1 /**************************************************************** 2 Copyright (C) Lucent Technologies 1997 3 All Rights Reserved 4 5 Permission to use, copy, modify, and distribute this software and 6 its documentation for any purpose and without fee is hereby 7 granted, provided that the above copyright notice appear in all 8 copies and that both that the copyright notice and this 9 permission notice and warranty disclaimer appear in supporting 10 documentation, and that the name Lucent Technologies or any of 11 its entities not be used in advertising or publicity pertaining 12 to distribution of the software without specific, written prior 13 permission. 14 15 LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, 16 INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. 17 IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY 18 SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES 19 WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER 20 IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, 21 ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF 22 THIS SOFTWARE. 23 ****************************************************************/ 24 25 #if HAVE_NBTOOL_CONFIG_H 26 #include "nbtool_config.h" 27 #endif 28 29 const char *version = "version 20240728"; 30 31 #define DEBUG 32 #include <stdio.h> 33 #include <ctype.h> 34 #include <locale.h> 35 #include <langinfo.h> 36 #include <stdlib.h> 37 #include <string.h> 38 #include <signal.h> 39 #include "awk.h" 40 #include "awkgram.h" 41 42 extern char **environ; 43 extern int nfields; 44 45 int dbg = 0; 46 Awkfloat srand_seed = 1; 47 char *cmdname; /* gets argv[0] for error messages */ 48 extern FILE *yyin; /* lex input file */ 49 char *lexprog; /* points to program argument if it exists */ 50 extern int errorflag; /* non-zero if any syntax errors; set by yyerror */ 51 enum compile_states compile_time = ERROR_PRINTING; 52 53 static char **pfile; /* program filenames from -f's */ 54 static size_t maxpfile; /* max program filename */ 55 static size_t npfile; /* number of filenames */ 56 static size_t curpfile; /* current filename */ 57 58 bool CSV = false; /* true for csv input */ 59 60 bool safe = false; /* true => "safe" mode */ 61 62 size_t awk_mb_cur_max = 1; 63 64 static noreturn void fpecatch(int n 65 #ifdef SA_SIGINFO 66 , siginfo_t *si, void *uc 67 #endif 68 ) 69 { 70 #ifdef SA_SIGINFO 71 const char *mesg = NULL; 72 73 switch (si->si_code) { 74 case FPE_INTDIV: 75 mesg = "Integer divide by zero"; 76 break; 77 case FPE_INTOVF: 78 mesg = "Integer overflow"; 79 break; 80 case FPE_FLTDIV: 81 mesg = "Floating point divide by zero"; 82 break; 83 case FPE_FLTOVF: 84 mesg = "Floating point overflow"; 85 break; 86 case FPE_FLTUND: 87 mesg = "Floating point underflow"; 88 break; 89 case FPE_FLTRES: 90 mesg = "Floating point inexact result"; 91 break; 92 case FPE_FLTINV: 93 mesg = "Invalid Floating point operation"; 94 break; 95 case FPE_FLTSUB: 96 mesg = "Subscript out of range"; 97 break; 98 case 0: 99 default: 100 mesg = "Unknown error"; 101 break; 102 } 103 #endif 104 FATAL("floating point exception" 105 #ifdef SA_SIGINFO 106 ": %s", mesg 107 #endif 108 ); 109 } 110 111 /* Can this work with recursive calls? I don't think so. 112 void segvcatch(int n) 113 { 114 FATAL("segfault. Do you have an unbounded recursive call?", n); 115 } 116 */ 117 118 static const char * 119 setfs(char *p) 120 { 121 /* wart: t=>\t */ 122 if (p[0] == 't' && p[1] == '\0') 123 return "\t"; 124 return p; 125 } 126 127 static char * 128 getarg(int *argc, char ***argv, const char *msg) 129 { 130 if ((*argv)[1][2] != '\0') { /* arg is -fsomething */ 131 return &(*argv)[1][2]; 132 } else { /* arg is -f something */ 133 (*argc)--; (*argv)++; 134 if (*argc <= 1) 135 FATAL("%s", msg); 136 return (*argv)[1]; 137 } 138 } 139 140 int main(int argc, char *argv[]) 141 { 142 const char *fs = NULL; 143 char *fn, *vn; 144 145 setlocale(LC_CTYPE, ""); 146 setlocale(LC_NUMERIC, "C"); /* for parsing cmdline & prog */ 147 /* Force C locale for non-UTF-8 */ 148 if (strcmp(nl_langinfo(CODESET), "UTF-8") != 0) 149 setlocale(LC_CTYPE, "C"); 150 awk_mb_cur_max = MB_CUR_MAX; 151 cmdname = argv[0]; 152 if (argc == 1) { 153 fprintf(stderr, 154 "usage: %s [-F fs | --csv] [-v var=value] [-f progfile | 'prog'] [file ...]\n", 155 cmdname); 156 exit(1); 157 } 158 #ifdef SA_SIGINFO 159 { 160 struct sigaction sa; 161 sa.sa_sigaction = fpecatch; 162 sa.sa_flags = SA_SIGINFO; 163 sigemptyset(&sa.sa_mask); 164 (void)sigaction(SIGFPE, &sa, NULL); 165 } 166 #else 167 (void)signal(SIGFPE, fpecatch); 168 #endif 169 /*signal(SIGSEGV, segvcatch); experiment */ 170 171 /* Set and keep track of the random seed */ 172 srand_seed = 1; 173 srandom((unsigned long) srand_seed); 174 175 yyin = NULL; 176 symtab = makesymtab(NSYMTAB/NSYMTAB); 177 while (argc > 1 && argv[1][0] == '-' && argv[1][1] != '\0') { 178 if (strcmp(argv[1], "-version") == 0 || strcmp(argv[1], "--version") == 0) { 179 printf("awk %s\n", version); 180 return 0; 181 } 182 if (strcmp(argv[1], "--") == 0) { /* explicit end of args */ 183 argc--; 184 argv++; 185 break; 186 } 187 if (strcmp(argv[1], "--csv") == 0) { /* turn on csv input processing */ 188 CSV = true; 189 argc--; 190 argv++; 191 continue; 192 } 193 switch (argv[1][1]) { 194 case 's': 195 if (strcmp(argv[1], "-safe") == 0) 196 safe = true; 197 break; 198 case 'f': /* next argument is program filename */ 199 fn = getarg(&argc, &argv, "no program filename"); 200 if (npfile >= maxpfile) { 201 maxpfile += 20; 202 pfile = (char **) realloc(pfile, maxpfile * sizeof(*pfile)); 203 if (pfile == NULL) 204 FATAL("error allocating space for -f options"); 205 } 206 pfile[npfile++] = fn; 207 break; 208 case 'F': /* set field separator */ 209 fs = setfs(getarg(&argc, &argv, "no field separator")); 210 break; 211 case 'v': /* -v a=1 to be done NOW. one -v for each */ 212 vn = getarg(&argc, &argv, "no variable name"); 213 if (isclvar(vn)) 214 setclvar(vn); 215 else 216 FATAL("invalid -v option argument: %s", vn); 217 break; 218 case 'd': 219 dbg = atoi(&argv[1][2]); 220 if (dbg == 0) 221 dbg = 1; 222 printf("awk %s\n", version); 223 break; 224 default: 225 WARNING("unknown option %s ignored", argv[1]); 226 break; 227 } 228 argc--; 229 argv++; 230 } 231 232 if (CSV && (fs != NULL || lookup("FS", symtab) != NULL)) 233 WARNING("danger: don't set FS when --csv is in effect"); 234 235 /* argv[1] is now the first argument */ 236 if (npfile == 0) { /* no -f; first argument is program */ 237 if (argc <= 1) { 238 if (dbg) 239 exit(0); 240 FATAL("no program given"); 241 } 242 DPRINTF("program = |%s|\n", argv[1]); 243 lexprog = argv[1]; 244 argc--; 245 argv++; 246 } 247 recinit(recsize); 248 syminit(); 249 compile_time = COMPILING; 250 argv[0] = cmdname; /* put prog name at front of arglist */ 251 DPRINTF("argc=%d, argv[0]=%s\n", argc, argv[0]); 252 arginit(argc, argv); 253 if (!safe) 254 envinit(environ); 255 yyparse(); 256 #if 0 257 // Doing this would comply with POSIX, but is not compatible with 258 // other awks and with what most users expect. So comment it out. 259 setlocale(LC_NUMERIC, ""); /* back to whatever it is locally */ 260 #endif 261 if (fs) 262 *FS = qstring(fs, '\0'); 263 DPRINTF("errorflag=%d\n", errorflag); 264 if (errorflag == 0) { 265 compile_time = RUNNING; 266 run(winner); 267 } else 268 bracecheck(); 269 return(errorflag); 270 } 271 272 int pgetc(void) /* get 1 character from awk program */ 273 { 274 int c; 275 276 for (;;) { 277 if (yyin == NULL) { 278 if (curpfile >= npfile) 279 return EOF; 280 if (strcmp(pfile[curpfile], "-") == 0) 281 yyin = stdin; 282 else if ((yyin = fopen(pfile[curpfile], "r")) == NULL) 283 FATAL("can't open file %s", pfile[curpfile]); 284 lineno = 1; 285 } 286 if ((c = getc(yyin)) != EOF) 287 return c; 288 if (yyin != stdin) 289 fclose(yyin); 290 yyin = NULL; 291 curpfile++; 292 } 293 } 294 295 char *cursource(void) /* current source file name */ 296 { 297 if (npfile > 0) 298 return pfile[curpfile < npfile ? curpfile : curpfile - 1]; 299 else 300 return NULL; 301 } 302