From 13e91621ca0177f2fffaa7e71bd9c0bb3217dbda Mon Sep 17 00:00:00 2001 From: tg Date: Thu, 27 Apr 2017 23:12:49 +0000 Subject: [PATCH] fixup the remaining issues and last optimisations --- edit.c | 10 ++--- eval.c | 4 +- expr.c | 4 +- funcs.c | 6 +-- misc.c | 6 +-- sh.h | 134 +++++++++++++++++++++++++++++++++----------------------- shf.c | 50 ++++++++------------- 7 files changed, 113 insertions(+), 101 deletions(-) diff --git a/edit.c b/edit.c index ab3e6aa..5102ab0 100644 --- a/edit.c +++ b/edit.c @@ -28,7 +28,7 @@ #ifndef MKSH_NO_CMDLINE_EDITING -__RCSID("$MirOS: src/bin/mksh/edit.c,v 1.327 2017/04/27 20:22:22 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/edit.c,v 1.328 2017/04/27 23:12:44 tg Exp $"); /* * in later versions we might use libtermcap for this, but since external @@ -553,8 +553,7 @@ x_command_glob(int flags, char *toglob, char ***wordsp) return (nwords); } -#define IS_WORDC(c) (!ctype(c, C_LEX1 | C_QC) && \ - (c) != '`' && (c) != '=' && (c) != ':') +#define IS_WORDC(c) (!ctype(c, C_EDNWC)) static int x_locate_word(const char *buf, int buflen, int pos, int *startp, @@ -591,7 +590,7 @@ x_locate_word(const char *buf, int buflen, int pos, int *startp, /* Figure out if this is a command */ while (p >= 0 && ctype(buf[p], C_SPACE)) p--; - iscmd = p < 0 || vstrchr(";|&()`", buf[p]); + iscmd = p < 0 || ctype(buf[p], C_EDCMD); if (iscmd) { /* * If command has a /, path, etc. is not searched; @@ -859,8 +858,7 @@ x_escape(const char *s, size_t len, int (*putbuf_func)(const char *, size_t)) int rval = 0; while (wlen - add > 0) - if (vstrchr("#*=?[\\`" ":{}", s[add]) || /*…1…*/ - ctype(s[add], C_IFS | C_QC | C_DOLAR | CiQCL)) { + if (ctype(s[add], C_IFS | C_EDQ)) { if (putbuf_func(s, add) != 0) { rval = -1; break; diff --git a/eval.c b/eval.c index e768f29..615cd87 100644 --- a/eval.c +++ b/eval.c @@ -23,7 +23,7 @@ #include "sh.h" -__RCSID("$MirOS: src/bin/mksh/eval.c,v 1.205 2017/04/27 19:33:47 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/eval.c,v 1.206 2017/04/27 23:12:46 tg Exp $"); /* * string expansion @@ -1728,7 +1728,7 @@ debunk(char *dp, const char *sp, size_t dlen) memmove(dp, sp, s - sp); for (d = dp + (s - sp); *s && (d - dp < (ssize_t)dlen); s++) if (!ISMAGIC(*s) || !(*++s & 0x80) || - !vstrchr("*+?@! ", *s & 0x7f)) + !ctype(*s & 0x7F, C_PATMO)) *d++ = *s; else { /* extended pattern operators: *+?@! */ diff --git a/expr.c b/expr.c index 6ee4000..3dd4fa4 100644 --- a/expr.c +++ b/expr.c @@ -23,7 +23,7 @@ #include "sh.h" -__RCSID("$MirOS: src/bin/mksh/expr.c,v 1.95 2017/04/27 20:22:24 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/expr.c,v 1.96 2017/04/27 23:12:46 tg Exp $"); #define EXPRTOK_DEFNS #include "exprtok.h" @@ -618,7 +618,7 @@ exprtoken(Expr_state *es) goto process_tvar; #endif } else if (ctype(c, C_DIGIT)) { - while (ctype(c, C_ALNUM) || c == '#') + while (ctype(c, C_ALNUM | C_HASH)) c = *cp++; strndupx(tvar, es->tokp, --cp - es->tokp, ATEMP); process_tvar: diff --git a/funcs.c b/funcs.c index 128dc6b..a33e1d5 100644 --- a/funcs.c +++ b/funcs.c @@ -38,7 +38,7 @@ #endif #endif -__RCSID("$MirOS: src/bin/mksh/funcs.c,v 1.344 2017/04/27 20:22:24 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/funcs.c,v 1.345 2017/04/27 23:12:46 tg Exp $"); #if HAVE_KILLPG /* @@ -1456,7 +1456,7 @@ c_umask(const char **wp) if (!positions) /* default is a */ positions = 0111; - if (!vstrchr("=+-", op = *cp)) + if (!ctype((op = *cp), C_EQUAL | C_MINUS | C_PLUS)) break; cp++; new_val = 0; @@ -1497,7 +1497,7 @@ c_umask(const char **wp) if (*cp == ',') { positions = 0; cp++; - } else if (!vstrchr("=+-", *cp)) + } else if (!ctype(*cp, C_EQUAL | C_MINUS | C_PLUS)) break; } if (*cp) { diff --git a/misc.c b/misc.c index 41b2519..5f4a06b 100644 --- a/misc.c +++ b/misc.c @@ -30,7 +30,7 @@ #include #endif -__RCSID("$MirOS: src/bin/mksh/misc.c,v 1.261 2017/04/27 20:22:26 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/misc.c,v 1.262 2017/04/27 23:12:47 tg Exp $"); #define KSH_CHVT_FLAG #ifdef MKSH_SMALL @@ -703,7 +703,7 @@ has_globbing(const char *xp, const char *xpe) return (0); in_bracket = false; } - } else if ((c & 0x80) && vstrchr("*+?@! ", c & 0x7f)) { + } else if ((c & 0x80) && ctype(c & 0x7F, C_PATMO)) { saw_glob = true; if (in_bracket) bnest++; @@ -919,7 +919,7 @@ pat_scan(const unsigned char *p, const unsigned char *pe, bool match_sep) if ((*++p == /*(*/ ')' && nest-- == 0) || (*p == '|' && match_sep && nest == 0)) return (p + 1); - if ((*p & 0x80) && vstrchr("*+?@! ", *p & 0x7f)) + if ((*p & 0x80) && ctype(*p & 0x7F, C_PATMO)) nest++; } return (NULL); diff --git a/sh.h b/sh.h index dd135da..b750b3e 100644 --- a/sh.h +++ b/sh.h @@ -175,7 +175,7 @@ #endif #ifdef EXTERN -__RCSID("$MirOS: src/bin/mksh/sh.h,v 1.817 2017/04/27 20:22:27 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/sh.h,v 1.818 2017/04/27 23:12:48 tg Exp $"); #endif #define MKSH_VERSION "R55 2017/04/27" @@ -498,7 +498,7 @@ extern int __cdecl setegid(gid_t); #endif /* define bit in flag */ -#define BIT(i) (1 << (i)) +#define BIT(i) (1U << (i)) #define NELEM(a) (sizeof(a) / sizeof((a)[0])) /* @@ -1288,30 +1288,40 @@ EXTERN bool really_exit; */ /* internal types, do not reference */ -#define CiCNTRL BIT(0) /* \x00‥\x1F\x7F */ -#define CiUPPER BIT(1) /* A‥Z */ -#define CiLOWER BIT(2) /* a‥z */ -#define CiHEXLT BIT(3) /* A‥Fa‥f */ -#define CiDIGIT BIT(4) /* 0‥9 */ -#define CiOCTAL BIT(5) /* 0‥7 */ -#define CiQCX BIT(6) /* #*=?[\\]` */ -#define CiQCM BIT(7) /* +/:^{}~ */ -#define CiQCL BIT(8) /* &();<>| */ -#define CiVAR1 BIT(9) /* !#$*-?@ */ -#define CiALIAS BIT(10) /* !%,-.@ */ -#define CiSUB1 BIT(11) /* +-=? */ -#define CiSPX BIT(12) /* \x0B\x0C */ -#define CiQC BIT(13) /* "' */ -#define CiSUB2 BIT(14) /* #% */ -#define CiUNDER BIT(15) /* _ */ -#define CiNUL BIT(16) /* \x00 */ -#define CiTAB BIT(17) /* \x09 */ -#define CiNL BIT(18) /* \x0A */ -#define CiSP BIT(19) /* \x20 */ -#define CiSS BIT(20) /* $ */ -#define CiCR BIT(21) /* \x0D */ + /* initially empty — filled at runtime from $IFS */ -#define CiIFS BIT(30) +#define CiIFS BIT(0) +#define CiCNTRL BIT(1) /* \x01‥\x08\x0E‥\x1F\x7F */ +#define CiUPPER BIT(2) /* A‥Z */ +#define CiLOWER BIT(3) /* a‥z */ +#define CiHEXLT BIT(4) /* A‥Fa‥f */ +#define CiOCTAL BIT(5) /* 0‥7 */ +#define CiQCL BIT(6) /* &();| */ +#define CiALIAS BIT(7) /* !,.@ */ +#define CiQCX BIT(8) /* *[\\ */ +#define CiVAR1 BIT(9) /* !*@ */ +#define CiQCM BIT(10) /* /^~ */ +#define CiDIGIT BIT(11) /* 89 */ +#define CiQC BIT(12) /* "' */ +#define CiSPX BIT(13) /* \x0B\x0C */ +#define CiCURLY BIT(14) /* {} */ +#define CiANGLE BIT(15) /* <> */ +#define CiNUL BIT(16) /* \x00 */ +#define CiTAB BIT(17) /* \x09 */ +#define CiNL BIT(18) /* \x0A */ +#define CiCR BIT(19) /* \x0D */ +#define CiSP BIT(20) /* \x20 */ +#define CiHASH BIT(21) /* # */ +#define CiSS BIT(22) /* $ */ +#define CiPERCT BIT(23) /* % */ +#define CiPLUS BIT(24) /* + */ +#define CiMINUS BIT(25) /* - */ +#define CiCOLON BIT(26) /* : */ +#define CiEQUAL BIT(27) /* = */ +#define CiQUEST BIT(28) /* ? */ +#define CiCBRK BIT(29) /* ] */ +#define CiUNDER BIT(30) /* _ */ +#define CiGRAVE BIT(31) /* ` */ /* compile-time initialised, ASCII only */ extern const uint32_t tpl_ctypes[128]; @@ -1322,68 +1332,81 @@ EXTERN char ifs0; /* external types */ -/* 0‥9A‥Za‥z!%,-.@ valid characters in alias name */ -#define C_ALIAS (CiUPPER | CiLOWER | CiDIGIT | CiOCTAL | CiUNDER | CiALIAS) +/* !%,-.0‥9@A‥Z_a‥z valid characters in alias name */ +#define C_ALIAS (CiALIAS | CiDIGIT | CiLOWER | CiMINUS | CiOCTAL | CiPERCT | CiUNDER | CiUPPER) /* 0‥9A‥Za‥z alphanumerical */ -#define C_ALNUM (CiUPPER | CiLOWER | CiDIGIT | CiOCTAL) -/* 0‥9A‥Za‥z_ alphanumerical plus underscore (“word character”) */ -#define C_ALNUX (CiUPPER | CiLOWER | CiDIGIT | CiOCTAL | CiUNDER) +#define C_ALNUM (CiDIGIT | CiLOWER | CiOCTAL | CiUPPER) +/* 0‥9A‥Z_a‥z alphanumerical plus underscore (“word character”) */ +#define C_ALNUX (CiDIGIT | CiLOWER | CiOCTAL | CiUNDER | CiUPPER) /* A‥Za‥z alphabetical (upper plus lower) */ -#define C_ALPHA (CiUPPER | CiLOWER) -/* A‥Za‥z_ alphabetical plus underscore (identifier lead) */ -#define C_ALPHX (CiUPPER | CiLOWER | CiUNDER) +#define C_ALPHA (CiLOWER | CiUPPER) +/* A‥Z_a‥z alphabetical plus underscore (identifier lead) */ +#define C_ALPHX (CiLOWER | CiUNDER | CiUPPER) /* \x09\x20 tab and space */ -#define C_BLANK (CiTAB | CiSP) +#define C_BLANK (CiSP | CiTAB) /* \x09\x20"' separator for completion */ -#define C_CFS (CiTAB | CiSP | CiQC) +#define C_CFS (CiQC | CiSP | CiTAB) /* \x00‥\x1F\x7F POSIX control characters */ -#define C_CNTRL CiCNTRL +#define C_CNTRL (CiCNTRL | CiCR | CiNL | CiNUL | CiSPX | CiTAB) /* 0‥9 decimal digits */ #define C_DIGIT (CiDIGIT | CiOCTAL) +/* &();`| editor x_locate_word() command */ +#define C_EDCMD (CiGRAVE | CiQCL) +/* \x09\x0A\x20"&'():;<=>`| editor non-word characters */ +#define C_EDNWC (CiANGLE | CiCOLON | CiEQUAL | CiGRAVE | CiNL | CiQC | CiQCL | CiSP | CiTAB) +/* "#$&'()*:;<=>?[\\`{|} editor quotes for tab completion */ +#define C_EDQ (CiANGLE | CiCOLON | CiCURLY | CiEQUAL | CiGRAVE | CiHASH | CiQC | CiQCL | CiQCX | CiQUEST | CiSS) /* !‥~ POSIX graphical (alphanumerical plus punctuation) */ -#define C_GRAPH (CiUPPER | CiLOWER | CiDIGIT | CiOCTAL | C_PUNCT) +#define C_GRAPH (C_PUNCT | CiDIGIT | CiLOWER | CiOCTAL | CiUPPER) /* A‥Fa‥f hex letter */ #define C_HEXLT CiHEXLT /* \x00 + $IFS IFS whitespace, IFS non-whitespace, NUL */ #define C_IFS (CiIFS | CiNUL) /* \x09\x0A\x20 IFS whitespace */ -#define C_IFSWS (CiTAB | CiSP | CiNL) +#define C_IFSWS (CiNL | CiSP | CiTAB) /* \x09\x0A\x20&();<>| (for the lexer) */ -#define C_LEX1 (CiTAB | CiSP | CiNL | CiQCL) +#define C_LEX1 (CiANGLE | CiNL | CiQCL | CiSP | CiTAB) /* a‥z lowercase letters */ #define C_LOWER CiLOWER /* not alnux or dollar separator for motion */ -#define C_MFS (CiCNTRL | CiSP | CiQCM | CiALIAS | CiQC | CiQCL | CiQCX) +#define C_MFS (CiALIAS | CiANGLE | CiCBRK | CiCNTRL | CiCOLON | CiCR | CiCURLY | CiEQUAL | CiGRAVE | CiHASH | CiMINUS | CiNL | CiNUL | CiPERCT | CiPLUS | CiQC | CiQCL | CiQCM | CiQCX | CiQUEST | CiSP | CiSPX | CiTAB) /* 0‥7 octal digit */ #define C_OCTAL CiOCTAL -/* \x20‥\x7E POSIX printable characters (graph plus space) */ +/* \x20!*+?@ pattern magical operator */ +#define C_PATMO (CiPLUS | CiQUEST | CiSP | CiVAR1) +/* \x20‥~ POSIX printable characters (graph plus space) */ #define C_PRINT (C_GRAPH | CiSP) /* !"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ POSIX punctuation */ -#define C_PUNCT (CiUNDER | CiALIAS | CiSS | CiQC | CiQCL | CiQCX) +#define C_PUNCT (CiALIAS | CiANGLE | CiCBRK | CiCOLON | CiCURLY | CiEQUAL | CiGRAVE | CiHASH | CiMINUS | CiPERCT | CiPLUS | CiQC | CiQCL | CiQCM | CiQCX | CiQUEST | CiSS | CiUNDER) /* \x09\x0A"#$&'()*;<=>?[\\]`| characters requiring quoting, minus space */ -#define C_QUOTE (CiTAB | CiNL | CiSS | CiQC | CiQCL | CiQCX) +#define C_QUOTE (CiANGLE | CiCBRK | CiEQUAL | CiGRAVE | CiHASH | CiNL | CiQC | CiQCL | CiQCX | CiQUEST | CiSS | CiTAB) /* 0‥9A‥Fa‥f hexadecimal digit */ -#define C_SEDEC (CiDIGIT | CiOCTAL | CiHEXLT) +#define C_SEDEC (CiDIGIT | CiHEXLT | CiOCTAL) /* \x09‥\x0D\x20 POSIX space class */ -#define C_SPACE (CiTAB | CiSP | CiNL | CiSPX | CiCR) +#define C_SPACE (CiCR | CiNL | CiSP | CiSPX | CiTAB) /* +-=? substitution operations with word */ -#define C_SUB1 CiSUB1 +#define C_SUB1 (CiEQUAL | CiMINUS | CiPLUS | CiQUEST) /* #% substitution operations with pattern */ -#define C_SUB2 CiSUB2 +#define C_SUB2 (CiHASH | CiPERCT) /* A‥Z uppercase letters */ #define C_UPPER CiUPPER /* !#$*-?@ substitution parameters, other than positional */ -#define C_VAR1 CiVAR1 +#define C_VAR1 (CiHASH | CiMINUS | CiQUEST | CiSS | CiVAR1) /* individual chars you might like */ +#define C_COLON CiCOLON /* : colon */ #define C_DOLAR CiSS /* $ dollar sign */ +#define C_EQUAL CiEQUAL /* = equals sign */ +#define C_HASH CiHASH /* # hash sign */ #define C_LF CiNL /* \x0A ASCII line feed */ +#define C_MINUS CiMINUS /* - hyphen-minus */ #ifdef MKSH_WITH_TEXTMODE -#define C_NL (CiNL | CiCR) /* CR or LF under OS/2 TEXTMODE */ +#define C_NL (CiNL | CiCR) /* CR or LF under OS/2 TEXTMODE */ #else -#define C_NL CiNL /* LF only like under Unix */ +#define C_NL CiNL /* LF only like under Unix */ #endif #define C_NUL CiNUL /* \x00 ASCII NUL */ +#define C_PLUS CiPLUS /* + plus sign */ #define C_QC CiQC /* "' quote characters */ #define C_SPC CiSP /* \x20 ASCII space */ #define C_TAB CiTAB /* \x09 ASCII horizontal tabulator */ @@ -1391,12 +1414,15 @@ EXTERN char ifs0; /* identity transform of octet */ #define ord(c) ((unsigned int)(unsigned char)(c)) -/* identity transformation in !EBCDIC; ASCII or high in EBCDIC */ +#ifdef MKSH_EBCDIC +/* asc(c) must do a table lookup, non-ASCII map high */ +#define ksh_eq(c,u,l) (ord(c) == ord(u) || ord(c) == ord(l)) +#else #define asc(c) ord(c) -/* EBCDIC needs to compare c with both u and l */ -#define ksh_eq(c,u,l) (((c) | 0x20) == (l)) +#define ksh_eq(c,u,l) ((ord(c) | 0x20) == ord(l)) +#endif /* new fast character classes */ -#define ctype(c, t) tobool(ksh_ctypes[ord(c)] & (t)) +#define ctype(c,t) tobool(ksh_ctypes[ord(c)] & (t)) /* helper functions */ #define ksh_isdash(s) tobool(ord((s)[0]) == '-' && ord((s)[1]) == '\0') /* invariant distance even in EBCDIC */ diff --git a/shf.c b/shf.c index f57e5a2..685c77a 100644 --- a/shf.c +++ b/shf.c @@ -25,7 +25,7 @@ #include "sh.h" -__RCSID("$MirOS: src/bin/mksh/shf.c,v 1.82 2017/04/27 19:33:53 tg Exp $"); +__RCSID("$MirOS: src/bin/mksh/shf.c,v 1.83 2017/04/27 23:12:49 tg Exp $"); /* flags to shf_emptybuf() */ #define EB_READSW 0x01 /* about to switch to reading */ @@ -1162,61 +1162,49 @@ cstrerror(int errnum) /* fast character classes */ const uint32_t tpl_ctypes[128] = { /* 0x00 */ - CiCNTRL | CiNUL, CiCNTRL, - CiCNTRL, CiCNTRL, - CiCNTRL, CiCNTRL, - CiCNTRL, CiCNTRL, - CiCNTRL, CiCNTRL | CiTAB, - CiCNTRL | CiNL, CiCNTRL | CiSPX, - CiCNTRL | CiSPX, CiCNTRL | CiCR, - CiCNTRL, CiCNTRL, + CiNUL, CiCNTRL, CiCNTRL, CiCNTRL, + CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, + CiCNTRL, CiTAB, CiNL, CiSPX, + CiSPX, CiCR, CiCNTRL, CiCNTRL, /* 0x10 */ CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, CiCNTRL, /* 0x20 */ - CiSP, CiALIAS | CiVAR1, - CiQC, CiQCX | CiVAR1 | CiSUB2, - CiVAR1 | CiSS, CiALIAS | CiSUB2, - CiQCL, CiQC, - CiQCL, CiQCL, - CiQCX | CiVAR1, CiQCM | CiSUB1, - CiALIAS, CiALIAS | CiVAR1 | CiSUB1, - CiALIAS, CiQCM, + CiSP, CiALIAS | CiVAR1, CiQC, CiHASH, + CiSS, CiPERCT, CiQCL, CiQC, + CiQCL, CiQCL, CiQCX | CiVAR1, CiPLUS, + CiALIAS, CiMINUS, CiALIAS, CiQCM, /* 0x30 */ CiOCTAL, CiOCTAL, CiOCTAL, CiOCTAL, CiOCTAL, CiOCTAL, CiOCTAL, CiOCTAL, - CiDIGIT, CiDIGIT, CiQCM, CiQCL, - CiQCL, CiQCX | CiSUB1, CiQCL, CiQCX | CiVAR1 | CiSUB1, + CiDIGIT, CiDIGIT, CiCOLON, CiQCL, + CiANGLE, CiEQUAL, CiANGLE, CiQUEST, /* 0x40 */ CiALIAS | CiVAR1, CiUPPER | CiHEXLT, CiUPPER | CiHEXLT, CiUPPER | CiHEXLT, CiUPPER | CiHEXLT, CiUPPER | CiHEXLT, CiUPPER | CiHEXLT, CiUPPER, - CiUPPER, CiUPPER, - CiUPPER, CiUPPER, - CiUPPER, CiUPPER, - CiUPPER, CiUPPER, + CiUPPER, CiUPPER, CiUPPER, CiUPPER, + CiUPPER, CiUPPER, CiUPPER, CiUPPER, /* 0x50 */ CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiUPPER, CiQCX, - CiQCX, CiQCX, CiQCM, CiUNDER, + CiQCX, CiCBRK, CiQCM, CiUNDER, /* 0x60 */ - CiQCX, CiLOWER | CiHEXLT, + CiGRAVE, CiLOWER | CiHEXLT, CiLOWER | CiHEXLT, CiLOWER | CiHEXLT, CiLOWER | CiHEXLT, CiLOWER | CiHEXLT, CiLOWER | CiHEXLT, CiLOWER, - CiLOWER, CiLOWER, - CiLOWER, CiLOWER, - CiLOWER, CiLOWER, - CiLOWER, CiLOWER, + CiLOWER, CiLOWER, CiLOWER, CiLOWER, + CiLOWER, CiLOWER, CiLOWER, CiLOWER, /* 0x70 */ CiLOWER, CiLOWER, CiLOWER, CiLOWER, CiLOWER, CiLOWER, CiLOWER, CiLOWER, - CiLOWER, CiLOWER, CiLOWER, CiQCM, - CiQCL, CiQCM, CiQCM, CiCNTRL + CiLOWER, CiLOWER, CiLOWER, CiCURLY, + CiQCL, CiCURLY, CiQCM, CiCNTRL }; void