP) && (c < DEL) && (c != '?') && (c != '=') && (c != '_') \ + && (c != '(') && (c != ')') && (c != '.') && (c != 0x22))) #define CP932_TABLE_BEGIN 0xFA #define CP932_TABLE_END 0xFC @@ -330,11 +475,9 @@ struct input_code{ int _file_stat; }; -static char *input_codename = ""; +static char *input_codename = NULL; /* NULL: unestablished, "": BINARY */ +static nkf_encoding *output_encoding; -#ifndef PERL_XS -static const char *CopyRight = COPY_RIGHT; -#endif #if !defined(PERL_XS) && !defined(WIN32DLL) static nkf_char noconvert(FILE *f); #endif @@ -343,9 +486,7 @@ static nkf_char kanji_convert(FILE *f); static nkf_char h_conv(FILE *f,nkf_char c2,nkf_char c1); static nkf_char push_hold_buf(nkf_char c2); static void set_iconv(nkf_char f, nkf_char (*iconv_func)(nkf_char c2,nkf_char c1,nkf_char c0)); -static nkf_char s_iconv(nkf_char c2,nkf_char c1,nkf_char c0); static nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1); -static nkf_char e_iconv(nkf_char c2,nkf_char c1,nkf_char c0); #if defined(UTF8_INPUT_ENABLE) || defined(UTF8_OUTPUT_ENABLE) /* UCS Mapping * 0: Shift_JIS, eucJP-ascii @@ -374,9 +515,6 @@ static void encode_fallback_perl(nkf_char c); static void encode_fallback_subchar(nkf_char c); static void (*encode_fallback)(nkf_char c) = NULL; static nkf_char w2e_conv(nkf_char c2,nkf_char c1,nkf_char c0,nkf_char *p2,nkf_char *p1); -static nkf_char w_iconv(nkf_char c2,nkf_char c1,nkf_char c0); -static nkf_char w_iconv16(nkf_char c2,nkf_char c1,nkf_char c0); -static nkf_char w_iconv32(nkf_char c2,nkf_char c1,nkf_char c0); static nkf_char unicode_to_jis_common(nkf_char c2,nkf_char c1,nkf_char c0,nkf_char *p2,nkf_char *p1); static nkf_char w_iconv_common(nkf_char c1,nkf_char c0,const unsigned short *const *pp,nkf_char psize,nkf_char *p2,nkf_char *p1); static void w16w_conv(nkf_char val, nkf_char *p2, nkf_char *p1, nkf_char *p0); @@ -388,16 +526,10 @@ static void w_status(struct input_code *, nkf_char); static int output_bom_f = FALSE; static int output_endian = ENDIAN_BIG; static nkf_char e2w_conv(nkf_char c2,nkf_char c1); -static void w_oconv(nkf_char c2,nkf_char c1); -static void w_oconv16(nkf_char c2,nkf_char c1); -static void w_oconv32(nkf_char c2,nkf_char c1); #endif -static void e_oconv(nkf_char c2,nkf_char c1); static nkf_char e2s_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1); -static void s_oconv(nkf_char c2,nkf_char c1); -static void j_oconv(nkf_char c2,nkf_char c1); static void fold_conv(nkf_char c2,nkf_char c1); -static void cr_conv(nkf_char c2,nkf_char c1); +static void nl_conv(nkf_char c2,nkf_char c1); static void z_conv(nkf_char c2,nkf_char c1); static void rot_conv(nkf_char c2,nkf_char c1); static void hira_conv(nkf_char c2,nkf_char c1); @@ -437,11 +569,10 @@ static void mimeout_addchar(nkf_char c); #ifndef PERL_XS static void usage(void); static void version(void); +static void show_configuration(void); #endif static void options(unsigned char *c); -#if defined(PERL_XS) || defined(WIN32DLL) static void reinit(void); -#endif /* buffers */ @@ -455,7 +586,7 @@ static int hold_count = 0; /* MIME preprocessor fifo */ #define MIME_BUF_SIZE (1024) /* 2^n ring buffer */ -#define MIME_BUF_MASK (MIME_BUF_SIZE-1) +#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; @@ -472,18 +603,14 @@ static int rot_f = FALSE; /* rot14/43 mode */ static int hira_f = FALSE; /* hira/kata henkan */ static int input_f = FALSE; /* non fixed input code */ static int alpha_f = FALSE; /* convert JIx0208 alphbet to ASCII */ -static int mime_f = STRICT_MIME; /* convert MIME B base64 or Q */ +static int mime_f = MIME_DECODE_DEFAULT; /* convert MIME B base64 or Q */ static int mime_decode_f = FALSE; /* mime decode is explicitly on */ static int mimebuf_f = FALSE; /* MIME buffered input */ static int broken_f = FALSE; /* convert ESC-less broken JIS */ static int iso8859_f = FALSE; /* ISO8859 through */ static int mimeout_f = FALSE; /* base64 mode */ -#if defined(MSDOS) || defined(__OS2__) -static int x0201_f = TRUE; /* Assume JISX0201 kana */ -#else -static int x0201_f = NO_X0201; /* Assume NO JISX0201 */ -#endif -static int iso2022jp_f = FALSE; /* convert ISO-2022-JP */ +static int x0201_f = X0201_DEFAULT; /* convert JIS X 0201 */ +static int iso2022jp_f = FALSE; /* replace non ISO-2022-JP with GETA */ #ifdef UNICODE_NORMALIZATION static int nfc_f = FALSE; @@ -531,18 +658,16 @@ static nkf_char numchar_ungetc(nkf_char c,FILE *f); #ifdef CHECK_OPTION static int noout_f = FALSE; static void no_putc(nkf_char c); -static nkf_char debug_f = FALSE; +static int debug_f = FALSE; static void debug(const char *str); static nkf_char (*iconv_for_check)(nkf_char c2,nkf_char c1,nkf_char c0) = 0; #endif -static int guess_f = FALSE; +static int guess_f = 0; /* 0: OFF, 1: ON, 2: VERBOSE */ #if !defined PERL_XS static void print_guessed_code(char *filename); #endif static void set_input_codename(char *codename); -static int is_inputcode_mixed = FALSE; -static int is_inputcode_set = FALSE; #ifdef EXEC_IO static int exec_f = 0; @@ -589,7 +714,7 @@ struct input_code input_code_list[] = { {0} }; -static int mimeout_mode = 0; +static int mimeout_mode = 0; /* 0, -1, 'Q', 'B', 1, 2 */ static int base64_count = 0; /* X0208 -> ASCII converter */ @@ -636,7 +761,7 @@ static nkf_char (*iconv)(nkf_char c2,nkf_char c1,nkf_char c0) = no_connection2; static void (*o_zconv)(nkf_char c2,nkf_char c1) = no_connection; static void (*o_fconv)(nkf_char c2,nkf_char c1) = no_connection; -static void (*o_crconv)(nkf_char c2,nkf_char c1) = no_connection; +static void (*o_nlconv)(nkf_char c2,nkf_char c1) = no_connection; static void (*o_rot_conv)(nkf_char c2,nkf_char c1) = no_connection; static void (*o_hira_conv)(nkf_char c2,nkf_char c1) = no_connection; static void (*o_base64conv)(nkf_char c2,nkf_char c1) = no_connection; @@ -671,8 +796,7 @@ static int mime_decode_mode = FALSE; /* MIME mode B base64, Q hex */ /* X0201 kana conversion table */ /* 90-9F A0-DF */ -static const -unsigned char cv[]= { +static const 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, @@ -694,8 +818,7 @@ unsigned char cv[]= { /* X0201 kana conversion table for daguten */ /* 90-9F A0-DF */ -static const -unsigned char dv[]= { +static const 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, @@ -716,8 +839,7 @@ unsigned char dv[]= { /* X0201 kana conversion table for han-daguten */ /* 90-9F A0-DF */ -static const -unsigned char ev[]= { +static const 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, @@ -739,8 +861,7 @@ unsigned char ev[]= { /* X0208 kigou conversion table */ /* 0x8140 - 0x819e */ -static const -unsigned char fv[] = { +static const unsigned char fv[] = { 0x00,0x00,0x00,0x00,0x2c,0x2e,0x00,0x3a, 0x3b,0x3f,0x21,0x00,0x00,0x27,0x60,0x00, @@ -757,7 +878,6 @@ unsigned char fv[] = { } ; -#define CRLF 1 static int file_out_f = FALSE; #ifdef OVERWRITE @@ -768,8 +888,9 @@ static char *backup_suffix = ""; static char *get_backup_filename(const char *suffix, const char *filename); #endif -static int crmode_f = 0; /* CR, NL, CRLF */ -static nkf_char prev_cr = 0; +static int nlmode_f = 0; /* CR, LF, CRLF */ +static int input_newline = 0; /* 0: unestablished, EOF: MIXED */ +static nkf_char prev_cr = 0; /* CR or 0 */ #ifdef EASYWIN /*Easy Win */ static int end_check; #endif /*Easy Win */ @@ -778,6 +899,57 @@ static int end_check; nkf_char std_gc_buf[STD_GC_BUFSIZE]; nkf_char std_gc_ndx; +char* nkf_strcpy(const char *str) +{ + char* result = malloc(strlen(str) + 1); + if (!result){ + perror(str); + return ""; + } + strcpy(result, str); + return result; +} + +static void nkf_str_upcase(const char *src, char *dest, size_t length) +{ + int i = 0; + for (; i < length && dest[i]; i++) { + dest[i] = nkf_toupper(src[i]); + } + dest[i] = 0; +} + +static nkf_encoding *nkf_enc_from_index(int idx) +{ + if (idx < 0 || NKF_ENCODING_TABLE_SIZE <= idx) { + return 0; + } + return &nkf_encoding_table[idx]; +} + +static int nkf_enc_find_index(const char *name) +{ + int i, index = -1; + for (i = 0; encoding_name_to_id_table[i].id >= 0; i++) { + if (strcmp(name, encoding_name_to_id_table[i].name) == 0) { + return encoding_name_to_id_table[i].id; + } + } + return index; +} + +static nkf_encoding *nkf_enc_find(const char *name) +{ + int idx = -1; + idx = nkf_enc_find_index(name); + if (idx < 0) return 0; + return nkf_enc_from_index(idx); +} + +#define nkf_enc_name(enc) (enc)->name +#define nkf_enc_to_index(enc) (enc)->id +#define nkf_enc_to_base_encoding(enc) (enc)->based_encoding + #ifdef WIN32DLL #include "nkf32dll.c" #elif defined(PERL_XS) @@ -797,6 +969,32 @@ int main(int argc, char **argv) for (argc--,argv++; (argc > 0) && **argv == '-'; argc--, argv++) { cp = (unsigned char *)*argv; options(cp); + if (guess_f) { +#ifdef CHECK_OPTION + int debug_f_back = debug_f; +#endif +#ifdef EXEC_IO + int exec_f_back = exec_f; +#endif +#ifdef X0212_ENABLE + int x0212_f_back = x0212_f; +#endif + int x0213_f_back = x0213_f; + int guess_f_back = guess_f; + reinit(); + guess_f = guess_f_back; + mime_f = FALSE; +#ifdef CHECK_OPTION + debug_f = debug_f_back; +#endif +#ifdef EXEC_IO + exec_f = exec_f_back; +#endif +#ifdef X0212_ENABLE + x0212_f = x0212_f_back; +#endif + x0213_f = x0213_f_back; + } #ifdef EXEC_IO if (exec_f){ int fds[2], pid; @@ -825,12 +1023,10 @@ int main(int argc, char **argv) } #endif } - if(x0201_f == WISH_TRUE) - x0201_f = ((!iso2022jp_f)? TRUE : NO_X0201); if (binmode_f == TRUE) #if defined(__OS2__) && (defined(__IBMC__) || defined(__IBMCPP__)) - if (freopen("","wb",stdout) == NULL) + if (freopen("","wb",stdout) == NULL) return (-1); #else setbinmode(stdout); @@ -859,15 +1055,13 @@ int main(int argc, char **argv) int nfiles = argc; int is_argument_error = FALSE; while (argc--) { - is_inputcode_mixed = FALSE; - is_inputcode_set = FALSE; - input_codename = ""; + input_codename = NULL; + input_newline = 0; #ifdef CHECK_OPTION iconv_for_check = 0; #endif if ((fin = fopen((origfname = *argv++), "r")) == NULL) { - perror(*--argv); - *argv++; + perror(*(argv-1)); is_argument_error = TRUE; continue; } else { @@ -916,7 +1110,7 @@ int main(int argc, char **argv) } }else #endif - if(argc == 1 ) { + if(argc == 1) { outfname = *argv++; argc--; } else { @@ -929,7 +1123,7 @@ int main(int argc, char **argv) } if (binmode_f == TRUE) { #if defined(__OS2__) && (defined(__IBMC__) || defined(__IBMCPP__)) - if (freopen("","wb",stdout) == NULL) + if (freopen("","wb",stdout) == NULL) return (-1); #else setbinmode(stdout); @@ -938,11 +1132,11 @@ int main(int argc, char **argv) } if (binmode_f == TRUE) #if defined(__OS2__) && (defined(__IBMC__) || defined(__IBMCPP__)) - if (freopen("","rb",fin) == NULL) + if (freopen("","rb",fin) == NULL) return (-1); #else setbinmode(fin); -#endif +#endif setvbuffer(fin, (char *) stdibuf, IOBUF_SIZE); if (nop_f) noconvert(fin); @@ -1021,12 +1215,12 @@ int main(int argc, char **argv) return(-1); } #ifdef EASYWIN /*Easy Win */ - if (file_out_f == FALSE) + if (file_out_f == FALSE) scanf("%d",&end_check); - else + else fclose(stdout); #else /* for Other OS */ - if (file_out_f == TRUE) + if (file_out_f == TRUE) fclose(stdout); #endif /*Easy Win */ return (0); @@ -1074,8 +1268,7 @@ char *get_backup_filename(const char *suffix, const char *filename) } #endif -static const -struct { +static const struct { const char *name; const char *alias; } long_option[] = { @@ -1100,7 +1293,8 @@ struct { {"hiragana","h1"}, {"katakana","h2"}, {"katakana-hiragana","h3"}, - {"guess", "g"}, + {"guess=", ""}, + {"guess", "g1"}, {"cp932", ""}, {"no-cp932", ""}, #ifdef X0212_ENABLE @@ -1162,6 +1356,7 @@ void options(unsigned char *cp) unsigned char *p; unsigned char *cp_back = NULL; char codeset[32]; + nkf_encoding *enc; if (option_mode==1) return; @@ -1175,36 +1370,38 @@ void options(unsigned char *cp) p = 0; switch (*cp++) { case '-': /* literal options */ - if (!*cp || *cp == SPACE) { /* ignore the rest of arguments */ + if (!*cp || *cp == SP) { /* ignore the rest of arguments */ option_mode = 1; return; } for (i=0;ihiragana bit:2 hiragana->katakana */ - if ('9'>= *cp && *cp>='0') + if ('9'>= *cp && *cp>='0') hira_f |= (*cp++ -'0'); - else + else hira_f |= 1; continue; case 'r': rot_f = TRUE; continue; -#if defined(MSDOS) || defined(__OS2__) +#if defined(MSDOS) || defined(__OS2__) case 'T': binmode_f = FALSE; continue; #endif #ifndef PERL_XS case 'V': - version(); + show_configuration(); exit(1); break; case 'v': @@ -1737,16 +1967,22 @@ void options(unsigned char *cp) output_conv = w_oconv; cp++; if (cp[0] == '0'){ cp++; + output_encoding = nkf_enc_from_index(UTF_8N); } else { output_bom_f = TRUE; + output_encoding = nkf_enc_from_index(UTF_8_BOM); } } else { + int enc_idx; if ('1'== cp[0] && '6'==cp[1]) { output_conv = w_oconv16; cp+=2; + enc_idx = UTF_16; } else if ('3'== cp[0] && '2'==cp[1]) { output_conv = w_oconv32; cp+=2; + enc_idx = UTF_32; } else { output_conv = w_oconv; + output_encoding = nkf_enc_from_index(UTF_8); continue; } if (cp[0]=='L') { @@ -1755,13 +1991,21 @@ void options(unsigned char *cp) } else if (cp[0] == 'B') { cp++; } else { + output_encoding = nkf_enc_from_index(enc_idx); continue; } if (cp[0] == '0'){ cp++; + enc_idx = enc_idx == UTF_16 + ? (output_endian == ENDIAN_LITTLE ? UTF_16LE : UTF_16BE) + : (output_endian == ENDIAN_LITTLE ? UTF_32LE : UTF_32BE); } else { output_bom_f = TRUE; + enc_idx = enc_idx == UTF_16 + ? (output_endian == ENDIAN_LITTLE ? UTF_16LE_BOM : UTF_16BE_BOM) + : (output_endian == ENDIAN_LITTLE ? UTF_32LE_BOM : UTF_32BE_BOM); } + output_encoding = nkf_enc_from_index(enc_idx); } continue; #endif @@ -1801,7 +2045,6 @@ void options(unsigned char *cp) 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 @@ -1811,10 +2054,10 @@ void options(unsigned char *cp) bit:3 Convert HTML Entity bit:4 Convert JIS X 0208 Katakana to JIS X 0201 Katakana */ - if ('9'>= *cp && *cp>='0') - alpha_f |= 1<<(*cp++ -'0'); - else - alpha_f |= TRUE; + while ('0'<= *cp && *cp <='9') { + alpha_f |= 1 << (*cp++ - '0'); + } + if (!alpha_f) alpha_f = 1; continue; case 'x': /* Convert X0201 kana to X0208 or X0201 Conversion */ x0201_f = FALSE; /* No X0201->X0208 conversion */ @@ -1822,15 +2065,14 @@ void options(unsigned char *cp) 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) + MS Kanji (0xa0-0xdf) output X0201 ESC-(-I in JIS (0x20-0x5f) SSO in EUC (0xa0-0xdf) - 0xa0-0xd in MS Kanji (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 */ + case 'X': /* Convert X0201 kana to X0208 */ x0201_f = TRUE; continue; case 'F': /* prserve new lines */ @@ -1842,7 +2084,7 @@ void options(unsigned char *cp) fold_len *= 10; fold_len += *cp++ - '0'; } - if (!(0= *cp && *cp>='0') + if ('9'>= *cp && *cp>='0') broken_f |= 1<<(*cp++ -'0'); - else + else broken_f |= TRUE; continue; #ifndef PERL_XS @@ -1894,35 +2136,44 @@ void options(unsigned char *cp) continue; #endif case 'c':/* add cr code */ - crmode_f = CRLF; + nlmode_f = CRLF; continue; case 'd':/* delete cr code */ - crmode_f = NL; - continue; - case 'I': /* ISO-2022-JP output */ - iso2022jp_f = TRUE; + nlmode_f = LF; continue; + case 'I': /* ISO-2022-JP output */ + iso2022jp_f = TRUE; + continue; case 'L': /* line mode */ if (*cp=='u') { /* unix */ - crmode_f = NL; cp++; + nlmode_f = LF; cp++; } else if (*cp=='m') { /* mac */ - crmode_f = CR; cp++; + nlmode_f = CR; cp++; } else if (*cp=='w') { /* windows */ - crmode_f = CRLF; cp++; + nlmode_f = CRLF; cp++; } else if (*cp=='0') { /* no conversion */ - crmode_f = 0; cp++; + nlmode_f = 0; cp++; } continue; - case 'g': #ifndef PERL_XS - guess_f = TRUE; -#endif + case 'g': + if (*cp == '1') { + guess_f = 2; + cp++; + } else if (*cp == '0') { + guess_f = 1; + cp++; + } else { + guess_f = 1; + } continue; - case ' ': +#endif + case SP: /* module muliple options in a string are allowed for Perl moudle */ while(*cp && *cp++!='-'); continue; default: + fprintf(stderr, "unknown option: -%c\n", *(cp-1)); /* bogus option but ignored */ continue; } @@ -1964,7 +2215,7 @@ void set_iconv(nkf_char f, nkf_char (*iconv_func)(nkf_char c2,nkf_char c1,nkf_ch struct input_code *p = find_inputcode_byfunc(iconv); if (p){ set_input_codename(p->name); - debug(input_codename); + debug(p->name); } iconv_for_check = iconv; } @@ -1974,29 +2225,26 @@ void set_iconv(nkf_char f, nkf_char (*iconv_func)(nkf_char c2,nkf_char c1,nkf_ch #define SCORE_L2 (1) /* 第2水準漢字 */ #define SCORE_KANA (SCORE_L2 << 1) /* いわゆる半角カナ */ #define SCORE_DEPEND (SCORE_KANA << 1) /* 機種依存文字 */ -#ifdef SHIFTJIS_CP932 -#define SCORE_CP932 (SCORE_DEPEND << 1) /* CP932 による読み換え */ -#define SCORE_NO_EXIST (SCORE_CP932 << 1) /* 存在しない文字 */ -#else -#define SCORE_NO_EXIST (SCORE_DEPEND << 1) /* 存在しない文字 */ -#endif +#define SCORE_CP932 (SCORE_DEPEND << 1) /* CP932 による読み換え (IBM extended characters) */ +#define SCORE_X0212 (SCORE_CP932 << 1) /* JIS X 0212 */ +#define SCORE_NO_EXIST (SCORE_X0212 << 1) /* 存在しない文字 */ #define SCORE_iMIME (SCORE_NO_EXIST << 1) /* MIME による指定 */ #define SCORE_ERROR (SCORE_iMIME << 1) /* エラー */ #define SCORE_INIT (SCORE_iMIME) -const nkf_char score_table_A0[] = { +static const char score_table_A0[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_NO_EXIST, }; -const nkf_char score_table_F0[] = { +static const char score_table_F0[] = { SCORE_L2, SCORE_L2, SCORE_L2, SCORE_L2, SCORE_L2, SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST, - SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, SCORE_DEPEND, - SCORE_DEPEND, SCORE_NO_EXIST, SCORE_NO_EXIST, SCORE_ERROR, + SCORE_DEPEND, SCORE_DEPEND, SCORE_CP932, SCORE_CP932, + SCORE_CP932, SCORE_NO_EXIST, SCORE_NO_EXIST, SCORE_ERROR, }; void set_code_score(struct input_code *ptr, nkf_char score) @@ -2023,6 +2271,8 @@ void code_score(struct input_code *ptr) set_code_score(ptr, SCORE_ERROR); }else if (c2 == SSO){ set_code_score(ptr, SCORE_KANA); + }else if (c2 == 0x8f){ + set_code_score(ptr, SCORE_X0212); #ifdef UTF8_OUTPUT_ENABLE }else if (!e2w_conv(c2, c1)){ set_code_score(ptr, SCORE_NO_EXIST); @@ -2092,17 +2342,19 @@ void s_status(struct input_code *ptr, nkf_char c) status_push_ch(ptr, c); code_score(ptr); status_clear(ptr); - }else if ((0x81 <= c && c < 0xa0) || (0xe0 <= c && c <= 0xef)){ + }else if ((0x81 <= c && c < 0xa0) || (0xe0 <= c && c <= 0xea)){ ptr->stat = 1; status_push_ch(ptr, c); + }else if (0xed <= c && c <= 0xee){ + ptr->stat = 3; + status_push_ch(ptr, c); #ifdef SHIFTJIS_CP932 - }else if (cp51932_f - && is_ibmext_in_sjis(c)){ + }else if (is_ibmext_in_sjis(c)){ ptr->stat = 2; status_push_ch(ptr, c); #endif /* SHIFTJIS_CP932 */ #ifdef X0212_ENABLE - }else if (x0212_f && 0xf0 <= c && c <= 0xfc){ + }else if (0xf0 <= c && c <= 0xfc){ ptr->stat = 1; status_push_ch(ptr, c); #endif /* X0212_ENABLE */ @@ -2122,18 +2374,26 @@ void s_status(struct input_code *ptr, nkf_char c) break; case 2: #ifdef SHIFTJIS_CP932 + if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)) { + status_push_ch(ptr, c); + if (s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]) == 0) { + set_code_score(ptr, SCORE_CP932); + status_clear(ptr); + break; + } + } +#endif /* SHIFTJIS_CP932 */ + status_disable(ptr); + break; + case 3: if ((0x40 <= c && c <= 0x7e) || (0x80 <= c && c <= 0xfc)){ status_push_ch(ptr, c); - if (s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]) == 0){ - set_code_score(ptr, SCORE_CP932); - status_clear(ptr); - break; - } + s2e_conv(ptr->buf[0], ptr->buf[1], &ptr->buf[0], &ptr->buf[1]); + set_code_score(ptr, SCORE_CP932); + status_clear(ptr); + }else{ + status_disable(ptr); } -#endif /* SHIFTJIS_CP932 */ -#ifndef X0212_ENABLE - status_disable(ptr); -#endif break; } } @@ -2325,7 +2585,7 @@ nkf_char noconvert(FILE *f) void module_connection(void) { - oconv = output_conv; + oconv = output_conv; o_putc = std_putc; /* replace continucation module, from output side */ @@ -2345,8 +2605,8 @@ void module_connection(void) /* base64_count = 0; */ } - if (crmode_f) { - o_crconv = oconv; oconv = cr_conv; + if (nlmode_f || guess_f) { + o_nlconv = oconv; oconv = nl_conv; } if (rot_f) { o_rot_conv = oconv; oconv = rot_conv; @@ -2534,7 +2794,7 @@ void check_bom(FILE *f) } /* - Conversion main loop. Code detection only. + Conversion main loop. Code detection only. */ nkf_char kanji_convert(FILE *f) @@ -2573,9 +2833,9 @@ nkf_char kanji_convert(FILE *f) if (!estab_f&&!mime_decode_mode) { /* in case of not established yet */ /* It is still ambiguious */ - if (h_conv(f, c2, c1)==EOF) + if (h_conv(f, c2, c1)==EOF) LAST; - else + else c2 = 0; NEXT; } else { @@ -2591,7 +2851,7 @@ nkf_char kanji_convert(FILE *f) } else /* second byte, 7 bit code */ /* it might be kanji shitfted */ - if ((c1 == DEL) || (c1 <= SPACE)) { + if ((c1 == DEL) || (c1 <= SP)) { /* ignore bogus first code */ c2 = 0; NEXT; @@ -2665,16 +2925,16 @@ nkf_char kanji_convert(FILE *f) NEXT; } else { /* estab_f==TRUE */ if (iso8859_f) { - c2 = ISO8859_1; + c2 = ISO_8859_1; c1 &= 0x7f; SEND; } else if (SSP<=c1 && c1<0xe0 && iconv == s_iconv) { /* SJIS X0201 Case... */ - if(iso2022jp_f && x0201_f==NO_X0201) { + if (iso2022jp_f && !x0201_f) { (*oconv)(GETA1, GETA2); NEXT; } else { - c2 = X0201; + c2 = JIS_X_0201; c1 &= 0x7f; SEND; } @@ -2683,11 +2943,11 @@ nkf_char kanji_convert(FILE *f) c1 = (*i_getc)(f); /* skip SSO */ code_status(c1); if (SSP<=c1 && c1<0xe0) { - if(iso2022jp_f && x0201_f==NO_X0201) { + if (iso2022jp_f && !x0201_f) { (*oconv)(GETA1, GETA2); NEXT; } else { - c2 = X0201; + c2 = JIS_X_0201; c1 &= 0x7f; SEND; } @@ -2697,7 +2957,7 @@ nkf_char kanji_convert(FILE *f) } else if (ms_ucs_map_f == UCS_MAP_CP10001 && (c1 == 0xFD || c1 == 0xFE)) { /* CP10001 */ - c2 = X0201; + c2 = JIS_X_0201; c1 &= 0x7f; SEND; } else { @@ -2706,32 +2966,32 @@ nkf_char kanji_convert(FILE *f) NEXT; } } - } else if ((c1 > SPACE) && (c1 != DEL)) { + } else if ((c1 > SP) && (c1 != DEL)) { /* in case of Roman characters */ - if (shift_mode) { + if (shift_mode) { /* output 1 shifted byte */ if (iso8859_f) { - c2 = ISO8859_1; + c2 = ISO_8859_1; SEND; - } else if (SPACE<=c1 && c1<(0xe0&0x7f) ){ + } else if (SP <= c1 && c1 < (0xe0&0x7f)){ /* output 1 shifted byte */ - if(iso2022jp_f && x0201_f==NO_X0201) { + if (iso2022jp_f && !x0201_f) { (*oconv)(GETA1, GETA2); NEXT; } else { - c2 = X0201; + c2 = JIS_X_0201; SEND; } } else { /* look like bogus code */ NEXT; } - } else if (input_mode == X0208 || input_mode == X0212 || - input_mode == X0213_1 || input_mode == X0213_2) { + } else if (input_mode == JIS_X_0208 || input_mode == JIS_X_0212 || + input_mode == JIS_X_0213_1 || input_mode == JIS_X_0213_2) { /* in case of Kanji shifted */ c2 = c1; NEXT; - } else if (c1 == '=' && mime_f && !mime_decode_mode ) { + } else if (c1 == '=' && mime_f && !mime_decode_mode) { /* Check MIME code */ if ((c1 = (*i_getc)(f)) == EOF) { (*oconv)(0, '='); @@ -2740,11 +3000,11 @@ nkf_char kanji_convert(FILE *f) /* =? is mime conversion start sequence */ if(mime_f == STRICT_MIME) { /* check in real detail */ - if (mime_begin_strict(f) == EOF) + if (mime_begin_strict(f) == EOF) LAST; else NEXT; - } else if (mime_begin(f) == EOF) + } else if (mime_begin(f) == EOF) LAST; else NEXT; @@ -2754,37 +3014,37 @@ nkf_char kanji_convert(FILE *f) NEXT; } } else { - /* normal ASCII code */ + /* normal ASCII code */ SEND; } - } else if (c1 == SI && (!is_8bit || mime_decode_mode)) { - shift_mode = FALSE; + } else if (c1 == SI && (!is_8bit || mime_decode_mode)) { + shift_mode = FALSE; NEXT; - } else if (c1 == SO && (!is_8bit || mime_decode_mode)) { - shift_mode = TRUE; + } else if (c1 == SO && (!is_8bit || mime_decode_mode)) { + shift_mode = TRUE; NEXT; - } else if (c1 == ESC && (!is_8bit || mime_decode_mode)) { + } else if (c1 == ESC && (!is_8bit || mime_decode_mode)) { if ((c1 = (*i_getc)(f)) == EOF) { /* (*oconv)(0, ESC); don't send bogus code */ LAST; } else if (c1 == '$') { if ((c1 = (*i_getc)(f)) == EOF) { /* - (*oconv)(0, ESC); don't send bogus code + (*oconv)(0, ESC); don't send bogus code (*oconv)(0, '$'); */ LAST; } else if (c1 == '@'|| c1 == 'B') { /* This is kanji introduction */ - input_mode = X0208; + input_mode = JIS_X_0208; shift_mode = FALSE; set_input_codename("ISO-2022-JP"); #ifdef CHECK_OPTION - debug(input_codename); + debug("ISO-2022-JP"); #endif NEXT; } else if (c1 == '(') { if ((c1 = (*i_getc)(f)) == EOF) { - /* don't send bogus code + /* don't send bogus code (*oconv)(0, ESC); (*oconv)(0, '$'); (*oconv)(0, '('); @@ -2792,21 +3052,21 @@ nkf_char kanji_convert(FILE *f) LAST; } else if (c1 == '@'|| c1 == 'B') { /* This is kanji introduction */ - input_mode = X0208; + input_mode = JIS_X_0208; shift_mode = FALSE; NEXT; #ifdef X0212_ENABLE } else if (c1 == 'D'){ - input_mode = X0212; + input_mode = JIS_X_0212; shift_mode = FALSE; NEXT; #endif /* X0212_ENABLE */ - } else if (c1 == (X0213_1&0x7F)){ - input_mode = X0213_1; + } else if (c1 == 0x4F){ + input_mode = JIS_X_0213_1; shift_mode = FALSE; NEXT; - } else if (c1 == (X0213_2&0x7F)){ - input_mode = X0213_2; + } else if (c1 == 0x50){ + input_mode = JIS_X_0213_2; shift_mode = FALSE; NEXT; } else { @@ -2819,7 +3079,7 @@ nkf_char kanji_convert(FILE *f) } } else if (broken_f&0x2) { /* accept any ESC-(-x as broken code ... */ - input_mode = X0208; + input_mode = JIS_X_0208; shift_mode = FALSE; NEXT; } else { @@ -2830,14 +3090,14 @@ nkf_char kanji_convert(FILE *f) } } else if (c1 == '(') { if ((c1 = (*i_getc)(f)) == EOF) { - /* don't send bogus code + /* don't send bogus code (*oconv)(0, ESC); (*oconv)(0, '('); */ LAST; } else { if (c1 == 'I') { /* This is X0201 kana introduction */ - input_mode = X0201; shift_mode = X0201; + input_mode = JIS_X_0201; shift_mode = JIS_X_0201; NEXT; } else if (c1 == 'B' || c1 == 'J' || c1 == 'H') { /* This is X0208 kanji introduction */ @@ -2853,12 +3113,12 @@ nkf_char kanji_convert(FILE *f) SEND; } } - } else if ( c1 == 'N' || c1 == 'n' ){ + } else if ( c1 == 'N' || c1 == 'n'){ /* SS2 */ c3 = (*i_getc)(f); /* skip SS2 */ - if ( (SPACE<=c3 && c3 < 0x60) || (0xa0<=c3 && c3 < 0xe0)){ + if ( (SP<=c3 && c3 < 0x60) || (0xa0<=c3 && c3 < 0xe0)){ c1 = c3; - c2 = X0201; + c2 = JIS_X_0201; SEND; }else{ (*i_ungetc)(c3, f); @@ -2880,7 +3140,7 @@ nkf_char kanji_convert(FILE *f) /* J-PHONE emoji */ if ((c1 = (*i_getc)(f)) == EOF) { /* - (*oconv)(0, ESC); don't send bogus code + (*oconv)(0, ESC); don't send bogus code (*oconv)(0, '$'); */ LAST; } else { @@ -2893,10 +3153,10 @@ nkf_char kanji_convert(FILE *f) C%7 : 0 1 2 3 4 5 6 NUM : 2 0 3 4 5 X 1 */ - static const int jphone_emoji_first_table[7] = {2, 0, 3, 4, 5, 0, 1}; - c0 = (jphone_emoji_first_table[c1 % 7] << 8) - SPACE + 0xE000 + CLASS_UNICODE; + static const char jphone_emoji_first_table[7] = {2, 0, 3, 4, 5, 0, 1}; + c0 = (jphone_emoji_first_table[c1 % 7] << 8) - SP + 0xE000 + CLASS_UNICODE; while ((c1 = (*i_getc)(f)) != EOF) { - if (SPACE <= c1 && c1 <= 'z') { + if (SP <= c1 && c1 <= 'z') { (*oconv)(0, c1 + c0); } else break; /* c1 == SO */ } @@ -2909,32 +3169,32 @@ nkf_char kanji_convert(FILE *f) (*oconv)(0, ESC); SEND; } - } else if (c1 == NL || c1 == CR) { + } else if (c1 == LF || c1 == CR) { if (broken_f&4) { input_mode = ASCII; set_iconv(FALSE, 0); SEND; } else if (mime_decode_f && !mime_decode_mode){ - if (c1 == NL) { - if ((c1=(*i_getc)(f))!=EOF && c1 == SPACE) { - i_ungetc(SPACE,f); + if (c1 == LF) { + if ((c1=(*i_getc)(f))!=EOF && c1 == SP) { + i_ungetc(SP,f); continue; } else { i_ungetc(c1,f); } - c1 = NL; + c1 = LF; SEND; } else { /* if (c1 == CR)*/ if ((c1=(*i_getc)(f))!=EOF) { - if (c1==SPACE) { - i_ungetc(SPACE,f); + if (c1==SP) { + i_ungetc(SP,f); continue; - } else if (c1 == NL && (c1=(*i_getc)(f))!=EOF && c1 == SPACE) { - i_ungetc(SPACE,f); + } else if (c1 == LF && (c1=(*i_getc)(f))!=EOF && c1 == SP) { + i_ungetc(SP,f); continue; } else { i_ungetc(c1,f); } - i_ungetc(NL,f); + i_ungetc(LF,f); } else { i_ungetc(c1,f); } @@ -2942,15 +3202,11 @@ nkf_char kanji_convert(FILE *f) SEND; } } - if (!crmode_f) { - if (prev_cr && c1 == NL) crmode_f = CRLF; - else crmode_f = c1; - } - } else if (c1 == DEL && input_mode == X0208 ) { + } else if (c1 == DEL && input_mode == JIS_X_0208) { /* CP5022x */ c2 = c1; NEXT; - } else + } else SEND; } /* send: */ @@ -2977,8 +3233,8 @@ nkf_char kanji_convert(FILE *f) break; } break; - case X0208: - case X0213_1: + case JIS_X_0208: + case JIS_X_0213_1: if (ms_ucs_map_f && 0x7F <= c2 && c2 <= 0x92 && 0x21 <= c1 && c1 <= 0x7E) { @@ -2990,11 +3246,11 @@ nkf_char kanji_convert(FILE *f) (*oconv)(c2, c1); /* this is JIS, not SJIS/EUC case */ break; #ifdef X0212_ENABLE - case X0212: + case JIS_X_0212: (*oconv)(PREFIX_EUCG3 | c2, c1); break; #endif /* X0212_ENABLE */ - case X0213_2: + case JIS_X_0213_2: (*oconv)(PREFIX_EUCG3 | c2, c1); break; default: @@ -3009,7 +3265,7 @@ nkf_char kanji_convert(FILE *f) /* epilogue */ (*iconv)(EOF, 0, 0); - if (!is_inputcode_set) + if (!input_codename) { if (is_8bit) { struct input_code *p = input_code_list; @@ -3019,6 +3275,9 @@ nkf_char kanji_convert(FILE *f) ++p; } set_input_codename(result->name); +#ifdef CHECK_OPTION + debug(result->name); +#endif } } return 1; @@ -3087,7 +3346,7 @@ h_conv(FILE *f, nkf_char c2, nkf_char c1) (*iconv)(0, c2, 0); continue; }else if (iconv == s_iconv && 0xa1 <= c2 && c2 <= 0xdf){ - (*iconv)(X0201, c2, 0); + (*iconv)(JIS_X_0201, c2, 0); continue; } if (hold_index < hold_count){ @@ -3154,7 +3413,7 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) #if defined(SHIFTJIS_CP932) || defined(X0212_ENABLE) nkf_char val; #endif - static const nkf_char shift_jisx0213_s1a3_table[5][2] ={ { 1, 8}, { 3, 4}, { 5,12}, {13,14}, {15, 0} }; + static const char shift_jisx0213_s1a3_table[5][2] ={ { 1, 8}, { 3, 4}, { 5,12}, {13,14}, {15, 0} }; #ifdef SHIFTJIS_CP932 if (!cp932inv_f && is_ibmext_in_sjis(c2)){ val = shiftjis_cp932[c2 - CP932_TABLE_BEGIN][c1 - 0x40]; @@ -3202,7 +3461,7 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) if (0x9E < c1) c2++; } if (c1 < 0x9F) - c1 = c1 - ((c1 > DEL) ? SPACE : 0x1F); + c1 = c1 - ((c1 > DEL) ? SP : 0x1F); else { c1 = c1 - 0x7E; } @@ -3218,9 +3477,9 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) nkf_char s_iconv(nkf_char c2, nkf_char c1, nkf_char c0) { - if (c2 == X0201) { + if (c2 == JIS_X_0201) { c1 &= 0x7f; - } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) { + } else if ((c2 == EOF) || (c2 == 0) || c2 < SP) { /* NOP */ } else if (!x0213_f && 0xF0 <= c2 && c2 <= 0xF9 && 0x40 <= c1 && c1 <= 0xFC) { /* CP932 UDC */ @@ -3237,7 +3496,7 @@ nkf_char s_iconv(nkf_char c2, nkf_char c1, nkf_char c0) nkf_char e_iconv(nkf_char c2, nkf_char c1, nkf_char c0) { - if (c2 == X0201) { + if (c2 == JIS_X_0201) { c1 &= 0x7f; #ifdef X0212_ENABLE }else if (c2 == 0x8f){ @@ -3266,9 +3525,9 @@ nkf_char e_iconv(nkf_char c2, nkf_char c1, nkf_char c0) } #endif /* X0212_ENABLE */ } else if (c2 == SSO){ - c2 = X0201; + c2 = JIS_X_0201; c1 &= 0x7f; - } else if ((c2 == EOF) || (c2 == 0) || c2 < SPACE) { + } else if ((c2 == EOF) || (c2 == 0) || c2 < SP) { /* NOP */ } else { if (!cp51932_f && ms_ucs_map_f && 0xF5 <= c2 && c2 <= 0xFE && 0xA1 <= c1 && c1 <= 0xFE) { @@ -3320,13 +3579,13 @@ nkf_char w2e_conv(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char *p2, nkf_char nkf_char w_iconv(nkf_char c2, nkf_char c1, nkf_char c0) { nkf_char ret = 0; - static const int w_iconv_utf8_1st_byte[] = + static const char w_iconv_utf8_1st_byte[] = { /* 0xC0 - 0xFF */ 20, 20, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 30, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 32, 33, 33, 40, 41, 41, 41, 42, 43, 43, 43, 50, 50, 50, 50, 60, 60, 70, 70}; - + if (c2 < 0 || 0xff < c2) { }else if (c2 == 0) { /* 0 : 1 byte*/ c0 = 0; @@ -3506,22 +3765,22 @@ nkf_char unicode_to_jis_common(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char * { const unsigned short *const *pp; const unsigned short *const *const *ppp; - static const int no_best_fit_chars_table_C2[] = + static const char no_best_fit_chars_table_C2[] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 2, 1, 1, 2, 0, 0, 1, 1, 0, 1, 0, 1, 2, 1, 1, 1, 1, 1, 1, 1}; - static const int no_best_fit_chars_table_C2_ms[] = + static const char no_best_fit_chars_table_C2_ms[] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 1, 0}; - static const int no_best_fit_chars_table_932_C2[] = + static const char no_best_fit_chars_table_932_C2[] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0}; - static const int no_best_fit_chars_table_932_C3[] = + static const char no_best_fit_chars_table_932_C3[] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, @@ -3684,7 +3943,7 @@ nkf_char w_iconv_common(nkf_char c1, nkf_char c0, const unsigned short *const *p c2 &= 0x7f; c2 |= PREFIX_EUCG3; } - if (c2 == SO) c2 = X0201; + if (c2 == SO) c2 = JIS_X_0201; c1 = val & 0x7f; if (p2) *p2 = c2; if (p1) *p1 = c1; @@ -3693,13 +3952,12 @@ nkf_char w_iconv_common(nkf_char c1, nkf_char c0, const unsigned short *const *p void nkf_each_char_to_hex(void (*f)(nkf_char c2,nkf_char c1), nkf_char c) { - const char *hex = "0123456789ABCDEF"; int shift = 20; c &= VALUE_MASK; while(shift >= 0){ if(c >= 1<= 0){ - (*f)(0, hex[(c>>shift)&0xF]); + (*f)(0, bin2hex(c>>shift)); shift -= 4; } }else{ @@ -3744,22 +4002,21 @@ void encode_fallback_xml(nkf_char c) void encode_fallback_java(nkf_char c) { - const char *hex = "0123456789ABCDEF"; (*oconv)(0, '\\'); c &= VALUE_MASK; if(!is_unicode_bmp(c)){ (*oconv)(0, 'U'); (*oconv)(0, '0'); (*oconv)(0, '0'); - (*oconv)(0, hex[(c>>20)&0xF]); - (*oconv)(0, hex[(c>>16)&0xF]); + (*oconv)(0, bin2hex(c>>20)); + (*oconv)(0, bin2hex(c>>16)); }else{ (*oconv)(0, 'u'); } - (*oconv)(0, hex[(c>>12)&0xF]); - (*oconv)(0, hex[(c>> 8)&0xF]); - (*oconv)(0, hex[(c>> 4)&0xF]); - (*oconv)(0, hex[ c &0xF]); + (*oconv)(0, bin2hex(c>>12)); + (*oconv)(0, bin2hex(c>> 8)); + (*oconv)(0, bin2hex(c>> 4)); + (*oconv)(0, bin2hex(c )); return; } @@ -3786,7 +4043,7 @@ nkf_char e2w_conv(nkf_char c2, nkf_char c1) { const unsigned short *p; - if (c2 == X0201) { + if (c2 == JIS_X_0201) { if (ms_ucs_map_f == UCS_MAP_CP10001) { switch (c1) { case 0x20: @@ -3864,14 +4121,14 @@ void w_oconv(nkf_char c2, nkf_char c1) } #endif - if (c2 == 0) { + if (c2 == 0) { output_mode = ASCII; (*o_putc)(c1); - } else if (c2 == ISO8859_1) { - output_mode = ISO8859_1; + } else if (c2 == ISO_8859_1) { + output_mode = UTF_8; (*o_putc)(c1 | 0x080); } else { - output_mode = UTF8; + output_mode = UTF_8; val = e2w_conv(c2, c1); if (val){ w16w_conv(val, &c2, &c1, &c0); @@ -3902,7 +4159,7 @@ void w_oconv16(nkf_char c2, nkf_char c1) return; } - if (c2 == ISO8859_1) { + if (c2 == ISO_8859_1) { c2 = 0; c1 |= 0x80; #ifdef NUMCHAR_OPTION @@ -3967,7 +4224,7 @@ void w_oconv32(nkf_char c2, nkf_char c1) return; } - if (c2 == ISO8859_1) { + if (c2 == ISO_8859_1) { c1 |= 0x80; #ifdef NUMCHAR_OPTION } else if (c2 == 0 && is_unicode_capsule(c1)) { @@ -4023,18 +4280,18 @@ void e_oconv(nkf_char c2, nkf_char c1) if (c2 == EOF) { (*o_putc)(EOF); return; - } else if (c2 == 0) { + } else if (c2 == 0) { output_mode = ASCII; (*o_putc)(c1); - } else if (c2 == X0201) { - output_mode = JAPANESE_EUC; + } else if (c2 == JIS_X_0201) { + output_mode = EUC_JP; (*o_putc)(SSO); (*o_putc)(c1|0x80); - } else if (c2 == ISO8859_1) { - output_mode = ISO8859_1; + } else if (c2 == ISO_8859_1) { + output_mode = ISO_8859_1; (*o_putc)(c1 | 0x080); #ifdef X0212_ENABLE } else if (is_eucg3(c2)){ - output_mode = JAPANESE_EUC; + output_mode = EUC_JP; #ifdef SHIFTJIS_CP932 if (!cp932inv_f){ nkf_char s2, s1; @@ -4062,7 +4319,7 @@ void e_oconv(nkf_char c2, nkf_char c1) set_iconv(FALSE, 0); return; /* too late to rescue this char */ } - output_mode = JAPANESE_EUC; + output_mode = EUC_JP; (*o_putc)(c2 | 0x080); (*o_putc)(c1 | 0x080); } @@ -4169,11 +4426,11 @@ void s_oconv(nkf_char c2, nkf_char c1) } else if (c2 == 0) { output_mode = ASCII; (*o_putc)(c1); - } else if (c2 == X0201) { + } else if (c2 == JIS_X_0201) { output_mode = SHIFT_JIS; (*o_putc)(c1|0x80); - } else if (c2 == ISO8859_1) { - output_mode = ISO8859_1; + } else if (c2 == ISO_8859_1) { + output_mode = ISO_8859_1; (*o_putc)(c1 | 0x080); #ifdef X0212_ENABLE } else if (is_eucg3(c2)){ @@ -4230,7 +4487,7 @@ void j_oconv(nkf_char c2, nkf_char c1) } #endif if (c2 == EOF) { - if (output_mode !=ASCII && output_mode!=ISO8859_1) { + if (output_mode !=ASCII && output_mode!=ISO_8859_1) { (*o_putc)(ESC); (*o_putc)('('); (*o_putc)(ascii_intro); @@ -4240,41 +4497,41 @@ void j_oconv(nkf_char c2, nkf_char c1) #ifdef X0212_ENABLE } else if (is_eucg3(c2)){ if(x0213_f){ - if(output_mode!=X0213_2){ - output_mode = X0213_2; + if(output_mode!=JIS_X_0213_2){ + output_mode = JIS_X_0213_2; (*o_putc)(ESC); (*o_putc)('$'); (*o_putc)('('); - (*o_putc)(X0213_2&0x7F); + (*o_putc)(0x50); } }else{ - if(output_mode!=X0212){ - output_mode = X0212; + if(output_mode!=JIS_X_0212){ + output_mode = JIS_X_0212; (*o_putc)(ESC); (*o_putc)('$'); (*o_putc)('('); - (*o_putc)(X0212&0x7F); + (*o_putc)(0x44); } } (*o_putc)(c2 & 0x7f); (*o_putc)(c1); #endif - } else if (c2==X0201) { - if (output_mode!=X0201) { - output_mode = X0201; + } else if (c2==JIS_X_0201) { + if (output_mode!=JIS_X_0201) { + output_mode = JIS_X_0201; (*o_putc)(ESC); (*o_putc)('('); (*o_putc)('I'); } (*o_putc)(c1); - } else if (c2==ISO8859_1) { + } else if (c2==ISO_8859_1) { /* iso8859 introduction, or 8th bit on */ - /* Can we convert in 7bit form using ESC-'-'-A ? + /* Can we convert in 7bit form using ESC-'-'-A ? Is this popular? */ - output_mode = ISO8859_1; + output_mode = ISO_8859_1; (*o_putc)(c1|0x80); } else if (c2 == 0) { - if (output_mode !=ASCII && output_mode!=ISO8859_1) { + if (output_mode !=ASCII && output_mode!=ISO_8859_1) { (*o_putc)(ESC); (*o_putc)('('); (*o_putc)(ascii_intro); @@ -4286,15 +4543,15 @@ void j_oconv(nkf_char c2, nkf_char c1) ? c2<0x20 || 0x920x80 Japanese (X0208/X0201) <0x80 ASCII - \n new line - ' ' space + LF new line + SP space This fold algorthm does not preserve heading space in a line. This is the main difference from fmt. @@ -4407,34 +4659,34 @@ void cr_conv(nkf_char c2, nkf_char c1) #define char_size(c2,c1) (c2?2:1) void fold_conv(nkf_char c2, nkf_char c1) -{ +{ nkf_char prev0; nkf_char fold_state; - if (c1== '\r' && !fold_preserve_f) { + if (c1== CR && !fold_preserve_f) { fold_state=0; /* ignore cr */ - }else if (c1== '\n'&&f_prev=='\r' && fold_preserve_f) { - f_prev = '\n'; + }else if (c1== LF&&f_prev==CR && fold_preserve_f) { + f_prev = LF; fold_state=0; /* ignore cr */ } else if (c1== BS) { if (f_line>0) f_line--; fold_state = 1; } else if (c2==EOF && f_line != 0) { /* close open last line */ - fold_state = '\n'; - } else if ((c1=='\n' && !fold_preserve_f) - || ((c1=='\r'||(c1=='\n'&&f_prev!='\r')) + fold_state = LF; + } else if ((c1==LF && !fold_preserve_f) + || ((c1==CR||(c1==LF&&f_prev!=CR)) && fold_preserve_f)) { /* new line */ - if (fold_preserve_f) { + if (fold_preserve_f) { f_prev = c1; f_line = 0; - fold_state = '\r'; + fold_state = CR; } else if ((f_prev == c1 && !fold_preserve_f) - || (f_prev == '\n' && fold_preserve_f) + || (f_prev == LF && fold_preserve_f) ) { /* duplicate newline */ if (f_line) { f_line = 0; - fold_state = '\n'; /* output two newline */ + fold_state = LF; /* output two newline */ } else { f_line = 0; fold_state = 1; @@ -4443,41 +4695,41 @@ void fold_conv(nkf_char c2, nkf_char c1) if (f_prev&0x80) { /* Japanese? */ f_prev = c1; fold_state = 0; /* ignore given single newline */ - } else if (f_prev==' ') { + } else if (f_prev==SP) { fold_state = 0; } else { f_prev = c1; - if (++f_line<=fold_len) - fold_state = ' '; + if (++f_line<=fold_len) + fold_state = SP; else { f_line = 0; - fold_state = '\r'; /* fold and output nothing */ + fold_state = CR; /* fold and output nothing */ } } } } else if (c1=='\f') { - f_prev = '\n'; + f_prev = LF; f_line = 0; - fold_state = '\n'; /* output newline and clear */ - } else if ( (c2==0 && c1==' ')|| - (c2==0 && c1=='\t')|| + fold_state = LF; /* output newline and clear */ + } else if ( (c2==0 && c1==SP)|| + (c2==0 && c1==TAB)|| (c2=='!'&& c1=='!')) { /* X0208 kankaku or ascii space */ - if (f_prev == ' ') { + if (f_prev == SP) { fold_state = 0; /* remove duplicate spaces */ } else { - f_prev = ' '; - if (++f_line<=fold_len) - fold_state = ' '; /* output ASCII space only */ + f_prev = SP; + if (++f_line<=fold_len) + fold_state = SP; /* output ASCII space only */ else { - f_prev = ' '; f_line = 0; - fold_state = '\r'; /* fold and output nothing */ + f_prev = SP; f_line = 0; + fold_state = CR; /* fold and output nothing */ } } } else { prev0 = f_prev; /* we still need this one... , but almost done */ f_prev = c1; - if (c2 || c2==X0201) + if (c2 || c2==JIS_X_0201) f_prev |= 0x80; /* this is Japanese */ f_line += char_size(c2,c1); if (f_line<=fold_len) { /* normal case */ @@ -4485,8 +4737,8 @@ void fold_conv(nkf_char c2, nkf_char c1) } else { if (f_line>fold_len+fold_margin) { /* too many kinsoku suspension */ f_line = char_size(c2,c1); - fold_state = '\n'; /* We can't wait, do fold now */ - } else if (c2==X0201) { + fold_state = LF; /* We can't wait, do fold now */ + } else if (c2==JIS_X_0201) { /* simple kinsoku rules return 1 means no folding */ if (c1==(0xde&0x7f)) fold_state = 1; /* ゛*/ else if (c1==(0xdf&0x7f)) fold_state = 1; /* ゜*/ @@ -4494,15 +4746,15 @@ void fold_conv(nkf_char c2, nkf_char c1) else if (c1==(0xa3&0x7f)) fold_state = 1; /* ，*/ else if (c1==(0xa1&0x7f)) fold_state = 1; /* 」*/ else if (c1==(0xb0&0x7f)) fold_state = 1; /* - */ - else if (SPACE<=c1 && c1<=(0xdf&0x7f)) { /* X0201 */ + else if (SP<=c1 && c1<=(0xdf&0x7f)) { /* X0201 */ f_line = 1; - fold_state = '\n';/* add one new f_line before this character */ + fold_state = LF;/* add one new f_line before this character */ } else { f_line = 1; - fold_state = '\n';/* add one new f_line before this character */ + fold_state = LF;/* add one new f_line before this character */ } } else if (c2==0) { - /* kinsoku point in ASCII */ + /* kinsoku point in ASCII */ if ( c1==')'|| /* { [ ( */ c1==']'|| c1=='}'|| @@ -4512,17 +4764,17 @@ void fold_conv(nkf_char c2, nkf_char c1) c1=='?'|| c1=='/'|| c1==':'|| - c1==';' ) { + c1==';') { fold_state = 1; /* just after special */ } else if (!is_alnum(prev0)) { f_line = char_size(c2,c1); - fold_state = '\n'; - } else if ((prev0==' ') || /* ignored new f_line */ - (prev0=='\n')|| /* ignored new f_line */ + fold_state = LF; + } else if ((prev0==SP) || /* ignored new f_line */ + (prev0==LF)|| /* ignored new f_line */ (prev0&0x80)) { /* X0208 - ASCII */ f_line = char_size(c2,c1); - fold_state = '\n';/* add one new f_line before this character */ + fold_state = LF;/* add one new f_line before this character */ } else { fold_state = 1; /* default no fold in ASCII */ } @@ -4541,14 +4793,14 @@ void fold_conv(nkf_char c2, nkf_char c1) else if (c1=='+') fold_state = 1; /* ゛ */ else if (c1==',') fold_state = 1; /* ゜ */ /* default no fold in kinsoku */ - else { - fold_state = '\n'; + else { + fold_state = LF; f_line = char_size(c2,c1); /* add one new f_line before this character */ } } else { f_line = char_size(c2,c1); - fold_state = '\n'; + fold_state = LF; /* add one new f_line before this character */ } } @@ -4556,18 +4808,18 @@ void fold_conv(nkf_char c2, nkf_char c1) } /* terminator process */ switch(fold_state) { - case '\n': - (*o_fconv)(0,'\n'); + case LF: + OCONV_NEWLINE((*o_fconv)); (*o_fconv)(c2,c1); break; - case 0: + case 0: return; - case '\r': - (*o_fconv)(0,'\n'); + case CR: + OCONV_NEWLINE((*o_fconv)); break; - case '\t': - case ' ': - (*o_fconv)(0,' '); + case TAB: + case SP: + (*o_fconv)(0,SP); break; default: (*o_fconv)(c2,c1); @@ -4581,105 +4833,155 @@ void z_conv(nkf_char c2, nkf_char c1) /* if (c2) c1 &= 0x7f; assertion */ - if (c2 == X0201 && (c1 == 0x20 || c1 == 0x7D || c1 == 0x7E)) { + if (c2 == JIS_X_0201 && (c1 == 0x20 || c1 == 0x7D || c1 == 0x7E)) { (*o_zconv)(c2,c1); return; } - if (x0201_f && z_prev2==X0201) { /* X0201 */ - if (c1==(0xde&0x7f)) { /* 濁点 */ - z_prev2=0; - (*o_zconv)(dv[(z_prev1-SPACE)*2],dv[(z_prev1-SPACE)*2+1]); - return; - } else if (c1==(0xdf&0x7f)&&ev[(z_prev1-SPACE)*2]) { /* 半濁点 */ - z_prev2=0; - (*o_zconv)(ev[(z_prev1-SPACE)*2],ev[(z_prev1-SPACE)*2+1]); - return; - } else { - z_prev2=0; - (*o_zconv)(cv[(z_prev1-SPACE)*2],cv[(z_prev1-SPACE)*2+1]); - } + if (x0201_f) { + if (z_prev2 == JIS_X_0201) { + if (c2 == JIS_X_0201) { + if (c1 == (0xde&0x7f)) { /* 濁点 */ + z_prev2 = 0; + (*o_zconv)(dv[(z_prev1-SP)*2], dv[(z_prev1-SP)*2+1]); + return; + } else if (c1 == (0xdf&0x7f) && ev[(z_prev1-SP)*2]) { /* 半濁点 */ + z_prev2 = 0; + (*o_zconv)(ev[(z_prev1-SP)*2], ev[(z_prev1-SP)*2+1]); + return; + } + } + z_prev2 = 0; + (*o_zconv)(cv[(z_prev1-SP)*2], cv[(z_prev1-SP)*2+1]); + } + if (c2 == JIS_X_0201) { + if (dv[(c1-SP)*2] || ev[(c1-SP)*2]) { + /* wait for 濁点 or 半濁点 */ + z_prev1 = c1; + z_prev2 = c2; + return; + } else { + (*o_zconv)(cv[(c1-SP)*2], cv[(c1-SP)*2+1]); + return; + } + } } - if (c2==EOF) { - (*o_zconv)(c2,c1); + if (c2 == EOF) { + (*o_zconv)(c2, c1); return; } - if (x0201_f && c2==X0201) { - if (dv[(c1-SPACE)*2]||ev[(c1-SPACE)*2]) { - /* wait for 濁点 or 半濁点 */ - z_prev1 = c1; z_prev2 = c2; - return; - } else { - (*o_zconv)(cv[(c1-SPACE)*2],cv[(c1-SPACE)*2+1]); - return; - } - } - - if (alpha_f && c2 == 0x23 ) { + if (alpha_f&1 && c2 == 0x23) { /* JISX0208 Alphabet */ c2 = 0; - } else if (alpha_f && c2 == 0x21 ) { + } else if (c2 == 0x21) { /* JISX0208 Kigou */ if (0x21==c1) { - if (alpha_f&0x2) { - c1 = ' '; + if (alpha_f&2) { c2 = 0; - } else if (alpha_f&0x4) { - (*o_zconv)(0,' '); - (*o_zconv)(0,' '); + c1 = SP; + } else if (alpha_f&4) { + (*o_zconv)(0, SP); + (*o_zconv)(0, SP); return; - } - } else if (0x20': entity = ">"; break; - case '<': entity = "<"; break; - case '\"': entity = """; break; - case '&': entity = "&"; break; - } - if (entity){ - while (*entity) (*o_zconv)(0, *entity++); - return; - } } - } - } else if (alpha_f & 0x10 && c2 == 0x25) { - /* JISX0208 Katakana */ - static const int fullwidth_to_halfwidth[] = - { - 0x0000, 0x2700, 0x3100, 0x2800, 0x3200, 0x2900, 0x3300, 0x2A00, - 0x3400, 0x2B00, 0x3500, 0x3600, 0x365E, 0x3700, 0x375E, 0x3800, - 0x385E, 0x3900, 0x395E, 0x3A00, 0x3A5E, 0x3B00, 0x3B5E, 0x3C00, - 0x3C5E, 0x3D00, 0x3D5E, 0x3E00, 0x3E5E, 0x3F00, 0x3F5E, 0x4000, - 0x405E, 0x4100, 0x415E, 0x2F00, 0x4200, 0x425E, 0x4300, 0x435E, - 0x4400, 0x445E, 0x4500, 0x4600, 0x4700, 0x4800, 0x4900, 0x4A00, - 0x4A5E, 0x4A5F, 0x4B00, 0x4B5E, 0x4B5F, 0x4C00, 0x4C5E, 0x4C5F, - 0x4D00, 0x4D5E, 0x4D5F, 0x4E00, 0x4E5E, 0x4E5F, 0x4F00, 0x5000, - 0x5100, 0x5200, 0x5300, 0x2C00, 0x5400, 0x2D00, 0x5500, 0x2E00, - 0x5600, 0x5700, 0x5800, 0x5900, 0x5A00, 0x5B00, 0x0000, 0x5C00, - 0x0000, 0x0000, 0x2600, 0x5D00, 0x0000, 0x0000, 0x0000, 0x0000, - 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000 - }; - if (fullwidth_to_halfwidth[c1-0x20]){ - c2 = fullwidth_to_halfwidth[c1-0x20]; - (*o_zconv)(X0201, c2>>8); - if (c2 & 0xFF) { - (*o_zconv)(X0201, c2&0xFF); - } + } else if (alpha_f&1 && 0x20': entity = ">"; break; + case '<': entity = "<"; break; + case '\"': entity = """; break; + case '&': entity = "&"; break; + } + if (entity){ + while (*entity) (*o_zconv)(0, *entity++); return; } } + + if (alpha_f & 16) { + /* JIS X 0208 Katakana to JIS X 0201 Katakana */ + if (c2 == 0x21) { + char c = 0; + switch (c1) { + case 0x23: + /* U+3002 (0x8142) Ideographic Full Stop -> U+FF61 (0xA1) Halfwidth Ideographic Full Stop */ + c = 0xA1; + break; + case 0x56: + /* U+300C (0x8175) Left Corner Bracket -> U+FF62 (0xA2) Halfwidth Left Corner Bracket */ + c = 0xA2; + break; + case 0x57: + /* U+300D (0x8176) Right Corner Bracket -> U+FF63 (0xA3) Halfwidth Right Corner Bracket */ + c = 0xA3; + break; + case 0x22: + /* U+3001 (0x8141) Ideographic Comma -> U+FF64 (0xA4) Halfwidth Ideographic Comma */ + c = 0xA4; + break; + case 0x26: + /* U+30FB (0x8145) Katakana Middle Dot -> U+FF65 (0xA5) Halfwidth Katakana Middle Dot */ + c = 0xA5; + break; + case 0x3C: + /* U+30FC (0x815B) Katakana-Hiragana Prolonged Sound Mark -> U+FF70 (0xB0) Halfwidth Katakana-Hiragana Prolonged Sound Mark */ + c = 0xB0; + break; + case 0x2B: + /* U+309B (0x814A) Katakana-Hiragana Voiced Sound Mark -> U+FF9E (0xDE) Halfwidth Katakana Voiced Sound Mark */ + c = 0xDE; + break; + case 0x2C: + /* U+309C (0x814B) Katakana-Hiragana Semi-Voiced Sound Mark -> U+FF9F (0xDF) Halfwidth Katakana Semi-Voiced Sound Mark */ + c = 0xDF; + break; + } + if (c) { + (*o_zconv)(JIS_X_0201, c); + return; + } + } else if (c2 == 0x25) { + /* JISX0208 Katakana */ + static const int fullwidth_to_halfwidth[] = + { + 0x0000, 0x2700, 0x3100, 0x2800, 0x3200, 0x2900, 0x3300, 0x2A00, + 0x3400, 0x2B00, 0x3500, 0x3600, 0x365E, 0x3700, 0x375E, 0x3800, + 0x385E, 0x3900, 0x395E, 0x3A00, 0x3A5E, 0x3B00, 0x3B5E, 0x3C00, + 0x3C5E, 0x3D00, 0x3D5E, 0x3E00, 0x3E5E, 0x3F00, 0x3F5E, 0x4000, + 0x405E, 0x4100, 0x415E, 0x2F00, 0x4200, 0x425E, 0x4300, 0x435E, + 0x4400, 0x445E, 0x4500, 0x4600, 0x4700, 0x4800, 0x4900, 0x4A00, + 0x4A5E, 0x4A5F, 0x4B00, 0x4B5E, 0x4B5F, 0x4C00, 0x4C5E, 0x4C5F, + 0x4D00, 0x4D5E, 0x4D5F, 0x4E00, 0x4E5E, 0x4E5F, 0x4F00, 0x5000, + 0x5100, 0x5200, 0x5300, 0x2C00, 0x5400, 0x2D00, 0x5500, 0x2E00, + 0x5600, 0x5700, 0x5800, 0x5900, 0x5A00, 0x5B00, 0x0000, 0x5C00, + 0x0000, 0x0000, 0x2600, 0x5D00, 0x335E, 0x0000, 0x0000, 0x0000, + 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000 + }; + if (fullwidth_to_halfwidth[c1-0x20]){ + c2 = fullwidth_to_halfwidth[c1-0x20]; + (*o_zconv)(JIS_X_0201, c2>>8); + if (c2 & 0xFF) { + (*o_zconv)(JIS_X_0201, c2&0xFF); + } + return; + } + } + } (*o_zconv)(c2,c1); } #define rot13(c) ( \ - ( c < 'A' ) ? c: \ + ( c < 'A') ? c: \ (c <= 'M') ? (c + 13): \ (c <= 'Z') ? (c - 13): \ (c < 'a') ? (c): \ @@ -4689,15 +4991,15 @@ void z_conv(nkf_char c2, nkf_char c1) ) #define rot47(c) ( \ - ( c < '!' ) ? c: \ - ( c <= 'O' ) ? (c + 47) : \ - ( c <= '~' ) ? (c - 47) : \ + ( c < '!') ? c: \ + ( c <= 'O') ? (c + 47) : \ + ( c <= '~') ? (c - 47) : \ c \ ) void rot_conv(nkf_char c2, nkf_char c1) { - if (c2==0 || c2==X0201 || c2==ISO8859_1) { + if (c2==0 || c2==JIS_X_0201 || c2==ISO_8859_1) { c1 = rot13(c1); } else if (c2) { c1 = rot47(c1); @@ -4789,7 +5091,7 @@ void iso2022jp_check_conv(nkf_char c2, nkf_char c1) /* This converts =?ISO-2022-JP?B?HOGE HOGE?= */ -const unsigned char *mime_pattern[] = { +static const unsigned char *mime_pattern[] = { (const unsigned char *)"\075?EUC-JP?B?", (const unsigned char *)"\075?SHIFT_JIS?B?", (const unsigned char *)"\075?ISO-8859-1?Q?", @@ -4814,16 +5116,16 @@ nkf_char (*mime_priority_func[])(nkf_char c2, nkf_char c1, nkf_char c0) = { 0, }; -const nkf_char mime_encode[] = { - JAPANESE_EUC, SHIFT_JIS,ISO8859_1, ISO8859_1, X0208, X0201, +static const nkf_char mime_encode[] = { + EUC_JP, SHIFT_JIS, ISO_8859_1, ISO_8859_1, JIS_X_0208, JIS_X_0201, #if defined(UTF8_INPUT_ENABLE) - UTF8, UTF8, + UTF_8, UTF_8, #endif ASCII, 0 }; -const nkf_char mime_encode_method[] = { +static const nkf_char mime_encode_method[] = { 'B', 'B','Q', 'B', 'B', 'Q', #if defined(UTF8_INPUT_ENABLE) 'B', 'Q', @@ -4872,8 +5174,8 @@ nkf_char mime_begin_strict(FILE *f) p = mime_pattern[j]; r[0]='='; r[1]='?'; - for(i=2;p[i]>' ';i++) { /* start at =? */ - if ( ((r[i] = c1 = (*i_getc)(f))==EOF) || nkf_toupper(c1) != p[i] ) { + for(i=2;p[i]>SP;i++) { /* start at =? */ + if (((r[i] = c1 = (*i_getc)(f))==EOF) || nkf_toupper(c1) != p[i]) { /* pattern fails, try next one */ q = p; while (mime_pattern[++j]) { @@ -4903,7 +5205,7 @@ nkf_char mime_begin_strict(FILE *f) if (!unbuf_f) { /* do MIME integrity check */ return mime_integrity(f,mime_pattern[j]); - } + } } switch_mime_getc(); mimebuf_f = TRUE; @@ -4922,7 +5224,7 @@ nkf_char mime_ungetc_buf(nkf_char c, FILE *f) { if (mimebuf_f) (*i_mungetc_buf)(c,f); - else + else Fifo(--mime_input) = (unsigned char)c; return c; } @@ -4941,8 +5243,8 @@ nkf_char mime_begin(FILE *f) for(i=2;iscore & (SCORE_DEPEND|SCORE_CP932)) + input_codename = "CP932"; + } else if (strcmp(input_codename, "EUC-JP") == 0) { + if (p->score & (SCORE_X0212)) + input_codename = "EUCJP-MS"; + else if (p->score & (SCORE_DEPEND|SCORE_CP932)) + input_codename = "CP51932"; + } else if (strcmp(input_codename, "ISO-2022-JP") == 0) { + if (p->score & (SCORE_KANA)) + input_codename = "CP50221"; + else if (p->score & (SCORE_DEPEND|SCORE_CP932)) + input_codename = "CP50220"; + } + } + return input_codename; } #if !defined(PERL_XS) && !defined(WIN32DLL) void print_guessed_code(char *filename) { - char *codename = "BINARY"; - char *str_crmode = NULL; - if (!is_inputcode_mixed) { - if (strcmp(input_codename, "") == 0) { - codename = "ASCII"; - } else { - codename = input_codename; - } - if (crmode_f == CR) str_crmode = "CR"; - else if (crmode_f == NL) str_crmode = "LF"; - else if (crmode_f == CRLF) str_crmode = "CRLF"; + if (filename != NULL) printf("%s: ", filename); + if (input_codename && !*input_codename) { + printf("BINARY\n"); + } else { + input_codename = get_guessed_code(); + if (guess_f == 1) { + printf("%s\n", input_codename); + } else { + printf("%s%s\n", + input_codename, + input_newline == CR ? " (CR)" : + input_newline == LF ? " (LF)" : + input_newline == CRLF ? " (CRLF)" : + input_newline == EOF ? " (MIXED NL)" : + ""); + } } - if (filename != NULL) printf("%s:", filename); - if (str_crmode != NULL) printf("%s (%s)\n", codename, str_crmode); - else printf("%s\n", codename); } #endif /*WIN32DLL*/ -#ifdef INPUT_OPTION +#ifdef INPUT_OPTION nkf_char hex_getc(nkf_char ch, FILE *f, nkf_char (*g)(FILE *f), nkf_char (*u)(nkf_char c, FILE *f)) { @@ -5145,7 +5471,7 @@ nkf_char nfc_getc(FILE *f) int i=0, j, k=1, lower, upper; nkf_char buf[9]; const nkf_nfchar *array; - + buf[i] = (*g)(f); while (k > 0 && ((buf[i] & 0xc0) != 0x80)){ lower=0, upper=NORMALIZATION_TABLE_LENGTH-1; @@ -5181,7 +5507,7 @@ nkf_char nfc_ungetc(nkf_char c, FILE *f) #endif /* UNICODE_NORMALIZATION */ -nkf_char +nkf_char mime_getc(FILE *f) { nkf_char c1, c2, c3, c4, cc; @@ -5207,15 +5533,15 @@ mime_getc(FILE *f) if (mime_decode_mode == 'Q') { if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF); restart_mime_q: - if (c1=='_' && mimebuf_f != FIXED_MIME) return ' '; - if (c1<=' ' || DEL<=c1) { + if (c1=='_' && mimebuf_f != FIXED_MIME) return SP; + if (c1<=SP || DEL<=c1) { mime_decode_mode = exit_mode; /* prepare for quit */ return c1; } if (c1!='=' && (c1!='?' || mimebuf_f == FIXED_MIME)) { return c1; } - + mime_decode_mode = exit_mode; /* prepare for quit */ if ((c2 = (*i_mgetc)(f)) == EOF) return (EOF); if (c1=='?'&&c2=='=' && mimebuf_f != FIXED_MIME) { @@ -5229,32 +5555,32 @@ restart_mime_q: } while ((c1=(*i_getc)(f))!=EOF) { switch (c1) { - case NL: + case LF: case CR: - if (c1==NL) { - if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) { - i_ungetc(SPACE,f); + if (c1==LF) { + if ((c1=(*i_getc)(f))!=EOF && (c1==SP||c1==TAB)) { + i_ungetc(SP,f); continue; } else { i_ungetc(c1,f); } - c1 = NL; + c1 = LF; } else { - if ((c1=(*i_getc)(f))!=EOF && c1 == NL) { - if ((c1=(*i_getc)(f))!=EOF && (c1==SPACE||c1==TAB)) { - i_ungetc(SPACE,f); + if ((c1=(*i_getc)(f))!=EOF && c1 == LF) { + if ((c1=(*i_getc)(f))!=EOF && (c1==SP||c1==TAB)) { + i_ungetc(SP,f); continue; } else { i_ungetc(c1,f); } - i_ungetc(NL,f); + i_ungetc(LF,f); } else { i_ungetc(c1,f); } c1 = CR; } break; - case SPACE: + case SP: case TAB: lwsp_buf[lwsp_count] = (unsigned char)c1; if (lwsp_count++>lwsp_size){ @@ -5271,7 +5597,7 @@ restart_mime_q: } break; } - if (lwsp_count > 0 && (c1 != '=' || (lwsp_buf[lwsp_count-1] != SPACE && lwsp_buf[lwsp_count-1] != TAB))) { + if (lwsp_count > 0 && (c1 != '=' || (lwsp_buf[lwsp_count-1] != SP && lwsp_buf[lwsp_count-1] != TAB))) { i_ungetc(c1,f); for(lwsp_count--;lwsp_count>0;lwsp_count--) i_ungetc(lwsp_buf[lwsp_count],f); @@ -5280,8 +5606,8 @@ restart_mime_q: free(lwsp_buf); return c1; } - if (c1=='='&&c2<' ') { /* this is soft wrap */ - while((c1 = (*i_mgetc)(f)) <=' ') { + if (c1=='='&&c2 0 && (c1 != '=' || (lwsp_buf[lwsp_count-1] != SP && lwsp_buf[lwsp_count-1] != TAB))) { i_ungetc(c1,f); for(lwsp_count--;lwsp_count>0;lwsp_count--) i_ungetc(lwsp_buf[lwsp_count],f); @@ -5392,19 +5718,19 @@ mime_c2_retry: return c1; } mime_c3_retry: - if ((c3 = (*i_mgetc)(f))<=' ') { + if ((c3 = (*i_mgetc)(f))<=SP) { if (c3==EOF) return (EOF); if (mime_f != STRICT_MIME) goto mime_c3_retry; - if (mimebuf_f!=FIXED_MIME) input_mode = ASCII; + if (mimebuf_f!=FIXED_MIME) input_mode = ASCII; return c3; } mime_c4_retry: - if ((c4 = (*i_mgetc)(f))<=' ') { + if ((c4 = (*i_mgetc)(f))<=SP) { if (c4==EOF) return (EOF); if (mime_f != STRICT_MIME) goto mime_c4_retry; - if (mimebuf_f!=FIXED_MIME) input_mode = ASCII; + if (mimebuf_f!=FIXED_MIME) input_mode = ASCII; return c4; } @@ -5423,7 +5749,7 @@ mime_c4_retry: if (c3 != '=') { Fifo(mime_last++) = (unsigned char)cc; cc = ((t3 << 6) & 0x0c0) | (t4 & 0x3f); - if (c4 != '=') + if (c4 != '=') Fifo(mime_last++) = (unsigned char)cc; } } else { @@ -5446,7 +5772,7 @@ nkf_char mime_integrity(FILE *f, const unsigned char *p) */ mime_input = mime_top; mime_last = mime_top; - + while(*p) Fifo(mime_input++) = *p++; d = 0; q = mime_input; @@ -5458,7 +5784,7 @@ nkf_char mime_integrity(FILE *f, const unsigned char *p) /* checked. skip header, start decode */ Fifo(mime_input++) = (unsigned char)c; /* mime_last_input = mime_input; */ - mime_input = q; + mime_input = q; switch_mime_getc(); return 1; } @@ -5482,13 +5808,15 @@ nkf_char base64decode(nkf_char c) if (c > '@') { if (c < '[') { i = c - 'A'; /* A..Z 0-25 */ + } else if (c == '_') { + i = '?' /* 63 */ ; /* _ 63 */ } 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 if (c == '+' || c == '-') { + i = '>' /* 62 */ ; /* + and - 62 */ } else { i = '?' /* 63 */ ; /* / 63 */ } @@ -5502,8 +5830,6 @@ static nkf_char b64c; #define MIMEOUT_BUF_LENGTH (60) char mimeout_buf[MIMEOUT_BUF_LENGTH+1]; int mimeout_buf_count = 0; -int mimeout_preserve_space = 0; -#define itoh4(c) (c>=10?c+'A'-10:c+'0') void open_mime(nkf_char mode) { @@ -5518,35 +5844,30 @@ void open_mime(nkf_char mode) } } mimeout_mode = mime_encode_method[i]; - i = 0; if (base64_count>45) { if (mimeout_buf_count>0 && nkf_isblank(mimeout_buf[i])){ (*o_mputc)(mimeout_buf[i]); i++; } - (*o_mputc)(NL); - (*o_mputc)(SPACE); + PUT_NEWLINE((*o_mputc)); + (*o_mputc)(SP); base64_count = 1; - if (!mimeout_preserve_space && mimeout_buf_count>0 - && (mimeout_buf[i]==SPACE || mimeout_buf[i]==TAB - || mimeout_buf[i]==CR || mimeout_buf[i]==NL )) { + if (mimeout_buf_count>0 + && (mimeout_buf[i]==SP || mimeout_buf[i]==TAB + || mimeout_buf[i]==CR || mimeout_buf[i]==LF)) { i++; } } - if (!mimeout_preserve_space) { - for (;i 0) { if (mimeout_f!=FIXED_MIME) { - close_mime(); + close_mime(); } else if (mimeout_mode != 'Q') mimeout_mode = 'B'; } @@ -5596,13 +5917,13 @@ void mimeout_addchar(nkf_char c) { switch(mimeout_mode) { case 'Q': - if (c==CR||c==NL) { + if (c==CR||c==LF) { (*o_mputc)(c); base64_count = 0; } else if(!nkf_isalnum(c)) { (*o_mputc)('='); - (*o_mputc)(itoh4(((c>>4)&0xf))); - (*o_mputc)(itoh4((c&0xf))); + (*o_mputc)(bin2hex(((c>>4)&0xf))); + (*o_mputc)(bin2hex((c&0xf))); base64_count += 3; } else { (*o_mputc)(c); @@ -5638,32 +5959,34 @@ void mimeout_addchar(nkf_char c) void mime_prechar(nkf_char c2, nkf_char c1) { - if (mimeout_mode){ + if (mimeout_mode > 0){ if (c2 == EOF){ if (base64_count + mimeout_buf_count/3*4> 73){ (*o_base64conv)(EOF,0); - (*o_base64conv)(0,NL); - (*o_base64conv)(0,SPACE); + OCONV_NEWLINE((*o_base64conv)); + (*o_base64conv)(0,SP); + base64_count = 1; } - } else if (c2){ - if (base64_count + mimeout_buf_count/3*4> 66){ + } else { + if (base64_count + mimeout_buf_count/3*4> 66) { (*o_base64conv)(EOF,0); - (*o_base64conv)(0,NL); - (*o_base64conv)(0,SPACE); + OCONV_NEWLINE((*o_base64conv)); + (*o_base64conv)(0,SP); + base64_count = 1; + mimeout_mode = -1; } - }/*else if (mime_lastchar2){ - if (c1 <=DEL && !nkf_isspace(c1)){ - (*o_base64conv)(0,SPACE); - } - }*/ - }/*else{ - if (c2 && mime_lastchar2 == 0 - && mime_lastchar1 && !nkf_isspace(mime_lastchar1)){ - (*o_base64conv)(0,SPACE); } - }*/ - /*mime_lastchar2 = c2; - mime_lastchar1 = c1;*/ + } else if (c2) { + if (c2 != EOF && base64_count + mimeout_buf_count/3*4> 60) { + mimeout_mode = (output_mode==ASCII ||output_mode == ISO_8859_1) ? 'Q' : 'B'; + open_mime(output_mode); + (*o_base64conv)(EOF,0); + OCONV_NEWLINE((*o_base64conv)); + (*o_base64conv)(0,SP); + base64_count = 1; + mimeout_mode = -1; + } + } } void mime_putc(nkf_char c) @@ -5674,16 +5997,16 @@ void mime_putc(nkf_char c) if (mimeout_f == FIXED_MIME){ if (mimeout_mode == 'Q'){ if (base64_count > 71){ - if (c!=CR && c!=NL) { + if (c!=CR && c!=LF) { (*o_mputc)('='); - (*o_mputc)(NL); + PUT_NEWLINE((*o_mputc)); } base64_count = 0; } }else{ if (base64_count > 71){ eof_mime(); - (*o_mputc)(NL); + PUT_NEWLINE((*o_mputc)); base64_count = 0; } if (c == EOF) { /* c==EOF */ @@ -5695,14 +6018,15 @@ void mime_putc(nkf_char c) } return; } - + /* mimeout_f != FIXED_MIME */ if (c == EOF) { /* c==EOF */ + if (mimeout_mode == -1 && mimeout_buf_count > 1) open_mime(output_mode); j = mimeout_buf_count; mimeout_buf_count = 0; i = 0; - if (mimeout_mode) { + if (mimeout_mode > 0) { if (!nkf_isblank(mimeout_buf[j-1])) { for (;i 0){ + lastchar = mimeout_buf[mimeout_buf_count - 1]; + }else{ + lastchar = -1; + } + if (mimeout_mode=='Q') { - if (c <= DEL && (output_mode==ASCII ||output_mode == ISO8859_1 ) ) { - if (c == CR || c == NL) { + if (c <= DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) { + if (c == CR || c == LF) { close_mime(); (*o_mputc)(c); base64_count = 0; return; - } else if (c <= SPACE) { + } else if (c <= SP) { close_mime(); if (base64_count > 70) { - (*o_mputc)(NL); + PUT_NEWLINE((*o_mputc)); base64_count = 0; } if (!nkf_isblank(c)) { - (*o_mputc)(SPACE); + (*o_mputc)(SP); base64_count++; } - } - (*o_mputc)(c); - base64_count++; + } else { + if (base64_count > 70) { + close_mime(); + PUT_NEWLINE((*o_mputc)); + (*o_mputc)(SP); + base64_count = 1; + open_mime(output_mode); + } + if (!nkf_noescape_mime(c)) { + mimeout_addchar(c); + return; + } + } + (*o_mputc)(c); + base64_count++; } return; } - if (mimeout_buf_count > 0){ - lastchar = mimeout_buf[mimeout_buf_count - 1]; - }else{ - lastchar = -1; - } - - if (!mimeout_mode) { - if (c <= DEL && (output_mode==ASCII ||output_mode == ISO8859_1)) { + if (mimeout_mode <= 0) { + if (c <= DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) { if (nkf_isspace(c)) { - if (c==CR || c==NL) { - base64_count=0; + int flag = 0; + if (mimeout_mode == -1) { + flag = 1; + } + if (c==CR || c==LF) { + if (flag) { + open_mime(output_mode); + output_mode = 0; + } else { + base64_count = 0; + } } for (i=0;i 1 && base64_count + mimeout_buf_count > 76 - && mimeout_buf[0] != CR && mimeout_buf[0] != NL){ - (*o_mputc)(NL); + && mimeout_buf[0] != CR && mimeout_buf[0] != LF){ + PUT_NEWLINE((*o_mputc)); base64_count = 0; if (!nkf_isspace(mimeout_buf[0])){ - (*o_mputc)(SPACE); + (*o_mputc)(SP); base64_count++; } } @@ -5792,33 +6142,33 @@ void mime_putc(nkf_char c) } return; }else{ - if (lastchar==CR || lastchar == NL){ + if (lastchar==CR || lastchar == LF){ for (i=0;i0 && SPACE0 && SPMIMEOUT_BUF_LENGTH) { j = mimeout_buf_count; @@ -5866,7 +6218,7 @@ void mime_putc(nkf_char c) j = mimeout_buf_count; mimeout_buf_count = 0; for (i=0;ihiragana, 2 hiragana->katakana, 3 both\n"); - fprintf(stderr,"v Show this usage. V: show version\n"); - fprintf(stderr,"m[BQN0] MIME decode [B:base64,Q:quoted,N:non-strict,0:no decode]\n"); - fprintf(stderr,"M[BQ] MIME encode [B:base64 Q:quoted]\n"); - fprintf(stderr,"l ISO8859-1 (Latin-1) support\n"); - fprintf(stderr,"f/F Folding: -f60 or -f or -f60-10 (fold margin 10) F preserve nl\n"); - fprintf(stderr,"Z[0-4] Default/0: Convert JISX0208 Alphabet to ASCII\n"); - fprintf(stderr," 1: Kankaku to one space 2: to two spaces 3: HTML Entity\n"); - fprintf(stderr," 4: JISX0208 Katakana to JISX0201 Katakana\n"); - fprintf(stderr,"X,x Assume X0201 kana in MS-Kanji, -x preserves X0201\n"); - fprintf(stderr,"B[0-2] Broken input 0: missing ESC,1: any X on ESC-[($]-X,2: ASCII on NL\n"); + fprintf(HELP_OUTPUT," After 'W' you can add more options. -W[ 8, 16 [BL] ] \n"); +#endif + fprintf(HELP_OUTPUT,"t no conversion\n"); + fprintf(HELP_OUTPUT,"i[@B] Specify the Esc Seq for JIS X 0208-1978/83 (DEFAULT B)\n"); + fprintf(HELP_OUTPUT,"o[BJH] Specify the Esc Seq for ASCII/Roman (DEFAULT B)\n"); + fprintf(HELP_OUTPUT,"r {de/en}crypt ROT13/47\n"); + fprintf(HELP_OUTPUT,"h 1 katakana->hiragana, 2 hiragana->katakana, 3 both\n"); + fprintf(HELP_OUTPUT,"m[BQN0] MIME decode [B:base64,Q:quoted,N:non-strict,0:no decode]\n"); + fprintf(HELP_OUTPUT,"M[BQ] MIME encode [B:base64 Q:quoted]\n"); + fprintf(HELP_OUTPUT,"l ISO8859-1 (Latin-1) support\n"); + fprintf(HELP_OUTPUT,"f/F Folding: -f60 or -f or -f60-10 (fold margin 10) F preserve nl\n"); + fprintf(HELP_OUTPUT,"Z[0-4] Default/0: Convert JISX0208 Alphabet to ASCII\n"); + fprintf(HELP_OUTPUT," 1: Kankaku to one space 2: to two spaces 3: HTML Entity\n"); + fprintf(HELP_OUTPUT," 4: JISX0208 Katakana to JISX0201 Katakana\n"); + fprintf(HELP_OUTPUT,"X,x Assume X0201 kana in MS-Kanji, -x preserves X0201\n"); + fprintf(HELP_OUTPUT,"B[0-2] Broken input 0: missing ESC,1: any X on ESC-[($]-X,2: ASCII on NL\n"); #ifdef MSDOS - fprintf(stderr,"T Text mode output\n"); -#endif - fprintf(stderr,"O Output to File (DEFAULT 'nkf.out')\n"); - fprintf(stderr,"I Convert non ISO-2022-JP charactor to GETA\n"); - fprintf(stderr,"d,c Convert line breaks -d: LF -c: CRLF\n"); - fprintf(stderr,"-L[uwm] line mode u:LF w:CRLF m:CR (DEFAULT noconversion)\n"); - fprintf(stderr,"\n"); - fprintf(stderr,"Long name options\n"); - fprintf(stderr," --ic= --oc=\n"); - fprintf(stderr," Specify the input or output codeset\n"); - fprintf(stderr," --fj --unix --mac --windows\n"); - fprintf(stderr," --jis --euc --sjis --utf8 --utf16 --mime --base64\n"); - fprintf(stderr," Convert for the system or code\n"); - fprintf(stderr," --hiragana --katakana --katakana-hiragana\n"); - fprintf(stderr," To Hiragana/Katakana Conversion\n"); - fprintf(stderr," --prefix= Insert escape before troublesome characters of Shift_JIS\n"); + fprintf(HELP_OUTPUT,"T Text mode output\n"); +#endif + fprintf(HELP_OUTPUT,"O Output to File (DEFAULT 'nkf.out')\n"); + fprintf(HELP_OUTPUT,"I Convert non ISO-2022-JP charactor to GETA\n"); + fprintf(HELP_OUTPUT,"d,c Convert line breaks -d: LF -c: CRLF\n"); + fprintf(HELP_OUTPUT,"-L[uwm] line mode u:LF w:CRLF m:CR (DEFAULT noconversion)\n"); + fprintf(HELP_OUTPUT,"v, V Show this usage. V: show configuration\n"); + fprintf(HELP_OUTPUT,"\n"); + fprintf(HELP_OUTPUT,"Long name options\n"); + fprintf(HELP_OUTPUT," --ic= --oc=\n"); + fprintf(HELP_OUTPUT," Specify the input or output codeset\n"); + fprintf(HELP_OUTPUT," --fj --unix --mac --windows\n"); + fprintf(HELP_OUTPUT," --jis --euc --sjis --utf8 --utf16 --mime --base64\n"); + fprintf(HELP_OUTPUT," Convert for the system or code\n"); + fprintf(HELP_OUTPUT," --hiragana --katakana --katakana-hiragana\n"); + fprintf(HELP_OUTPUT," To Hiragana/Katakana Conversion\n"); + fprintf(HELP_OUTPUT," --prefix= Insert escape before troublesome characters of Shift_JIS\n"); #ifdef INPUT_OPTION - fprintf(stderr," --cap-input, --url-input Convert hex after ':' or '%%'\n"); + fprintf(HELP_OUTPUT," --cap-input, --url-input Convert hex after ':' or '%%'\n"); #endif #ifdef NUMCHAR_OPTION - fprintf(stderr," --numchar-input Convert Unicode Character Reference\n"); + fprintf(HELP_OUTPUT," --numchar-input Convert Unicode Character Reference\n"); #endif #ifdef UTF8_INPUT_ENABLE - fprintf(stderr," --fb-{skip, html, xml, perl, java, subchar}\n"); - fprintf(stderr," Specify how nkf handles unassigned characters\n"); + fprintf(HELP_OUTPUT," --fb-{skip, html, xml, perl, java, subchar}\n"); + fprintf(HELP_OUTPUT," Specify how nkf handles unassigned characters\n"); #endif #ifdef OVERWRITE - fprintf(stderr," --in-place[=SUFFIX] --overwrite[=SUFFIX]\n"); - fprintf(stderr," Overwrite original listed files by filtered result\n"); - fprintf(stderr," --overwrite preserves timestamp of original files\n"); -#endif - fprintf(stderr," -g --guess Guess the input code\n"); - fprintf(stderr," --help --version Show this help/the version\n"); - fprintf(stderr," For more information, see also man nkf\n"); - fprintf(stderr,"\n"); + fprintf(HELP_OUTPUT," --in-place[=SUFFIX] --overwrite[=SUFFIX]\n"); + fprintf(HELP_OUTPUT," Overwrite original listed files by filtered result\n"); + fprintf(HELP_OUTPUT," --overwrite preserves timestamp of original files\n"); +#endif + fprintf(HELP_OUTPUT," -g --guess Guess the input code\n"); + fprintf(HELP_OUTPUT," --help --version Show this help/the version\n"); + fprintf(HELP_OUTPUT," For more information, see also man nkf\n"); + fprintf(HELP_OUTPUT,"\n"); version(); } -void version(void) +void show_configuration(void) { - fprintf(stderr,"Network Kanji Filter Version %s (%s) " -#if defined(MSDOS) && !defined(__WIN32__) && !defined(__WIN16__) && !defined(__OS2__) - "for DOS" + fprintf(HELP_OUTPUT, "Summary of my nkf " NKF_VERSION " (" NKF_RELEASE_DATE ") configuration:\n"); + fprintf(HELP_OUTPUT, " Compile-time options:\n"); + fprintf(HELP_OUTPUT, " Default output encoding: " +#if defined(DEFAULT_CODE_JIS) + "ISO-2022-JP" +#elif defined(DEFAULT_CODE_SJIS) + "Shift_JIS" +#elif defined(DEFAULT_CODE_EUC) + "EUC-JP" +#elif defined(DEFAULT_CODE_UTF8) + "UTF-8" +#endif + "\n"); + fprintf(HELP_OUTPUT, " Default output newline: " +#if DEFAULT_NEWLINE == CR + "CR" +#elif DEFAULT_NEWLINE == CRLF + "CRLF" +#else + "LF" #endif -#if defined(MSDOS) && defined(__WIN16__) - "for Win16" + "\n"); + fprintf(HELP_OUTPUT, " Decode MIME encoded string: " +#if MIME_DECODE_DEFAULT + "ON" +#else + "OFF" #endif -#if defined(MSDOS) && defined(__WIN32__) - "for Win32" + "\n"); + fprintf(HELP_OUTPUT, " Convert JIS X 0201 Katakana: " +#if X0201_DEFAULT + "ON" +#else + "OFF" #endif -#ifdef __OS2__ - "for OS/2" + "\n"); +fprintf(HELP_OUTPUT, " --help, --version output: " +#if HELP_OUTPUT_HELP_OUTPUT +"HELP_OUTPUT" +#else +"STDOUT" #endif - ,NKF_VERSION,NKF_RELEASE_DATE); - fprintf(stderr,"\n%s\n",CopyRight); +"\n"); } -#endif /*PERL_XS*/ -/** - ** 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@ie.u-ryukyu.ac.jp (Shinji Kono) - ** GHG00637@nifty-serve.or.jp (COW) - ** - **/ - -/* end */ +void version(void) +{ + fprintf(HELP_OUTPUT,"Network Kanji Filter Version " NKF_VERSION " (" NKF_RELEASE_DATE ") \n" COPY_RIGHT "\n"); +} +#endif /*PERL_XS*/