]> Zhao Yanbai Git Server - minix.git/commitdiff
original netbsd's tr(1)
authorBen Gras <ben@minix3.org>
Thu, 8 Apr 2010 15:04:57 +0000 (15:04 +0000)
committerBen Gras <ben@minix3.org>
Thu, 8 Apr 2010 15:04:57 +0000 (15:04 +0000)
commands/simple/str.c [new file with mode: 0644]
commands/simple/tr.c [new file with mode: 0644]

diff --git a/commands/simple/str.c b/commands/simple/str.c
new file mode 100644 (file)
index 0000000..052e2b3
--- /dev/null
@@ -0,0 +1,330 @@
+/*     $NetBSD: str.c,v 1.12 2009/04/13 23:50:49 lukem Exp $   */
+
+/*-
+ * Copyright (c) 1991, 1993
+ *     The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <sys/cdefs.h>
+#ifndef lint
+#if 0
+static char sccsid[] = "@(#)str.c      8.2 (Berkeley) 4/28/95";
+#endif
+__RCSID("$NetBSD: str.c,v 1.12 2009/04/13 23:50:49 lukem Exp $");
+#endif /* not lint */
+
+#include <sys/types.h>
+
+#include <err.h>
+#include <errno.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+
+#include "extern.h"
+
+static int     backslash __P((STR *));
+static int     bracket __P((STR *));
+static int     c_class __P((const void *, const void *));
+static void    genclass __P((STR *));
+static void    genequiv __P((STR *));
+static int     genrange __P((STR *));
+static void    genseq __P((STR *));
+
+int
+next(s)
+       STR *s;
+{
+       int ch;
+
+       switch (s->state) {
+       case EOS:
+               return (0);
+       case INFINITE:
+               return (1);
+       case NORMAL:
+               switch (ch = *s->str) {
+               case '\0':
+                       s->state = EOS;
+                       return (0);
+               case '\\':
+                       s->lastch = backslash(s);
+                       break;
+               case '[':
+                       if (bracket(s))
+                               return (next(s));
+                       /* FALLTHROUGH */
+               default:
+                       ++s->str;
+                       s->lastch = ch;
+                       break;
+               }
+
+               /* We can start a range at any time. */
+               if (s->str[0] == '-' && genrange(s))
+                       return (next(s));
+               return (1);
+       case RANGE:
+               if (s->cnt-- == 0) {
+                       s->state = NORMAL;
+                       return (next(s));
+               }
+               ++s->lastch;
+               return (1);
+       case SEQUENCE:
+               if (s->cnt-- == 0) {
+                       s->state = NORMAL;
+                       return (next(s));
+               }
+               return (1);
+       case SET:
+               if ((s->lastch = s->set[s->cnt++]) == OOBCH) {
+                       s->state = NORMAL;
+                       return (next(s));
+               }
+               return (1);
+       }
+       /* NOTREACHED */
+       return (0);
+}
+
+static int
+bracket(s)
+       STR *s;
+{
+       char *p;
+
+       switch (s->str[1]) {
+       case ':':                               /* "[:class:]" */
+               if ((p = strstr(s->str + 2, ":]")) == NULL)
+                       return (0);
+               *p = '\0';
+               s->str += 2;
+               genclass(s);
+               s->str = p + 2;
+               return (1);
+       case '=':                               /* "[=equiv=]" */
+               if ((p = strstr(s->str + 2, "=]")) == NULL)
+                       return (0);
+               s->str += 2;
+               genequiv(s);
+               return (1);
+       default:                                /* "[\###*n]" or "[#*n]" */
+               if ((p = strpbrk(s->str + 2, "*]")) == NULL)
+                       return (0);
+               if (p[0] != '*' || strchr(p, ']') == NULL)
+                       return (0);
+               s->str += 1;
+               genseq(s);
+               return (1);
+       }
+       /* NOTREACHED */
+}
+
+typedef struct {
+       const char *name;
+       int (*func) __P((int));
+       int *set;
+} CLASS;
+
+static CLASS classes[] = {
+       { "alnum",  isalnum,  NULL, },
+       { "alpha",  isalpha,  NULL, },
+       { "blank",  isblank,  NULL, },
+       { "cntrl",  iscntrl,  NULL, },
+       { "digit",  isdigit,  NULL, },
+       { "graph",  isgraph,  NULL, },
+       { "lower",  islower,  NULL, },
+       { "print",  isprint,  NULL, },
+       { "punct",  ispunct,  NULL, },
+       { "space",  isspace,  NULL, },
+       { "upper",  isupper,  NULL, },
+       { "xdigit", isxdigit, NULL, },
+};
+
+static void
+genclass(s)
+       STR *s;
+{
+       int cnt, (*func) __P((int));
+       CLASS *cp, tmp;
+       int *p;
+
+       tmp.name = s->str;
+       if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) /
+           sizeof(CLASS), sizeof(CLASS), c_class)) == NULL)
+               errx(1, "unknown class %s", s->str);
+
+       if ((cp->set = p = malloc((NCHARS + 1) * sizeof(int))) == NULL)
+               err(1, "malloc");
+       memset(p, 0, (NCHARS + 1) * sizeof(int));
+       for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt)
+               if ((func)(cnt))
+                       *p++ = cnt;
+       *p = OOBCH;
+
+       s->cnt = 0;
+       s->state = SET;
+       s->set = cp->set;
+}
+
+static int
+c_class(a, b)
+       const void *a, *b;
+{
+       return (strcmp(((const CLASS *)a)->name, ((const CLASS *)b)->name));
+}
+
+/*
+ * English doesn't have any equivalence classes, so for now
+ * we just syntax check and grab the character.
+ */
+static void
+genequiv(s)
+       STR *s;
+{
+       if (*s->str == '\\') {
+               s->equiv[0] = backslash(s);
+               if (*s->str != '=')
+                       errx(1, "misplaced equivalence equals sign");
+       } else {
+               s->equiv[0] = s->str[0];
+               if (s->str[1] != '=')
+                       errx(1, "misplaced equivalence equals sign");
+       }
+       s->str += 2;
+       s->cnt = 0;
+       s->state = SET;
+       s->set = s->equiv;
+}
+
+static int
+genrange(s)
+       STR *s;
+{
+       int stopval;
+       char *savestart;
+
+       savestart = s->str;
+       stopval = *++s->str == '\\' ? backslash(s) : *s->str++;
+       if (stopval < (u_char)s->lastch) {
+               s->str = savestart;
+               return (0);
+       }
+       s->cnt = stopval - s->lastch + 1;
+       s->state = RANGE;
+       --s->lastch;
+       return (1);
+}
+
+static void
+genseq(s)
+       STR *s;
+{
+       char *ep;
+
+       if (s->which == STRING1)
+               errx(1, "sequences only valid in string2");
+
+       if (*s->str == '\\')
+               s->lastch = backslash(s);
+       else
+               s->lastch = *s->str++;
+       if (*s->str != '*')
+               errx(1, "misplaced sequence asterisk");
+
+       switch (*++s->str) {
+       case '\\':
+               s->cnt = backslash(s);
+               break;
+       case ']':
+               s->cnt = 0;
+               ++s->str;
+               break;
+       default:
+               if (isdigit(*s->str)) {
+                       s->cnt = strtol(s->str, &ep, 0);
+                       if (*ep == ']') {
+                               s->str = ep + 1;
+                               break;
+                       }
+               }
+               errx(1, "illegal sequence count");
+               /* NOTREACHED */
+       }
+
+       s->state = s->cnt ? SEQUENCE : INFINITE;
+}
+
+/*
+ * Translate \??? into a character.  Up to 3 octal digits, if no digits either
+ * an escape code or a literal character.
+ */
+static int
+backslash(s)
+       STR *s;
+{
+       int ch, cnt, val;
+
+       for (cnt = val = 0;;) {
+               ch = *++s->str;
+               if (!isascii(ch) || !isdigit(ch))
+                       break;
+               val = val * 8 + ch - '0';
+               if (++cnt == 3) {
+                       ++s->str;
+                       break;
+               }
+       }
+       if (cnt)
+               return (val);
+       if (ch != '\0')
+               ++s->str;
+       switch (ch) {
+               case 'a':                       /* escape characters */
+                       return ('\7');
+               case 'b':
+                       return ('\b');
+               case 'f':
+                       return ('\f');
+               case 'n':
+                       return ('\n');
+               case 'r':
+                       return ('\r');
+               case 't':
+                       return ('\t');
+               case 'v':
+                       return ('\13');
+               case '\0':                      /*  \" -> \ */
+                       s->state = EOS;
+                       return ('\\');
+               default:                        /* \x" -> x */
+                       return (ch);
+       }
+}
diff --git a/commands/simple/tr.c b/commands/simple/tr.c
new file mode 100644 (file)
index 0000000..814f245
--- /dev/null
@@ -0,0 +1,264 @@
+/*     $NetBSD: tr.c,v 1.8 2008/07/21 14:19:27 lukem Exp $     */
+
+/*
+ * Copyright (c) 1988, 1993
+ *     The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <sys/cdefs.h>
+#ifndef lint
+__COPYRIGHT("@(#) Copyright (c) 1988, 1993\
+ The Regents of the University of California.  All rights reserved.");
+#endif /* not lint */
+
+#ifndef lint
+#if 0
+static char sccsid[] = "@(#)tr.c       8.2 (Berkeley) 5/4/95";
+#endif
+__RCSID("$NetBSD: tr.c,v 1.8 2008/07/21 14:19:27 lukem Exp $");
+#endif /* not lint */
+
+#include <sys/types.h>
+
+#include <err.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include "extern.h"
+
+static int string1[NCHARS] = {
+       0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,         /* ASCII */
+       0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
+       0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
+       0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
+       0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
+       0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
+       0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
+       0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
+       0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
+       0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f,
+       0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
+       0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
+       0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
+       0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
+       0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
+       0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
+       0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
+       0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
+       0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
+       0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
+       0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7,
+       0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
+       0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7,
+       0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
+       0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
+       0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
+       0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7,
+       0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
+       0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7,
+       0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
+       0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
+       0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
+}, string2[NCHARS];
+
+STR s1 = { STRING1, NORMAL, 0, OOBCH, { 0, OOBCH }, NULL, NULL };
+STR s2 = { STRING2, NORMAL, 0, OOBCH, { 0, OOBCH }, NULL, NULL };
+
+int    main __P((int, char **));
+static void setup __P((int *, char *, STR *, int));
+static void usage __P((void));
+
+int
+main(argc, argv)
+       int argc;
+       char **argv;
+{
+       int ch, cnt, lastch, *p;
+       int cflag, dflag, sflag, isstring2;
+
+       cflag = dflag = sflag = 0;
+       while ((ch = getopt(argc, argv, "cds")) != -1)
+               switch((char)ch) {
+               case 'c':
+                       cflag = 1;
+                       break;
+               case 'd':
+                       dflag = 1;
+                       break;
+               case 's':
+                       sflag = 1;
+                       break;
+               case '?':
+               default:
+                       usage();
+               }
+       argc -= optind;
+       argv += optind;
+
+       switch(argc) {
+       case 0:
+       default:
+               usage();
+               /* NOTREACHED */
+       case 1:
+               isstring2 = 0;
+               break;
+       case 2:
+               isstring2 = 1;
+               break;
+       }
+
+       /*
+        * tr -ds [-c] string1 string2
+        * Delete all characters (or complemented characters) in string1.
+        * Squeeze all characters in string2.
+        */
+       if (dflag && sflag) {
+               if (!isstring2)
+                       usage();
+
+               setup(string1, argv[0], &s1, cflag);
+               setup(string2, argv[1], &s2, 0);
+               
+               for (lastch = OOBCH; (ch = getchar()) != EOF;)
+                       if (!string1[ch] && (!string2[ch] || lastch != ch)) {
+                               lastch = ch;
+                               (void)putchar(ch);
+                       }
+               exit(0);
+       }
+
+       /*
+        * tr -d [-c] string1
+        * Delete all characters (or complemented characters) in string1.
+        */
+       if (dflag) {
+               if (isstring2)
+                       usage();
+
+               setup(string1, argv[0], &s1, cflag);
+
+               while ((ch = getchar()) != EOF)
+                       if (!string1[ch])
+                               (void)putchar(ch);
+               exit(0);
+       }
+
+       /*
+        * tr -s [-c] string1
+        * Squeeze all characters (or complemented characters) in string1.
+        */
+       if (sflag && !isstring2) {
+               setup(string1, argv[0], &s1, cflag);
+
+               for (lastch = OOBCH; (ch = getchar()) != EOF;)
+                       if (!string1[ch] || lastch != ch) {
+                               lastch = ch;
+                               (void)putchar(ch);
+                       }
+               exit(0);
+       }
+
+       /*
+        * tr [-cs] string1 string2
+        * Replace all characters (or complemented characters) in string1 with
+        * the character in the same position in string2.  If the -s option is
+        * specified, squeeze all the characters in string2.
+        */
+       if (!isstring2)
+               usage();
+
+       s1.str = argv[0];
+       s2.str = argv[1];
+
+       if (cflag)
+               for (cnt = NCHARS, p = string1; cnt--;)
+                       *p++ = OOBCH;
+
+       if (!next(&s2))
+               errx(1, "empty string2");
+
+       /* If string2 runs out of characters, use the last one specified. */
+       if (sflag)
+               while (next(&s1)) {
+                       string1[s1.lastch] = ch = s2.lastch;
+                       string2[ch] = 1;
+                       (void)next(&s2);
+               }
+       else
+               while (next(&s1)) {
+                       string1[s1.lastch] = ch = s2.lastch;
+                       (void)next(&s2);
+               }
+
+       if (cflag)
+               for (cnt = 0, p = string1; cnt < NCHARS; ++p, ++cnt)
+                       *p = *p == OOBCH ? ch : cnt;
+
+       if (sflag)
+               for (lastch = OOBCH; (ch = getchar()) != EOF;) {
+                       ch = string1[ch];
+                       if (!string2[ch] || lastch != ch) {
+                               lastch = ch;
+                               (void)putchar(ch);
+                       }
+               }
+       else
+               while ((ch = getchar()) != EOF)
+                       (void)putchar(string1[ch]);
+       exit (0);
+}
+
+static void
+setup(string, arg, str, cflag)
+       int *string;
+       char *arg;
+       STR *str;
+       int cflag;
+{
+       int cnt, *p;
+
+       str->str = arg;
+       memset(string, 0, NCHARS * sizeof(int));
+       while (next(str))
+               string[str->lastch] = 1;
+       if (cflag)
+               for (p = string, cnt = NCHARS; cnt--; ++p)
+                       *p = !*p;
+}
+
+static void
+usage()
+{
+       (void)fprintf(stderr, "usage: tr [-cs] string1 string2\n");
+       (void)fprintf(stderr, "       tr [-c] -d string1\n");
+       (void)fprintf(stderr, "       tr [-c] -s string1\n");
+       (void)fprintf(stderr, "       tr [-c] -ds string1 string2\n");
+       exit(1);
+}