]> Zhao Yanbai Git Server - minix.git/commitdiff
Porting grep from NetBSD 37/2837/1
authorSky Liu <blackskygg@gmail.com>
Sat, 13 Sep 2014 14:59:14 +0000 (22:59 +0800)
committerLionel Sambuc <lionel@minix3.org>
Wed, 24 Sep 2014 13:33:57 +0000 (15:33 +0200)
Change-Id: I6fa9e3c0a4edc4d3ff292dc5c96ac16b34d7c792

30 files changed:
minix/commands/Makefile
minix/commands/grep/Makefile [deleted file]
minix/commands/grep/binary.c [deleted file]
minix/commands/grep/file.c [deleted file]
minix/commands/grep/grep.c [deleted file]
minix/commands/grep/grep.h [deleted file]
minix/commands/grep/mmfile.c [deleted file]
minix/commands/grep/readme.md [deleted file]
minix/commands/grep/util.c [deleted file]
minix/drivers/storage/ramdisk/Makefile
usr.bin/Makefile
usr.bin/grep/Makefile [new file with mode: 0644]
usr.bin/grep/fastgrep.c [new file with mode: 0644]
usr.bin/grep/file.c [new file with mode: 0644]
usr.bin/grep/grep.1 [moved from minix/commands/grep/grep.1 with 67% similarity]
usr.bin/grep/grep.c [new file with mode: 0644]
usr.bin/grep/grep.h [new file with mode: 0644]
usr.bin/grep/nls/C.msg [new file with mode: 0644]
usr.bin/grep/nls/es_ES.ISO8859-1.msg [new file with mode: 0644]
usr.bin/grep/nls/gl_ES.ISO8859-1.msg [new file with mode: 0644]
usr.bin/grep/nls/hu_HU.ISO8859-2.msg [new file with mode: 0644]
usr.bin/grep/nls/ja_JP.SJIS.msg [new file with mode: 0644]
usr.bin/grep/nls/ja_JP.UTF-8.msg [new file with mode: 0644]
usr.bin/grep/nls/ja_JP.eucJP.msg [new file with mode: 0644]
usr.bin/grep/nls/pt_BR.ISO8859-1.msg [new file with mode: 0644]
usr.bin/grep/nls/ru_RU.KOI8-R.msg [new file with mode: 0644]
usr.bin/grep/nls/uk_UA.UTF-8.msg [new file with mode: 0644]
usr.bin/grep/nls/zh_CN.UTF-8.msg [new file with mode: 0644]
usr.bin/grep/queue.c [moved from minix/commands/grep/queue.c with 58% similarity]
usr.bin/grep/util.c [new file with mode: 0644]

index 9882d297c921338086e4c5c7e4693aaecf05e2c6..5f50a1ca5eddbc429fe5fad36eabca8b375c0906 100644 (file)
@@ -10,7 +10,7 @@ SUBDIR=       add_route arp at backup btrace \
        dhrystone diff diskctl \
        eject fbdctl \
        find fix format fsck.mfs \
-       gcov-pull grep host \
+       gcov-pull host \
        hostaddr ifconfig ifdef \
        intr ipcrm ipcs irdpd isoread \
        loadkeys loadramdisk logger look lp \
diff --git a/minix/commands/grep/Makefile b/minix/commands/grep/Makefile
deleted file mode 100644 (file)
index db64ccd..0000000
+++ /dev/null
@@ -1,23 +0,0 @@
-#      $OpenBSD: Makefile,v 1.5 2003/06/23 07:52:18 deraadt Exp $
-
-PROG=  grep
-SRCS=  binary.c file.c grep.c mmfile.c queue.c util.c
-LINKS= ${BINDIR}/grep ${BINDIR}/egrep \
-       ${BINDIR}/grep ${BINDIR}/fgrep \
-       ${BINDIR}/grep ${BINDIR}/zgrep \
-       ${BINDIR}/grep ${BINDIR}/zegrep \
-       ${BINDIR}/grep ${BINDIR}/zfgrep \
-
-OPSYS!= uname
-MLINKS= grep.1 egrep.1 \
-       grep.1 fgrep.1 \
-       grep.1 zgrep.1 \
-       grep.1 zegrep.1 \
-       grep.1 zfgrep.1
-
-LDADD=  -lz
-DPADD=  ${LIBZ}
-
-CFLAGS+= -Wall
-
-.include <bsd.prog.mk>
diff --git a/minix/commands/grep/binary.c b/minix/commands/grep/binary.c
deleted file mode 100644 (file)
index 245059f..0000000
+++ /dev/null
@@ -1,96 +0,0 @@
-/*     $OpenBSD: binary.c,v 1.14 2005/02/07 08:47:18 otto Exp $        */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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 <ctype.h>
-#include <err.h>
-#include <stdio.h>
-
-#include "grep.h"
-
-#define        isbinary(ch)    (!isprint((ch)) && !isspace((ch)) && (ch) != '\b')
-
-int
-bin_file(FILE *f)
-{
-       unsigned char   buf[BUFSIZ];
-       size_t          i, m;
-       int             ret = 0;
-
-       if (fseek(f, 0L, SEEK_SET) == -1)
-               return 0;
-
-       if ((m = fread(buf, 1, BUFSIZ, f)) == 0)
-               return 0;
-
-       for (i = 0; i < m; i++)
-               if (isbinary(buf[i])) {
-                       ret = 1;
-                       break;
-               }
-
-       rewind(f);
-       return ret;
-}
-
-#ifndef NOZ
-int
-gzbin_file(gzFile *f)
-{
-       unsigned char   buf[BUFSIZ];
-       int             i, m;
-       int             ret = 0;
-
-       if (gzseek(f, (z_off_t)0, SEEK_SET) == -1)
-               return 0;
-
-       if ((m = gzread(f, buf, BUFSIZ)) <= 0)
-               return 0;
-
-       for (i = 0; i < m; i++)
-               if (isbinary(buf[i])) {
-                       ret = 1;
-                       break;
-               }
-
-       if (gzrewind(f) != 0)
-               err(1, "gzbin_file");
-       return ret;
-}
-#endif
-
-int
-mmbin_file(mmf_t *f)
-{
-       int i;
-
-       /* XXX knows too much about mmf internals */
-       for (i = 0; i < BUFSIZ && i < f->len; i++)
-               if (isbinary(f->base[i]))
-                       return 1;
-       return 0;
-}
diff --git a/minix/commands/grep/file.c b/minix/commands/grep/file.c
deleted file mode 100644 (file)
index 3982f07..0000000
+++ /dev/null
@@ -1,226 +0,0 @@
-/*     $OpenBSD: file.c,v 1.9 2006/02/09 09:54:46 otto Exp $   */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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/param.h>
-
-#include <err.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <unistd.h>
-
-#include "grep.h"
-
-static char     fname[MAXPATHLEN];
-#ifndef NOZ
-static char    *lnbuf;
-static size_t   lnbuflen;
-#endif
-
-#define FILE_STDIO     0
-#define FILE_MMAP      1
-#define FILE_GZIP      2
-
-struct file {
-       int      type;
-       int      noseek;
-       FILE    *f;
-       mmf_t   *mmf;
-#ifndef NOZ
-       gzFile  *gzf;
-#endif
-};
-
-#ifndef NOZ
-static char *
-gzfgetln(gzFile *f, size_t *len)
-{
-       size_t          n;
-       int             c;
-
-       for (n = 0; ; ++n) {
-               c = gzgetc(f);
-               if (c == -1) {
-                       const char *gzerrstr;
-                       int gzerr;
-
-                       if (gzeof(f))
-                               break;
-
-                       gzerrstr = gzerror(f, &gzerr);
-                       if (gzerr == Z_ERRNO)
-                               err(2, "%s", fname);
-                       else
-                               errx(2, "%s: %s", fname, gzerrstr);
-               }
-               if (n >= lnbuflen) {
-                       lnbuflen *= 2;
-                       lnbuf = grep_realloc(lnbuf, ++lnbuflen);
-               }
-               if (c == '\n')
-                       break;
-               lnbuf[n] = c;
-       }
-
-       if (gzeof(f) && n == 0)
-               return NULL;
-       *len = n;
-       return lnbuf;
-}
-#endif
-
-file_t *
-grep_fdopen(int fd, char *mode)
-{
-       file_t *f;
-
-       if (fd == STDIN_FILENO)
-               snprintf(fname, sizeof fname, "(standard input)");
-       else
-               snprintf(fname, sizeof fname, "(fd %d)", fd);
-
-       f = grep_malloc(sizeof *f);
-
-#ifndef NOZ
-       if (Zflag) {
-               f->type = FILE_GZIP;
-               f->noseek = lseek(fd, 0L, SEEK_SET) == -1;
-               if ((f->gzf = gzdopen(fd, mode)) != NULL)
-                       return f;
-       } else
-#endif
-       {
-               f->type = FILE_STDIO;
-               f->noseek = isatty(fd);
-               if ((f->f = fdopen(fd, mode)) != NULL)
-                       return f;
-       }
-
-       free(f);
-       return NULL;
-}
-
-file_t *
-grep_open(char *path, char *mode)
-{
-       file_t *f;
-
-       snprintf(fname, sizeof fname, "%s", path);
-
-       f = grep_malloc(sizeof *f);
-       f->noseek = 0;
-
-#ifndef NOZ
-       if (Zflag) {
-               f->type = FILE_GZIP;
-               if ((f->gzf = gzopen(fname, mode)) != NULL)
-                       return f;
-       } else
-#endif
-       {
-#ifdef FILE_MMAP
-               /* try mmap first; if it fails, try stdio */
-               if ((f->mmf = mmopen(fname, mode)) != NULL) {
-                       f->type = FILE_MMAP;
-                       return f;
-               }
-#endif
-               f->type = FILE_STDIO;
-               if ((f->f = fopen(path, mode)) != NULL)
-                       return f;
-       }
-
-       free(f);
-       return NULL;
-}
-
-int
-grep_bin_file(file_t *f)
-{
-       if (f->noseek)
-               return 0;
-
-       switch (f->type) {
-       case FILE_STDIO:
-               return bin_file(f->f);
-#ifdef FILE_MMAP
-       case FILE_MMAP:
-               return mmbin_file(f->mmf);
-#endif
-#ifndef NOZ
-       case FILE_GZIP:
-               return gzbin_file(f->gzf);
-#endif
-       default:
-               /* can't happen */
-               errx(2, "invalid file type");
-       }
-}
-
-char *
-grep_fgetln(file_t *f, size_t *l)
-{
-       switch (f->type) {
-       case FILE_STDIO:
-               return fgetln(f->f, l);
-#ifdef FILE_MMAP
-       case FILE_MMAP:
-               return mmfgetln(f->mmf, l);
-#endif
-#ifndef NOZ
-       case FILE_GZIP:
-               return gzfgetln(f->gzf, l);
-#endif
-       default:
-               /* can't happen */
-               errx(2, "invalid file type");
-       }
-}
-
-void
-grep_close(file_t *f)
-{
-       switch (f->type) {
-       case FILE_STDIO:
-               fclose(f->f);
-               break;
-#ifdef FILE_MMAP
-       case FILE_MMAP:
-               mmclose(f->mmf);
-               break;
-#endif
-#ifndef NOZ
-       case FILE_GZIP:
-               gzclose(f->gzf);
-               break;
-#endif
-       default:
-               /* can't happen */
-               errx(2, "invalid file type");
-       }
-       free(f);
-}
diff --git a/minix/commands/grep/grep.c b/minix/commands/grep/grep.c
deleted file mode 100644 (file)
index 310abba..0000000
+++ /dev/null
@@ -1,488 +0,0 @@
-/*     $OpenBSD: grep.c,v 1.38 2007/02/13 21:48:20 kili Exp $  */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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/types.h>
-#include <sys/stat.h>
-#include <sys/queue.h>
-
-#include <ctype.h>
-#include <err.h>
-#include <errno.h>
-#include <getopt.h>
-#include <regex.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <unistd.h>
-
-#include "grep.h"
-
-/* Flags passed to regcomp() and regexec() */
-int     cflags;
-int     eflags = REG_STARTEND;
-
-int     matchall;      /* shortcut */
-int     patterns, pattern_sz;
-char   **pattern;
-regex_t        *r_pattern;
-fastgrep_t *fg_pattern;
-
-/* For regex errors  */
-char    re_error[RE_ERROR_BUF + 1];
-
-/* Command-line flags */
-int     Aflag;         /* -A x: print x lines trailing each match */
-int     Bflag;         /* -B x: print x lines leading each match */
-int     Eflag;         /* -E: interpret pattern as extended regexp */
-int     Fflag;         /* -F: interpret pattern as list of fixed strings */
-int     Gflag;         /* -G: interpret pattern as basic regexp */
-int     Hflag;         /* -H: if -R, follow explicitly listed symlinks */
-int     Lflag;         /* -L: only show names of files with no matches */
-int     Pflag;         /* -P: if -R, no symlinks are followed */
-int     Rflag;         /* -R: recursively search directory trees */
-int     Sflag;         /* -S: if -R, follow all symlinks */
-#ifndef NOZ
-int     Zflag;         /* -Z: decompress input before processing */
-#endif
-int     bflag;         /* -b: show block numbers for each match */
-int     cflag;         /* -c: only show a count of matching lines */
-int     hflag;         /* -h: don't print filename headers */
-int     iflag;         /* -i: ignore case */
-int     lflag;         /* -l: only show names of files with matches */
-int     nflag;         /* -n: show line numbers in front of matching lines */
-int     oflag;         /* -o: always print file name */
-int     qflag;         /* -q: quiet mode (don't output anything) */
-int     sflag;         /* -s: silent mode (ignore errors) */
-int     vflag;         /* -v: only show non-matching lines */
-int     wflag;         /* -w: pattern must start and end on word boundaries */
-int     xflag;         /* -x: pattern must match entire line */
-int     lbflag;        /* --line-buffered */
-
-int binbehave = BIN_FILE_BIN;
-
-enum {
-       BIN_OPT = CHAR_MAX + 1,
-       HELP_OPT,
-       MMAP_OPT,
-       LINEBUF_OPT
-};
-
-/* Housekeeping */
-int     first;         /* flag whether or not this is our first match */
-int     tail;          /* lines left to print */
-
-struct patfile {
-       const char              *pf_file;
-       SLIST_ENTRY(patfile)     pf_next;
-};
-SLIST_HEAD(, patfile)           patfilelh;
-
-static void
-usage(void)
-{
-       fprintf(stderr,
-#ifdef NOZ
-           "usage: %s [-abcEFGHhIiLlnoPqRSsUVvwx] [-A num] [-B num] [-C[num]]\n"
-#else
-           "usage: %s [-abcEFGHhIiLlnoPqRSsUVvwxZ] [-A num] [-B num] [-C[num]]\n"
-#endif
-           "\t[-e pattern] [-f file] [--binary-files=value] [--context[=num]]\n"
-           "\t[--line-buffered] [pattern] [file ...]\n", getprogname());
-       exit(2);
-}
-
-#ifdef NOZ
-static char *optstr = "0123456789A:B:CEFGHILPSRUVabce:f:hilnoqrsuvwxy";
-#else
-static char *optstr = "0123456789A:B:CEFGHILPSRUVZabce:f:hilnoqrsuvwxy";
-#endif
-
-struct option long_options[] =
-{
-       {"binary-files",        required_argument,      NULL, BIN_OPT},
-       {"help",                no_argument,            NULL, HELP_OPT},
-       {"mmap",                no_argument,            NULL, MMAP_OPT},
-       {"line-buffered",       no_argument,            NULL, LINEBUF_OPT},
-       {"after-context",       required_argument,      NULL, 'A'},
-       {"before-context",      required_argument,      NULL, 'B'},
-       {"context",             optional_argument,      NULL, 'C'},
-       {"devices",             required_argument,      NULL, 'D'},
-       {"extended-regexp",     no_argument,            NULL, 'E'},
-       {"fixed-strings",       no_argument,            NULL, 'F'},
-       {"basic-regexp",        no_argument,            NULL, 'G'},
-       {"binary",              no_argument,            NULL, 'U'},
-       {"version",             no_argument,            NULL, 'V'},
-       {"text",                no_argument,            NULL, 'a'},
-       {"byte-offset",         no_argument,            NULL, 'b'},
-       {"count",               no_argument,            NULL, 'c'},
-       {"regexp",              required_argument,      NULL, 'e'},
-       {"file",                required_argument,      NULL, 'f'},
-       {"no-filename",         no_argument,            NULL, 'h'},
-       {"ignore-case",         no_argument,            NULL, 'i'},
-       {"files-without-match", no_argument,            NULL, 'L'},
-       {"files-with-matches",  no_argument,            NULL, 'l'},
-       {"line-number",         no_argument,            NULL, 'n'},
-       {"quiet",               no_argument,            NULL, 'q'},
-       {"silent",              no_argument,            NULL, 'q'},
-       {"recursive",           no_argument,            NULL, 'r'},
-       {"no-messages",         no_argument,            NULL, 's'},
-       {"revert-match",        no_argument,            NULL, 'v'},
-       {"word-regexp",         no_argument,            NULL, 'w'},
-       {"line-regexp",         no_argument,            NULL, 'x'},
-       {"unix-byte-offsets",   no_argument,            NULL, 'u'},
-#ifndef NOZ
-       {"decompress",          no_argument,            NULL, 'Z'},
-#endif
-       {NULL,                  no_argument,            NULL, 0}
-};
-
-
-static void
-add_pattern(char *pat, size_t len)
-{
-       if (!xflag && (len == 0 || matchall)) {
-               matchall = 1;
-               return;
-       }
-       if (patterns == pattern_sz) {
-               pattern_sz *= 2;
-               pattern = grep_realloc(pattern, ++pattern_sz * sizeof(*pattern));
-       }
-       if (len > 0 && pat[len - 1] == '\n')
-               --len;
-       /* pat may not be NUL-terminated */
-       if (wflag && !Fflag) {
-               int bol = 0, eol = 0, extra;
-               if (pat[0] == '^')
-                       bol = 1;
-               if (len > 0 && pat[len - 1] == '$')
-                       eol = 1;
-               extra = Eflag ? 2 : 4;
-               pattern[patterns] = grep_malloc(len + 15 + extra);
-               snprintf(pattern[patterns], len + 15 + extra,
-                  "%s[[:<:]]%s%.*s%s[[:>:]]%s",
-                   bol ? "^" : "",
-                   Eflag ? "(" : "\\(",
-                   (int)len - bol - eol, pat + bol,
-                   Eflag ? ")" : "\\)",
-                   eol ? "$" : "");
-               len += 14 + extra;
-       } else {
-               pattern[patterns] = grep_malloc(len + 1);
-               memcpy(pattern[patterns], pat, len);
-               pattern[patterns][len] = '\0';
-       }
-       ++patterns;
-}
-
-static void
-add_patterns(char *pats)
-{
-       char *nl;
-
-       while ((nl = strchr(pats, '\n')) != NULL) {
-               add_pattern(pats, nl - pats);
-               pats = nl + 1;
-       }
-       add_pattern(pats, strlen(pats));
-}
-
-static void
-read_patterns(const char *fn)
-{
-       FILE *f;
-       char *line;
-       size_t len;
-
-       if ((f = fopen(fn, "r")) == NULL)
-               err(2, "%s", fn);
-       while ((line = fgetln(f, &len)) != NULL)
-               add_pattern(line, *line == '\n' ? 0 : len);
-       if (ferror(f))
-               err(2, "%s", fn);
-       fclose(f);
-}
-
-int
-main(int argc, char *argv[])
-{
-       int c, lastc, prevoptind, newarg, i, needpattern;
-       struct patfile *patfile, *pf_next;
-       long l;
-       char *ep;
-
-       SLIST_INIT(&patfilelh);
-
-       switch (getprogname()[0]) {
-       case 'e':
-               Eflag++;
-               break;
-       case 'f':
-               Fflag++;
-               break;
-       case 'g':
-               Gflag++;
-               break;
-#ifndef NOZ
-       case 'z':
-               Zflag++;
-               switch(getprogname()[1]) {
-               case 'e':
-                       Eflag++;
-                       break;
-               case 'f':
-                       Fflag++;
-                       break;
-               case 'g':
-                       Gflag++;
-                       break;
-               }
-               break;
-#endif
-       }
-
-       lastc = '\0';
-       newarg = 1;
-       prevoptind = 1;
-       needpattern = 1;
-       while ((c = getopt_long(argc, argv, optstr,
-                               long_options, NULL)) != -1) {
-               switch (c) {
-               case '0': case '1': case '2': case '3': case '4':
-               case '5': case '6': case '7': case '8': case '9':
-                       if (newarg || !isdigit(lastc))
-                               Aflag = 0;
-                       else if (Aflag > INT_MAX / 10)
-                               errx(2, "context out of range");
-                       Aflag = Bflag = (Aflag * 10) + (c - '0');
-                       break;
-               case 'A':
-               case 'B':
-                       l = strtol(optarg, &ep, 10);
-                       if (ep == optarg || *ep != '\0' ||
-                           l <= 0 || l >= INT_MAX)
-                               errx(2, "context out of range");
-                       if (c == 'A')
-                               Aflag = (int)l;
-                       else
-                               Bflag = (int)l;
-                       break;
-               case 'C':
-                       if (optarg == NULL)
-                               Aflag = Bflag = 2;
-                       else {
-                               l = strtol(optarg, &ep, 10);
-                               if (ep == optarg || *ep != '\0' ||
-                                   l <= 0 || l >= INT_MAX)
-                                       errx(2, "context out of range");
-                               Aflag = Bflag = (int)l;
-                       }
-                       break;
-               case 'E':
-                       Fflag = Gflag = 0;
-                       Eflag++;
-                       break;
-               case 'F':
-                       Eflag = Gflag = 0;
-                       Fflag++;
-                       break;
-               case 'G':
-                       Eflag = Fflag = 0;
-                       Gflag++;
-                       break;
-               case 'H':
-                       Hflag++;
-                       break;
-               case 'I':
-                       binbehave = BIN_FILE_SKIP;
-                       break;
-               case 'L':
-                       lflag = 0;
-                       Lflag = qflag = 1;
-                       break;
-               case 'P':
-                       Pflag++;
-                       break;
-               case 'S':
-                       Sflag++;
-                       break;
-               case 'R':
-               case 'r':
-                       Rflag++;
-                       oflag++;
-                       break;
-               case 'U':
-                       binbehave = BIN_FILE_BIN;
-                       break;
-               case 'V':
-                       fprintf(stderr, "grep version %u.%u\n", VER_MAJ, VER_MIN);
-                       exit(0);
-                       break;
-#ifndef NOZ
-               case 'Z':
-                       Zflag++;
-                       break;
-#endif
-               case 'a':
-                       binbehave = BIN_FILE_TEXT;
-                       break;
-               case 'b':
-                       bflag = 1;
-                       break;
-               case 'c':
-                       cflag = 1;
-                       break;
-               case 'e':
-                       add_patterns(optarg);
-                       needpattern = 0;
-                       break;
-               case 'f':
-                       patfile = grep_malloc(sizeof(*patfile));
-                       patfile->pf_file = optarg;
-                       SLIST_INSERT_HEAD(&patfilelh, patfile, pf_next);
-                       needpattern = 0;
-                       break;
-               case 'h':
-                       oflag = 0;
-                       hflag = 1;
-                       break;
-               case 'i':
-               case 'y':
-                       iflag = 1;
-                       cflags |= REG_ICASE;
-                       break;
-               case 'l':
-                       Lflag = 0;
-                       lflag = qflag = 1;
-                       break;
-               case 'n':
-                       nflag = 1;
-                       break;
-               case 'o':
-                       hflag = 0;
-                       oflag = 1;
-                       break;
-               case 'q':
-                       qflag = 1;
-                       break;
-               case 's':
-                       sflag = 1;
-                       break;
-               case 'v':
-                       vflag = 1;
-                       break;
-               case 'w':
-                       wflag = 1;
-                       break;
-               case 'x':
-                       xflag = 1;
-                       break;
-               case BIN_OPT:
-                       if (strcmp("binary", optarg) == 0)
-                               binbehave = BIN_FILE_BIN;
-                       else if (strcmp("without-match", optarg) == 0)
-                               binbehave = BIN_FILE_SKIP;
-                       else if (strcmp("text", optarg) == 0)
-                               binbehave = BIN_FILE_TEXT;
-                       else
-                               errx(2, "Unknown binary-files option");
-                       break;
-               case 'u':
-               case MMAP_OPT:
-                       /* default, compatibility */
-                       break;
-               case LINEBUF_OPT:
-                       lbflag = 1;
-                       break;
-               case HELP_OPT:
-               default:
-                       usage();
-               }
-               lastc = c;
-               newarg = optind != prevoptind;
-               prevoptind = optind;
-       }
-       argc -= optind;
-       argv += optind;
-
-       for (patfile = SLIST_FIRST(&patfilelh); patfile != NULL;
-           patfile = pf_next) {
-               pf_next = SLIST_NEXT(patfile, pf_next);
-               read_patterns(patfile->pf_file);
-               free(patfile);
-       }
-
-       if (argc == 0 && needpattern)
-               usage();
-
-       if (argc != 0 && needpattern) {
-               add_patterns(*argv);
-               --argc;
-               ++argv;
-       }
-
-       if (Eflag)
-               cflags |= REG_EXTENDED;
-       fg_pattern = grep_calloc(patterns, sizeof(*fg_pattern));
-       r_pattern = grep_calloc(patterns, sizeof(*r_pattern));
-       for (i = 0; i < patterns; ++i) {
-               /* Check if cheating is allowed (always is for fgrep). */
-               if (Fflag) {
-                       fgrepcomp(&fg_pattern[i], pattern[i]);
-               } else {
-                       if (fastcomp(&fg_pattern[i], pattern[i])) {
-                               /* Fall back to full regex library */
-                               c = regcomp(&r_pattern[i], pattern[i], cflags);
-                               if (c != 0) {
-                                       regerror(c, &r_pattern[i], re_error,
-                                           RE_ERROR_BUF);
-                                       errx(2, "%s", re_error);
-                               }
-                       }
-               }
-       }
-
-#ifndef __minix
-       if (lbflag)
-               setlinebuf(stdout);
-#endif
-
-       if ((argc == 0 || argc == 1) && !oflag)
-               hflag = 1;
-
-       if (argc == 0)
-               exit(!procfile(NULL));
-
-       if (Rflag)
-               c = grep_tree(argv);
-       else
-               for (c = 0; argc--; ++argv)
-                       c += procfile(*argv);
-
-       exit(!c);
-}
diff --git a/minix/commands/grep/grep.h b/minix/commands/grep/grep.h
deleted file mode 100644 (file)
index bfbca23..0000000
+++ /dev/null
@@ -1,127 +0,0 @@
-/*     $OpenBSD: grep.h,v 1.13 2006/02/09 09:54:47 otto Exp $  */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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/types.h>
-
-#include <limits.h>
-#include <regex.h>
-#include <stdio.h>
-
-#ifndef NOZ
-#include <zlib.h>
-#endif
-
-#define VER_MAJ 1
-#define VER_MIN 2
-
-#define BIN_FILE_BIN   0
-#define BIN_FILE_SKIP  1
-#define BIN_FILE_TEXT  2
-
-typedef struct {
-       size_t           len;
-       int              line_no;
-       off_t            off;
-       char            *file;
-       char            *dat;
-} str_t;
-
-typedef struct {
-       unsigned char   *pattern;
-       int              patternLen;
-       int              qsBc[UCHAR_MAX + 1];
-       /* flags */
-       int              bol;
-       int              eol;
-       int              wmatch;
-       int              reversedSearch;
-} fastgrep_t;
-
-/* Flags passed to regcomp() and regexec() */
-extern int      cflags, eflags;
-
-/* Command line flags */
-extern int      Aflag, Bflag, Eflag, Fflag, Gflag, Hflag, Lflag, Pflag,
-                Sflag, Rflag, Zflag,
-                bflag, cflag, hflag, iflag, lflag, nflag, qflag, sflag,
-                vflag, wflag, xflag;
-extern int      binbehave;
-
-extern int      first, matchall, patterns, tail;
-extern char    **pattern;
-extern fastgrep_t *fg_pattern;
-extern regex_t *r_pattern;
-
-/* For regex errors  */
-#define RE_ERROR_BUF 512
-extern char     re_error[RE_ERROR_BUF + 1];    /* Seems big enough */
-
-/* util.c */
-int             procfile(char *fn);
-int             grep_tree(char **argv);
-void           *grep_malloc(size_t size);
-void           *grep_calloc(size_t nmemb, size_t size);
-void           *grep_realloc(void *ptr, size_t size);
-void            printline(str_t *line, int sep);
-int             fastcomp(fastgrep_t *, const char *);
-void            fgrepcomp(fastgrep_t *, const char *);
-
-/* queue.c */
-void            initqueue(void);
-void            enqueue(str_t *x);
-void            printqueue(void);
-void            clearqueue(void);
-
-/* mmfile.c */
-typedef struct mmfile {
-       int      fd;
-       size_t   len;
-       unsigned char   *base, *end, *ptr;
-} mmf_t;
-
-mmf_t          *mmopen(char *fn, char *mode);
-void            mmclose(mmf_t *mmf);
-char           *mmfgetln(mmf_t *mmf, size_t *l);
-
-/* file.c */
-struct file;
-typedef struct file file_t;
-
-file_t         *grep_fdopen(int fd, char *mode);
-file_t         *grep_open(char *path, char *mode);
-int             grep_bin_file(file_t *f);
-char           *grep_fgetln(file_t *f, size_t *l);
-void            grep_close(file_t *f);
-
-/* binary.c */
-int             bin_file(FILE * f);
-#ifndef NOZ
-int             gzbin_file(gzFile * f);
-#endif
-int             mmbin_file(mmf_t *f);
-
diff --git a/minix/commands/grep/mmfile.c b/minix/commands/grep/mmfile.c
deleted file mode 100644 (file)
index 8b6faab..0000000
+++ /dev/null
@@ -1,102 +0,0 @@
-/*     $OpenBSD: mmfile.c,v 1.11 2006/09/19 05:52:23 otto Exp $        */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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/param.h>
-#include <sys/mman.h>
-#include <sys/stat.h>
-
-#include <err.h>
-#include <fcntl.h>
-#include <stdlib.h>
-#include <unistd.h>
-
-#include "grep.h"
-
-#define MAX_MAP_LEN 1048576
-
-mmf_t *
-mmopen(char *fn, char *mode)
-{
-       mmf_t *mmf;
-       struct stat st;
-
-       /* XXX ignore mode for now */
-       mode = mode;
-
-       mmf = grep_malloc(sizeof *mmf);
-       if ((mmf->fd = open(fn, O_RDONLY)) == -1)
-               goto ouch1;
-       if (fstat(mmf->fd, &st) == -1)
-               goto ouch2;
-       if (st.st_size > SIZE_T_MAX) /* too big to mmap */
-               goto ouch2;
-       if (!S_ISREG(st.st_mode)) /* only mmap regular files */
-               goto ouch2;
-       mmf->len = (size_t)st.st_size;
-       mmf->base = mmap(NULL, mmf->len, PROT_READ, MAP_PRIVATE, mmf->fd, (off_t)0);
-       if (mmf->base == MAP_FAILED)
-               goto ouch2;
-       mmf->ptr = mmf->base;
-       mmf->end = mmf->base + mmf->len;
-#ifndef __minix
-       madvise(mmf->base, mmf->len, MADV_SEQUENTIAL);
-#endif
-       return mmf;
-
-ouch2:
-       close(mmf->fd);
-ouch1:
-       free(mmf);
-       return NULL;
-}
-
-void
-mmclose(mmf_t *mmf)
-{
-       munmap(mmf->base, mmf->len);
-       close(mmf->fd);
-       free(mmf);
-}
-
-char *
-mmfgetln(mmf_t *mmf, size_t *l)
-{
-       static unsigned char *p;
-       char *start = mmf->ptr;          /* Remove speed bump */
-       char *end = mmf->end;            /* Remove speed bump */
-
-       if (start >= end)
-               return NULL;
-       for (p = mmf->ptr; mmf->ptr < mmf->end; ++mmf->ptr)
-               if (*mmf->ptr == '\n')
-                       break;
-
-       *l = mmf->ptr - p;
-       ++mmf->ptr;
-       return p;
-}
diff --git a/minix/commands/grep/readme.md b/minix/commands/grep/readme.md
deleted file mode 100644 (file)
index 5f8d9c1..0000000
+++ /dev/null
@@ -1,17 +0,0 @@
-FreeGrep
-========
-
-The grep utility searches any given input files, selecting lines
-that match one or more patterns. By default, a pattern matches an
-input line if the regular expression in the pattern matches the
-input line without its trailing newline. An empty expression matches
-every line. Each input line that matches at least one of the patterns
-is written to the standard output. grep is used for simple patterns
-and basic regular expressions; egrep can handle extended regular
-expressions. fgrep is quicker than both grep and egrep, but can
-only handle fixed patterns (i.e. it does not interpret regular
-expressions). Patterns may consist of one or more lines, allowing
-any of the pattern lines to match a portion of the input. zgrep,
-zegrep, and zfgrep act like grep, egrep, and fgrep, respectively,
-but accept input files compressed with the compress or gzip compression
-utilities.
diff --git a/minix/commands/grep/util.c b/minix/commands/grep/util.c
deleted file mode 100644 (file)
index bbfff15..0000000
+++ /dev/null
@@ -1,594 +0,0 @@
-/*     $OpenBSD: util.c,v 1.35 2007/09/02 15:19:32 deraadt Exp $       */
-
-/*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
- * 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.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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/types.h>
-#include <sys/stat.h>
-
-#include <ctype.h>
-#include <err.h>
-#include <errno.h>
-#include <fts.h>
-#include <regex.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <unistd.h>
-
-#include "grep.h"
-
-/*
- * Process a file line by line...
- */
-
-static int     linesqueued;
-static int     procline(str_t *l, int);
-static int     grep_search(fastgrep_t *, unsigned char *, size_t, regmatch_t *pmatch);
-static int     grep_cmp(const unsigned char *, const unsigned char *, size_t);
-static void    grep_revstr(unsigned char *, int);
-
-int
-grep_tree(char **argv)
-{
-       FTS     *fts;
-       FTSENT  *p;
-       int     c, fts_flags;
-
-       c = fts_flags = 0;
-
-       if (Hflag)
-               fts_flags = FTS_COMFOLLOW;
-       if (Pflag)
-               fts_flags = FTS_PHYSICAL;
-       if (Sflag)
-               fts_flags = FTS_LOGICAL;
-
-       fts_flags |= FTS_NOSTAT | FTS_NOCHDIR;
-
-       if (!(fts = fts_open(argv, fts_flags, NULL)))
-               err(2, NULL);
-       while ((p = fts_read(fts)) != NULL) {
-               switch (p->fts_info) {
-               case FTS_DNR:
-                       /* FALL THROUGH */
-               case FTS_ERR:
-                       errx(2, "%s: %s", p->fts_path, strerror(p->fts_errno));
-                       break;
-               case FTS_D:
-               case FTS_DP:
-                       break;
-               default:
-                       c += procfile(p->fts_path);
-                       break;
-               }
-       }
-       if (errno)
-               err(2, "fts_read");
-
-       return c;
-}
-
-int
-procfile(char *fn)
-{
-       str_t ln;
-       file_t *f;
-       int c, t, z, nottext;
-
-       if (fn == NULL) {
-               fn = "(standard input)";
-               f = grep_fdopen(STDIN_FILENO, "r");
-       } else {
-               f = grep_open(fn, "r");
-       }
-       if (f == NULL) {
-               if (!sflag)
-                       warn("%s", fn);
-               return 0;
-       }
-
-       nottext = grep_bin_file(f);
-       if (nottext && binbehave == BIN_FILE_SKIP) {
-               grep_close(f);
-               return 0;
-       }
-
-       ln.file = fn;
-       ln.line_no = 0;
-       ln.len = 0;
-       linesqueued = 0;
-       tail = 0;
-       ln.off = -1;
-
-       if (Bflag > 0)
-               initqueue();
-       for (c = 0;  c == 0 || !(lflag || qflag); ) {
-               ln.off += ln.len + 1;
-               if ((ln.dat = grep_fgetln(f, &ln.len)) == NULL)
-                       break;
-               if (ln.len > 0 && ln.dat[ln.len - 1] == '\n')
-                       --ln.len;
-               ln.line_no++;
-
-               z = tail;
-
-               if ((t = procline(&ln, nottext)) == 0 && Bflag > 0 && z == 0) {
-                       enqueue(&ln);
-                       linesqueued++;
-               }
-               c += t;
-       }
-       if (Bflag > 0)
-               clearqueue();
-       grep_close(f);
-
-       if (cflag) {
-               if (!hflag)
-                       printf("%s:", ln.file);
-               printf("%u\n", c);
-       }
-       if (lflag && c != 0)
-               printf("%s\n", fn);
-       if (Lflag && c == 0)
-               printf("%s\n", fn);
-       if (c && !cflag && !lflag && !Lflag &&
-           binbehave == BIN_FILE_BIN && nottext && !qflag)
-               printf("Binary file %s matches\n", fn);
-
-       return c;
-}
-
-
-/*
- * Process an individual line in a file. Return non-zero if it matches.
- */
-
-#define isword(x) (isalnum(x) || (x) == '_')
-
-static int
-procline(str_t *l, int nottext)
-{
-       regmatch_t      pmatch;
-       int             c, i, r;
-
-       if (matchall) {
-               c = !vflag;
-               goto print;
-       }
-
-       for (c = i = 0; i < patterns; i++) {
-               if (fg_pattern[i].pattern) {
-                       r = grep_search(&fg_pattern[i], (unsigned char *)l->dat,
-                           l->len, &pmatch);
-               } else {
-                       pmatch.rm_so = 0;
-                       pmatch.rm_eo = l->len;
-                       r = regexec(&r_pattern[i], l->dat, 1, &pmatch, eflags);
-               }
-               if (r == 0 && xflag) {
-                       if (pmatch.rm_so != 0 || pmatch.rm_eo != l->len)
-                               r = REG_NOMATCH;
-               }
-               if (r == 0) {
-                       c++;
-                       break;
-               }
-       }
-       if (vflag)
-               c = !c;
-
-print:
-       if (c && binbehave == BIN_FILE_BIN && nottext)
-               return c; /* Binary file */
-
-       if ((tail > 0 || c) && !cflag && !qflag) {
-               if (c) {
-                       if (first > 0 && tail == 0 && (Bflag < linesqueued) &&
-                           (Aflag || Bflag))
-                               printf("--\n");
-                       first = 1;
-                       tail = Aflag;
-                       if (Bflag > 0)
-                               printqueue();
-                       linesqueued = 0;
-                       printline(l, ':');
-               } else {
-                       printline(l, '-');
-                       tail--;
-               }
-       }
-       return c;
-}
-
-void
-fgrepcomp(fastgrep_t *fg, const char *pattern)
-{
-       int i;
-
-       /* Initialize. */
-       fg->patternLen = strlen(pattern);
-       fg->bol = 0;
-       fg->eol = 0;
-       fg->wmatch = wflag;
-       fg->reversedSearch = 0;
-
-       /*
-        * Make a copy and upper case it for later if in -i mode,
-        * else just copy the pointer.
-        */
-       if (iflag) {
-               fg->pattern = grep_malloc(fg->patternLen + 1);
-               for (i = 0; i < fg->patternLen; i++)
-                       fg->pattern[i] = toupper(pattern[i]);
-               fg->pattern[fg->patternLen] = '\0';
-       } else
-               fg->pattern = (unsigned char *)pattern; /* really const */
-
-       /* Preprocess pattern. */
-       for (i = 0; i <= UCHAR_MAX; i++)
-               fg->qsBc[i] = fg->patternLen;
-       for (i = 1; i < fg->patternLen; i++) {
-               fg->qsBc[fg->pattern[i]] = fg->patternLen - i;
-               /*
-                * If case is ignored, make the jump apply to both upper and
-                * lower cased characters.  As the pattern is stored in upper
-                * case, apply the same to the lower case equivalents.
-                */
-               if (iflag)
-                       fg->qsBc[tolower(fg->pattern[i])] = fg->patternLen - i;
-       }
-}
-
-/*
- * Returns: -1 on failure, 0 on success
- */
-int
-fastcomp(fastgrep_t *fg, const char *pattern)
-{
-       int i;
-       int bol = 0;
-       int eol = 0;
-       int shiftPatternLen;
-       int hasDot = 0;
-       int firstHalfDot = -1;
-       int firstLastHalfDot = -1;
-       int lastHalfDot = 0;
-
-       /* Initialize. */
-       fg->patternLen = strlen(pattern);
-       fg->bol = 0;
-       fg->eol = 0;
-       fg->wmatch = 0;
-       fg->reversedSearch = 0;
-
-       /* Remove end-of-line character ('$'). */
-       if (pattern[fg->patternLen - 1] == '$') {
-               eol++;
-               fg->eol = 1;
-               fg->patternLen--;
-       }
-
-       /* Remove beginning-of-line character ('^'). */
-       if (pattern[0] == '^') {
-               bol++;
-               fg->bol = 1;
-               fg->patternLen--;
-       }
-
-       /* Remove enclosing [[:<:]] and [[:>:]] (word match). */
-       if (wflag) {
-               /* basic re's use \( \), extended re's ( ) */
-               int extra = Eflag ? 1 : 2;
-               fg->patternLen -= 14 + 2 * extra;
-               fg->wmatch = 7 + extra;
-       } else if (fg->patternLen >= 14 &&
-           strncmp(pattern + fg->bol, "[[:<:]]", 7) == 0 &&
-           strncmp(pattern + fg->bol + fg->patternLen - 7, "[[:>:]]", 7) == 0) {
-               fg->patternLen -= 14;
-               fg->wmatch = 7;
-       }
-
-       /*
-        * Copy pattern minus '^' and '$' characters as well as word
-        * match character classes at the beginning and ending of the
-        * string respectively.
-        */
-       fg->pattern = grep_malloc(fg->patternLen + 1);
-       memcpy(fg->pattern, pattern + bol + fg->wmatch, fg->patternLen);
-       fg->pattern[fg->patternLen] = '\0';
-
-       /* Look for ways to cheat...er...avoid the full regex engine. */
-       for (i = 0; i < fg->patternLen; i++)
-       {
-               /* Can still cheat? */
-               if ((isalnum(fg->pattern[i])) || isspace(fg->pattern[i]) ||
-                   (fg->pattern[i] == '_') || (fg->pattern[i] == ',') ||
-                   (fg->pattern[i] == '=') || (fg->pattern[i] == '-') ||
-                   (fg->pattern[i] == ':') || (fg->pattern[i] == '/')) {
-                       /* As long as it is good, upper case it for later. */
-                       if (iflag)
-                               fg->pattern[i] = toupper(fg->pattern[i]);
-               } else if (fg->pattern[i] == '.') {
-                       hasDot = i;
-                       if (i < fg->patternLen / 2) {
-                               if (firstHalfDot < 0)
-                                       /* Closest dot to the beginning */
-                                       firstHalfDot = i;
-                       } else {
-                               /* Closest dot to the end of the pattern. */
-                               lastHalfDot = i;
-                               if (firstLastHalfDot < 0)
-                                       firstLastHalfDot = i;
-                       }
-               } else {
-                       /* Free memory and let others know this is empty. */
-                       free(fg->pattern);
-                       fg->pattern = NULL;
-                       return (-1);
-               }
-       }
-
-       /*
-        * Determine if a reverse search would be faster based on the placement
-        * of the dots.
-        */
-       if ((!(lflag || cflag)) && ((!(bol || eol)) &&
-           ((lastHalfDot) && ((firstHalfDot < 0) ||
-           ((fg->patternLen - (lastHalfDot + 1)) < firstHalfDot))))) {
-               fg->reversedSearch = 1;
-               hasDot = fg->patternLen - (firstHalfDot < 0 ?
-                   firstLastHalfDot : firstHalfDot) - 1;
-               grep_revstr(fg->pattern, fg->patternLen);
-       }
-
-       /*
-        * Normal Quick Search would require a shift based on the position the
-        * next character after the comparison is within the pattern.  With
-        * wildcards, the position of the last dot effects the maximum shift
-        * distance.
-        * The closer to the end the wild card is the slower the search.  A
-        * reverse version of this algorithm would be useful for wildcards near
-        * the end of the string.
-        *
-        * Examples:
-        * Pattern      Max shift
-        * -------      ---------
-        * this         5
-        * .his         4
-        * t.is         3
-        * th.s         2
-        * thi.         1
-        */
-
-       /* Adjust the shift based on location of the last dot ('.'). */
-       shiftPatternLen = fg->patternLen - hasDot;
-
-       /* Preprocess pattern. */
-       for (i = 0; i <= UCHAR_MAX; i++)
-               fg->qsBc[i] = shiftPatternLen;
-       for (i = hasDot + 1; i < fg->patternLen; i++) {
-               fg->qsBc[fg->pattern[i]] = fg->patternLen - i;
-               /*
-                * If case is ignored, make the jump apply to both upper and
-                * lower cased characters.  As the pattern is stored in upper
-                * case, apply the same to the lower case equivalents.
-                */
-               if (iflag)
-                       fg->qsBc[tolower(fg->pattern[i])] = fg->patternLen - i;
-       }
-
-       /*
-        * Put pattern back to normal after pre-processing to allow for easy
-        * comparisons later.
-        */
-       if (fg->reversedSearch)
-               grep_revstr(fg->pattern, fg->patternLen);
-
-       return (0);
-}
-
-/*
- * Word boundaries using regular expressions are defined as the point
- * of transition from a non-word char to a word char, or vice versa.
- * This means that grep -w +a and grep -w a+ never match anything,
- * because they lack a starting or ending transition, but grep -w a+b
- * does match a line containing a+b.
- */
-#define wmatch(d, l, s, e)     \
-       ((s == 0 || !isword(d[s-1])) && (e == l || !isword(d[e])) && \
-         e > s && isword(d[s]) && isword(d[e-1]))
-
-static int
-grep_search(fastgrep_t *fg, unsigned char *data, size_t dataLen, regmatch_t *pmatch)
-{
-       int j;
-       int rtrnVal = REG_NOMATCH;
-
-       pmatch->rm_so = -1;
-       pmatch->rm_eo = -1;
-
-       /* No point in going farther if we do not have enough data. */
-       if (dataLen < fg->patternLen)
-               return (rtrnVal);
-
-       /* Only try once at the beginning or ending of the line. */
-       if (fg->bol || fg->eol) {
-               /* Simple text comparison. */
-               /* Verify data is >= pattern length before searching on it. */
-               if (dataLen >= fg->patternLen) {
-                       /* Determine where in data to start search at. */
-                       if (fg->eol)
-                               j = dataLen - fg->patternLen;
-                       else
-                               j = 0;
-                       if (!((fg->bol && fg->eol) && (dataLen != fg->patternLen)))
-                               if (grep_cmp(fg->pattern, data + j,
-                                   fg->patternLen) == -1) {
-                                       pmatch->rm_so = j;
-                                       pmatch->rm_eo = j + fg->patternLen;
-                                       if (!fg->wmatch || wmatch(data, dataLen,
-                                           pmatch->rm_so, pmatch->rm_eo))
-                                               rtrnVal = 0;
-                               }
-               }
-       } else if (fg->reversedSearch) {
-               /* Quick Search algorithm. */
-               j = dataLen;
-               do {
-                       if (grep_cmp(fg->pattern, data + j - fg->patternLen,
-                           fg->patternLen) == -1) {
-                               pmatch->rm_so = j - fg->patternLen;
-                               pmatch->rm_eo = j;
-                               if (!fg->wmatch || wmatch(data, dataLen,
-                                   pmatch->rm_so, pmatch->rm_eo)) {
-                                       rtrnVal = 0;
-                                       break;
-                               }
-                       }
-                       /* Shift if within bounds, otherwise, we are done. */
-                       if (j == fg->patternLen)
-                               break;
-                       j -= fg->qsBc[data[j - fg->patternLen - 1]];
-               } while (j >= fg->patternLen);
-       } else {
-               /* Quick Search algorithm. */
-               j = 0;
-               do {
-                       if (grep_cmp(fg->pattern, data + j, fg->patternLen) == -1) {
-                               pmatch->rm_so = j;
-                               pmatch->rm_eo = j + fg->patternLen;
-                               if (fg->patternLen == 0 || !fg->wmatch ||
-                                   wmatch(data, dataLen, pmatch->rm_so,
-                                   pmatch->rm_eo)) {
-                                       rtrnVal = 0;
-                                       break;
-                               }
-                       }
-
-                       /* Shift if within bounds, otherwise, we are done. */
-                       if (j + fg->patternLen == dataLen)
-                               break;
-                       else
-                               j += fg->qsBc[data[j + fg->patternLen]];
-               } while (j <= (dataLen - fg->patternLen));
-       }
-
-       return (rtrnVal);
-}
-
-
-void *
-grep_malloc(size_t size)
-{
-       void    *ptr;
-
-       if ((ptr = malloc(size)) == NULL)
-               err(2, "malloc");
-       return ptr;
-}
-
-void *
-grep_calloc(size_t nmemb, size_t size)
-{
-       void    *ptr;
-
-       if ((ptr = calloc(nmemb, size)) == NULL && (size > 0) && (nmemb > 0))
-               err(2, "calloc");
-       return ptr;
-}
-
-void *
-grep_realloc(void *ptr, size_t size)
-{
-       if ((ptr = realloc(ptr, size)) == NULL)
-               err(2, "realloc");
-       return ptr;
-}
-
-/*
- * Returns:    i >= 0 on failure (position that it failed)
- *             -1 on success
- */
-static int
-grep_cmp(const unsigned char *pattern, const unsigned char *data, size_t len)
-{
-       int i;
-
-       for (i = 0; i < len; i++) {
-               if (((pattern[i] == data[i]) || (!Fflag && pattern[i] == '.'))
-                   || (iflag && pattern[i] == toupper(data[i])))
-                       continue;
-               return (i);
-       }
-
-       return (-1);
-}
-
-static void
-grep_revstr(unsigned char *str, int len)
-{
-       int i;
-       char c;
-
-       for (i = 0; i < len / 2; i++) {
-               c = str[i];
-               str[i] = str[len - i - 1];
-               str[len - i - 1] = c;
-       }
-}
-
-void
-printline(str_t *line, int sep)
-{
-       int n;
-
-       n = 0;
-       if (!hflag) {
-               fputs(line->file, stdout);
-               ++n;
-       }
-       if (nflag) {
-               if (n)
-                       putchar(sep);
-               printf("%d", line->line_no);
-               ++n;
-       }
-       if (bflag) {
-               if (n)
-                       putchar(sep);
-#ifndef __minix
-               printf("%lld", (long long)line->off);
-#else
-               printf("%ld", (long)line->off);
-#endif
-               ++n;
-       }
-       if (n)
-               putchar(sep);
-       fwrite(line->dat, line->len, 1, stdout);
-       putchar('\n');
-}
index 84a732a8baf9f22d18c62f555ac71e1f7d488be2..72d2d0c7aec23d14c82b3a6367044e02caee6744 100644 (file)
@@ -34,7 +34,7 @@ PROGRAMS=             # defined
 PROGRAMS+=             fsck_mfs
 dir.fsck_mfs:= minix/commands/fsck.mfs
 PROGRAMS+=             grep
-dir.grep:=     minix/commands/grep
+dir.grep:=     usr.bin/grep
 PROGRAMS+=             input
 dir.input:=    minix/servers/input
 PROGRAMS+=             loadramdisk
index 792ba7e783436dc5beaf9c12e94a3319cfe752e0..8e77e7b908f0695b5cea8159ce16b35fc7a02eb9 100644 (file)
@@ -11,7 +11,7 @@ SUBDIR= asa \
        deroff dirname du \
        env expand \
        false finger fold fpr from \
-       fsplit ftp genassym getopt \
+       fsplit ftp genassym getopt grep \
        head hexdump id indent infocmp join jot \
        lam last ldd leave \
        lock login logname lorder m4 \
diff --git a/usr.bin/grep/Makefile b/usr.bin/grep/Makefile
new file mode 100644 (file)
index 0000000..5b00257
--- /dev/null
@@ -0,0 +1,40 @@
+#      $NetBSD: Makefile,v 1.5 2011/08/17 15:32:20 christos Exp $
+#      $FreeBSD: head/usr.bin/grep/Makefile 210389 2010-07-22 19:11:57Z gabor $
+#      $OpenBSD: Makefile,v 1.6 2003/06/25 15:00:04 millert Exp $
+
+PROG=  grep
+SRCS=  fastgrep.c file.c grep.c queue.c util.c
+
+LINKS= ${BINDIR}/grep ${BINDIR}/egrep  \
+       ${BINDIR}/grep ${BINDIR}/fgrep  \
+       ${BINDIR}/grep ${BINDIR}/zgrep  \
+       ${BINDIR}/grep ${BINDIR}/zegrep \
+       ${BINDIR}/grep ${BINDIR}/zfgrep
+
+MLINKS=        grep.1 egrep.1  \
+       grep.1 fgrep.1  \
+       grep.1 zgrep.1  \
+       grep.1 zegrep.1 \
+       grep.1 zfgrep.1
+
+LDADD= -lz -lbz2
+DPADD= ${LIBZ} ${LIBBZ2}
+
+.PATH: ${.CURDIR}/nls
+
+NLS=   C.msg \
+       es_ES.ISO8859-1.msg \
+       gl_ES.ISO8859-1.msg \
+       hu_HU.ISO8859-2.msg \
+       ja_JP.eucJP.msg \
+       ja_JP.SJIS.msg \
+       ja_JP.UTF-8.msg \
+       pt_BR.ISO8859-1.msg \
+       ru_RU.KOI8-R.msg \
+       uk_UA.UTF-8.msg \
+       zh_CN.UTF-8.msg
+
+COPTS.grep.c += -Wno-format-nonliteral
+COPTS.util.c += -Wno-format-nonliteral
+
+.include <bsd.prog.mk>
diff --git a/usr.bin/grep/fastgrep.c b/usr.bin/grep/fastgrep.c
new file mode 100644 (file)
index 0000000..2fcd864
--- /dev/null
@@ -0,0 +1,336 @@
+/*     $OpenBSD: util.c,v 1.36 2007/10/02 17:59:18 otto Exp $  */
+/*     $FreeBSD: head/usr.bin/grep/fastgrep.c 211496 2010-08-19 09:28:59Z des $ */
+
+/*-
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (C) 2008 Gabor Kovesdan <gabor@FreeBSD.org>
+ * 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.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ */
+
+/*
+ * XXX: This file is a speed up for grep to cover the defects of the
+ * regex library.  These optimizations should practically be implemented
+ * there keeping this code clean.  This is a future TODO, but for the
+ * meantime, we need to use this workaround.
+ */
+
+#if HAVE_NBTOOL_CONFIG_H
+#include "nbtool_config.h"
+#endif
+
+#include <sys/cdefs.h>
+__RCSID("$NetBSD: fastgrep.c,v 1.5 2011/04/18 03:27:40 joerg Exp $");
+
+#include <limits.h>
+#include <stdbool.h>
+#include <stdlib.h>
+#include <string.h>
+#include <wchar.h>
+#include <wctype.h>
+
+#include "grep.h"
+
+static inline int      grep_cmp(const unsigned char *, const unsigned char *, size_t);
+static inline void     grep_revstr(unsigned char *, int);
+
+void
+fgrepcomp(fastgrep_t *fg, const char *pat)
+{
+       unsigned int i;
+
+       /* Initialize. */
+       fg->len = strlen(pat);
+       fg->bol = false;
+       fg->eol = false;
+       fg->reversed = false;
+
+       fg->pattern = (unsigned char *)grep_strdup(pat);
+
+       /* Preprocess pattern. */
+       for (i = 0; i <= UCHAR_MAX; i++)
+               fg->qsBc[i] = fg->len;
+       for (i = 1; i < fg->len; i++)
+               fg->qsBc[fg->pattern[i]] = fg->len - i;
+}
+
+/*
+ * Returns: -1 on failure, 0 on success
+ */
+int
+fastcomp(fastgrep_t *fg, const char *pat)
+{
+       unsigned int i;
+       int firstHalfDot = -1;
+       int firstLastHalfDot = -1;
+       int hasDot = 0;
+       int lastHalfDot = 0;
+       int shiftPatternLen;
+
+       /* Initialize. */
+       fg->len = strlen(pat);
+       fg->bol = false;
+       fg->eol = false;
+       fg->reversed = false;
+       fg->word = wflag;
+
+       /* Remove end-of-line character ('$'). */
+       if (fg->len > 0 && pat[fg->len - 1] == '$') {
+               fg->eol = true;
+               fg->len--;
+       }
+
+       /* Remove beginning-of-line character ('^'). */
+       if (pat[0] == '^') {
+               fg->bol = true;
+               fg->len--;
+               pat++;
+       }
+
+       if (fg->len >= 14 &&
+           memcmp(pat, "[[:<:]]", 7) == 0 &&
+           memcmp(pat + fg->len - 7, "[[:>:]]", 7) == 0) {
+               fg->len -= 14;
+               pat += 7;
+               /* Word boundary is handled separately in util.c */
+               fg->word = true;
+       }
+
+       /*
+        * pat has been adjusted earlier to not include '^', '$' or
+        * the word match character classes at the beginning and ending
+        * of the string respectively.
+        */
+       fg->pattern = grep_malloc(fg->len + 1);
+       memcpy(fg->pattern, pat, fg->len);
+       fg->pattern[fg->len] = '\0';
+
+       /* Look for ways to cheat...er...avoid the full regex engine. */
+       for (i = 0; i < fg->len; i++) {
+               /* Can still cheat? */
+               if (fg->pattern[i] == '.') {
+                       hasDot = i;
+                       if (i < fg->len / 2) {
+                               if (firstHalfDot < 0)
+                                       /* Closest dot to the beginning */
+                                       firstHalfDot = i;
+                       } else {
+                               /* Closest dot to the end of the pattern. */
+                               lastHalfDot = i;
+                               if (firstLastHalfDot < 0)
+                                       firstLastHalfDot = i;
+                       }
+               } else {
+                       /* Free memory and let others know this is empty. */
+                       free(fg->pattern);
+                       fg->pattern = NULL;
+                       return (-1);
+               }
+       }
+
+       /*
+        * Determine if a reverse search would be faster based on the placement
+        * of the dots.
+        */
+       if ((!(lflag || cflag)) && ((!(fg->bol || fg->eol)) &&
+           ((lastHalfDot) && ((firstHalfDot < 0) ||
+           ((fg->len - (lastHalfDot + 1)) < (size_t)firstHalfDot)))) &&
+           !oflag && !color) {
+               fg->reversed = true;
+               hasDot = fg->len - (firstHalfDot < 0 ?
+                   firstLastHalfDot : firstHalfDot) - 1;
+               grep_revstr(fg->pattern, fg->len);
+       }
+
+       /*
+        * Normal Quick Search would require a shift based on the position the
+        * next character after the comparison is within the pattern.  With
+        * wildcards, the position of the last dot effects the maximum shift
+        * distance.
+        * The closer to the end the wild card is the slower the search.  A
+        * reverse version of this algorithm would be useful for wildcards near
+        * the end of the string.
+        *
+        * Examples:
+        * Pattern      Max shift
+        * -------      ---------
+        * this         5
+        * .his         4
+        * t.is         3
+        * th.s         2
+        * thi.         1
+        */
+
+       /* Adjust the shift based on location of the last dot ('.'). */
+       shiftPatternLen = fg->len - hasDot;
+
+       /* Preprocess pattern. */
+       for (i = 0; i <= (signed)UCHAR_MAX; i++)
+               fg->qsBc[i] = shiftPatternLen;
+       for (i = hasDot + 1; i < fg->len; i++) {
+               fg->qsBc[fg->pattern[i]] = fg->len - i;
+       }
+
+       /*
+        * Put pattern back to normal after pre-processing to allow for easy
+        * comparisons later.
+        */
+       if (fg->reversed)
+               grep_revstr(fg->pattern, fg->len);
+
+       return (0);
+}
+
+int
+grep_search(fastgrep_t *fg, const unsigned char *data, size_t len, regmatch_t *pmatch)
+{
+       unsigned int j;
+       int ret = REG_NOMATCH;
+
+       if (pmatch->rm_so == (ssize_t)len)
+               return (ret);
+
+       if (fg->bol && pmatch->rm_so != 0) {
+               pmatch->rm_so = len;
+               pmatch->rm_eo = len;
+               return (ret);
+       }
+
+       /* No point in going farther if we do not have enough data. */
+       if (len < fg->len)
+               return (ret);
+
+       /* Only try once at the beginning or ending of the line. */
+       if (fg->bol || fg->eol) {
+               /* Simple text comparison. */
+               /* Verify data is >= pattern length before searching on it. */
+               if (len >= fg->len) {
+                       /* Determine where in data to start search at. */
+                       j = fg->eol ? len - fg->len : 0;
+                       if (!((fg->bol && fg->eol) && (len != fg->len)))
+                               if (grep_cmp(fg->pattern, data + j,
+                                   fg->len) == -1) {
+                                       pmatch->rm_so = j;
+                                       pmatch->rm_eo = j + fg->len;
+                                               ret = 0;
+                               }
+               }
+       } else if (fg->reversed) {
+               /* Quick Search algorithm. */
+               j = len;
+               do {
+                       if (grep_cmp(fg->pattern, data + j - fg->len,
+                               fg->len) == -1) {
+                               pmatch->rm_so = j - fg->len;
+                               pmatch->rm_eo = j;
+                               ret = 0;
+                               break;
+                       }
+                       /* Shift if within bounds, otherwise, we are done. */
+                       if (j == fg->len)
+                               break;
+                       j -= fg->qsBc[data[j - fg->len - 1]];
+               } while (j >= fg->len);
+       } else {
+               /* Quick Search algorithm. */
+               j = pmatch->rm_so;
+               do {
+                       if (grep_cmp(fg->pattern, data + j, fg->len) == -1) {
+                               pmatch->rm_so = j;
+                               pmatch->rm_eo = j + fg->len;
+                               ret = 0;
+                               break;
+                       }
+
+                       /* Shift if within bounds, otherwise, we are done. */
+                       if (j + fg->len == len)
+                               break;
+                       else
+                               j += fg->qsBc[data[j + fg->len]];
+               } while (j <= (len - fg->len));
+       }
+
+       return (ret);
+}
+
+/*
+ * Returns:    i >= 0 on failure (position that it failed)
+ *             -1 on success
+ */
+static inline int
+grep_cmp(const unsigned char *pat, const unsigned char *data, size_t len)
+{
+       size_t size;
+       wchar_t *wdata, *wpat;
+       unsigned int i;
+
+       if (iflag) {
+               if ((size = mbstowcs(NULL, (const char *)data, 0)) ==
+                   ((size_t) - 1))
+                       return (-1);
+
+               wdata = grep_malloc(size * sizeof(wint_t));
+
+               if (mbstowcs(wdata, (const char *)data, size) ==
+                   ((size_t) - 1))
+                       return (-1);
+
+               if ((size = mbstowcs(NULL, (const char *)pat, 0)) ==
+                   ((size_t) - 1))
+                       return (-1);
+
+               wpat = grep_malloc(size * sizeof(wint_t));
+
+               if (mbstowcs(wpat, (const char *)pat, size) == ((size_t) - 1))
+                       return (-1);
+               for (i = 0; i < len; i++) {
+                       if ((towlower(wpat[i]) == towlower(wdata[i])) ||
+                           ((grepbehave != GREP_FIXED) && wpat[i] == L'.'))
+                               continue;
+                       free(wpat);
+                       free(wdata);
+                               return (i);
+               }
+       } else {
+               for (i = 0; i < len; i++) {
+                       if ((pat[i] == data[i]) || ((grepbehave != GREP_FIXED) &&
+                           pat[i] == '.'))
+                               continue;
+                       return (i);
+               }
+       }
+       return (-1);
+}
+
+static inline void
+grep_revstr(unsigned char *str, int len)
+{
+       int i;
+       char c;
+
+       for (i = 0; i < len / 2; i++) {
+               c = str[i];
+               str[i] = str[len - i - 1];
+               str[len - i - 1] = c;
+       }
+}
diff --git a/usr.bin/grep/file.c b/usr.bin/grep/file.c
new file mode 100644 (file)
index 0000000..60b016f
--- /dev/null
@@ -0,0 +1,259 @@
+/*     $NetBSD: file.c,v 1.7 2011/04/18 22:46:48 joerg Exp $   */
+/*     $FreeBSD: head/usr.bin/grep/file.c 211496 2010-08-19 09:28:59Z des $    */
+/*     $OpenBSD: file.c,v 1.11 2010/07/02 20:48:48 nicm Exp $  */
+
+/*-
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (C) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org>
+ * Copyright (C) 2010 Dimitry Andric <dimitry@andric.com>
+ * 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.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ */
+
+#if HAVE_NBTOOL_CONFIG_H
+#include "nbtool_config.h"
+#endif
+
+#include <sys/cdefs.h>
+__RCSID("$NetBSD: file.c,v 1.7 2011/04/18 22:46:48 joerg Exp $");
+
+#include <sys/param.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#include <bzlib.h>
+#include <err.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <wchar.h>
+#include <wctype.h>
+#include <zlib.h>
+
+#include "grep.h"
+
+#define        MAXBUFSIZ       (32 * 1024)
+#define        LNBUFBUMP       80
+
+static gzFile gzbufdesc;
+static BZFILE* bzbufdesc;
+
+static unsigned char buffer[MAXBUFSIZ];
+static unsigned char *bufpos;
+static size_t bufrem;
+
+static unsigned char *lnbuf;
+static size_t lnbuflen;
+
+static inline int
+grep_refill(struct file *f)
+{
+       ssize_t nr;
+       int bzerr;
+
+       bufpos = buffer;
+       bufrem = 0;
+
+       if (filebehave == FILE_GZIP)
+               nr = gzread(gzbufdesc, buffer, MAXBUFSIZ);
+       else if (filebehave == FILE_BZIP && bzbufdesc != NULL) {
+               nr = BZ2_bzRead(&bzerr, bzbufdesc, buffer, MAXBUFSIZ);
+               switch (bzerr) {
+               case BZ_OK:
+               case BZ_STREAM_END:
+                       /* No problem, nr will be okay */
+                       break;
+               case BZ_DATA_ERROR_MAGIC:
+                       /*
+                        * As opposed to gzread(), which simply returns the
+                        * plain file data, if it is not in the correct
+                        * compressed format, BZ2_bzRead() instead aborts.
+                        *
+                        * So, just restart at the beginning of the file again,
+                        * and use plain reads from now on.
+                        */
+                       BZ2_bzReadClose(&bzerr, bzbufdesc);
+                       bzbufdesc = NULL;
+                       if (lseek(f->fd, 0, SEEK_SET) == -1)
+                               return (-1);
+                       nr = read(f->fd, buffer, MAXBUFSIZ);
+                       break;
+               default:
+                       /* Make sure we exit with an error */
+                       nr = -1;
+               }
+       } else
+               nr = read(f->fd, buffer, MAXBUFSIZ);
+
+       if (nr < 0)
+               return (-1);
+
+       bufrem = nr;
+       return (0);
+}
+
+static inline int
+grep_lnbufgrow(size_t newlen)
+{
+
+       if (lnbuflen < newlen) {
+               lnbuf = grep_realloc(lnbuf, newlen);
+               lnbuflen = newlen;
+       }
+
+       return (0);
+}
+
+char *
+grep_fgetln(struct file *f, size_t *lenp)
+{
+       unsigned char *p;
+       char *ret;
+       size_t len;
+       size_t off;
+       ptrdiff_t diff;
+
+       /* Fill the buffer, if necessary */
+       if (bufrem == 0 && grep_refill(f) != 0)
+               goto error;
+
+       if (bufrem == 0) {
+               /* Return zero length to indicate EOF */
+               *lenp = 0;
+               return ((char *)bufpos);
+       }
+
+       /* Look for a newline in the remaining part of the buffer */
+       if ((p = memchr(bufpos, line_sep, bufrem)) != NULL) {
+               ++p; /* advance over newline */
+               ret = (char *)bufpos;
+               len = p - bufpos;
+               bufrem -= len;
+               bufpos = p;
+               *lenp = len;
+               return (ret);
+       }
+
+       /* We have to copy the current buffered data to the line buffer */
+       for (len = bufrem, off = 0; ; len += bufrem) {
+               /* Make sure there is room for more data */
+               if (grep_lnbufgrow(len + LNBUFBUMP))
+                       goto error;
+               memcpy(lnbuf + off, bufpos, len - off);
+               off = len;
+               if (grep_refill(f) != 0)
+                       goto error;
+               if (bufrem == 0)
+                       /* EOF: return partial line */
+                       break;
+               if ((p = memchr(bufpos, line_sep, bufrem)) == NULL)
+                       continue;
+               /* got it: finish up the line (like code above) */
+               ++p;
+               diff = p - bufpos;
+               len += diff;
+               if (grep_lnbufgrow(len))
+                   goto error;
+               memcpy(lnbuf + off, bufpos, diff);
+               bufrem -= diff;
+               bufpos = p;
+               break;
+       }
+       *lenp = len;
+       return ((char *)lnbuf);
+
+error:
+       *lenp = 0;
+       return (NULL);
+}
+
+static inline struct file *
+grep_file_init(struct file *f)
+{
+
+       if (filebehave == FILE_GZIP &&
+           (gzbufdesc = gzdopen(f->fd, "r")) == NULL)
+               goto error;
+
+       if (filebehave == FILE_BZIP &&
+           (bzbufdesc = BZ2_bzdopen(f->fd, "r")) == NULL)
+               goto error;
+
+       /* Fill read buffer, also catches errors early */
+       if (grep_refill(f) != 0)
+               goto error;
+
+       /* Check for binary stuff, if necessary */
+       if (!nulldataflag && binbehave != BINFILE_TEXT &&
+           memchr(bufpos, '\0', bufrem) != NULL)
+               f->binary = true;
+
+       return (f);
+error:
+       close(f->fd);
+       free(f);
+       return (NULL);
+}
+
+/*
+ * Opens a file for processing.
+ */
+struct file *
+grep_open(const char *path)
+{
+       struct file *f;
+
+       f = grep_malloc(sizeof *f);
+       memset(f, 0, sizeof *f);
+       if (path == NULL) {
+               /* Processing stdin implies --line-buffered. */
+               lbflag = true;
+               f->fd = STDIN_FILENO;
+       } else if ((f->fd = open(path, O_RDONLY)) == -1) {
+               free(f);
+               return (NULL);
+       }
+
+       return (grep_file_init(f));
+}
+
+/*
+ * Closes a file.
+ */
+void
+grep_close(struct file *f)
+{
+
+       close(f->fd);
+
+       /* Reset read buffer and line buffer */
+       bufpos = buffer;
+       bufrem = 0;
+
+       free(lnbuf);
+       lnbuf = NULL;
+       lnbuflen = 0;
+}
similarity index 67%
rename from minix/commands/grep/grep.1
rename to usr.bin/grep/grep.1
index 3ae8c5e1fbd7fd7b643e64bf211318752e5ffeec..7446cadd85b9856e5cd7d041d2bda5476b781467 100644 (file)
@@ -1,4 +1,6 @@
-.\"    $OpenBSD: grep.1,v 1.35 2007/05/31 19:20:10 jmc Exp $
+.\"    $NetBSD: grep.1,v 1.4 2012/04/08 22:00:38 wiz Exp $
+.\"    $FreeBSD: head/usr.bin/grep/grep.1 210652 2010-07-30 14:05:20Z joel $
+.\"    $OpenBSD: grep.1,v 1.38 2010/04/05 06:30:59 jmc Exp $
 .\" Copyright (c) 1980, 1990, 1993
 .\"    The Regents of the University of California.  All rights reserved.
 .\"
@@ -28,7 +30,7 @@
 .\"
 .\"    @(#)grep.1      8.3 (Berkeley) 4/18/94
 .\"
-.Dd $Mdocdate: May 31 2007 $
+.Dd April 19, 2011
 .Dt GREP 1
 .Os
 .Sh NAME
 .Nd file pattern searcher
 .Sh SYNOPSIS
 .Nm grep
-.Bk -words
-.Op Fl abcEFGHhIiLlnoPqRSsUVvwxZ
+.Op Fl abcdDEFGHhIiJLlmnOopqRSsUVvwxZz
 .Op Fl A Ar num
 .Op Fl B Ar num
 .Op Fl C Ns Op Ar num
 .Op Fl e Ar pattern
 .Op Fl f Ar file
-.Op Fl -binary-files Ns = Ns Ar value
-.Op Fl -context Ns Op = Ns Ar num
-.Op Fl -line-buffered
+.Op Fl Fl binary-files Ns = Ns Ar value
+.Op Fl Fl color Ns Op = Ns Ar when
+.Op Fl Fl colour Ns Op = Ns Ar when
+.Op Fl Fl context Ns Op = Ns Ar num
+.Op Fl Fl decompress
+.Op Fl Fl label
+.Op Fl Fl line-buffered
 .Op Ar pattern
 .Op Ar
-.Ek
 .Sh DESCRIPTION
 The
 .Nm grep
@@ -99,7 +103,7 @@ compression utilities.
 .Pp
 The following options are available:
 .Bl -tag -width indent
-.It Fl A Ar num
+.It Fl A Ar num , Fl Fl after-context Ns = Ns Ar num
 Print
 .Ar num
 lines of trailing context after each match.
@@ -108,7 +112,7 @@ See also the
 and
 .Fl C
 options.
-.It Fl a
+.It Fl a , Fl Fl text
 Treat all files as ASCII text.
 Normally
 .Nm
@@ -118,7 +122,7 @@ if files contain binary characters.
 Use of this option forces
 .Nm
 to output lines matching the specified pattern.
-.It Fl B Ar num
+.It Fl B Ar num , Fl Fl before-context Ns = Ns Ar num
 Print
 .Ar num
 lines of leading context before each match.
@@ -127,10 +131,10 @@ See also the
 and
 .Fl C
 options.
-.It Fl b
+.It Fl b , Fl Fl byte-offset
 The offset in bytes of a matched pattern is
 displayed in front of the respective matched line.
-.It Fl C Ns Op Ar num
+.It Fl C Ns Op Ar num , Fl Fl context Ns = Ns Ar num
 Print
 .Ar num
 lines of leading and trailing context surrounding each match.
@@ -141,9 +145,30 @@ The default is 2 and is equivalent to
 .Ar 2 .
 Note:
 no whitespace may be given between the option and its argument.
-.It Fl c
+.It Fl c , Fl Fl count
 Only a count of selected lines is written to standard output.
-.It Fl E
+.It Fl Fl colour Ns = Ns Op Ar when , Fl Fl color Ns = Ns Op Ar when
+Mark up the matching text with the expression stored in
+.Ev GREP_COLOR
+environment variable.
+The possible values of when can be `never', `always' or `auto'.
+.It Fl D Ar action , Fl Fl devices Ns = Ns Ar action
+Specify the demanded action for devices, FIFOs and sockets.
+The default action is `read', which means, that they are read
+as if they were normal files.
+If the action is set to `skip', devices will be silently skipped.
+.It Fl d Ar action , Fl Fl directories Ns = Ns Ar action
+Specify the demanded action for directories.
+It is `read' by default, which means that the directories
+are read in the same manner as normal files.
+Other possible values are `skip' to silently ignore the
+directories, and `recurse' to read them recursively, which
+has the same effect as the
+.Fl R
+and
+.Fl r
+option.
+.It Fl E , Fl Fl extended-regexp
 Interpret
 .Ar pattern
 as an extended regular expression
@@ -151,7 +176,7 @@ as an extended regular expression
 .Nm grep
 to behave as
 .Nm egrep ) .
-.It Fl e Ar pattern
+.It Fl e Ar pattern , Fl Fl regexp Ns = Ns Ar pattern
 Specify a pattern used during the search of the input:
 an input line is selected if it matches any of the specified patterns.
 This option is most useful when multiple
@@ -159,7 +184,33 @@ This option is most useful when multiple
 options are used to specify multiple patterns,
 or when a pattern begins with a dash
 .Pq Sq - .
-.It Fl F
+.It Fl Fl exclude
+If specified, it excludes files matching the given
+filename pattern from the search.
+Note that
+.Fl Fl exclude
+patterns take priority over
+.Fl Fl include
+patterns, and if no
+.Fl Fl include
+pattern is specified, all files are searched that are
+not excluded.
+Patterns are matched to the full path specified,
+not only to the filename component.
+.It Fl Fl exclude-dir
+If
+.Fl R
+is specified, it excludes directories matching the
+given filename pattern from the search.
+Note that
+.Fl Fl exclude-dir
+patterns take priority over
+.Fl Fl include-dir
+patterns, and if no
+.Fl Fl include-dir
+pattern is specified, all directories are searched that are
+not excluded.
+.It Fl F , Fl Fl fixed-strings
 Interpret
 .Ar pattern
 as a set of fixed strings
@@ -167,7 +218,7 @@ as a set of fixed strings
 .Nm grep
 to behave as
 .Nm fgrep ) .
-.It Fl f Ar file
+.It Fl f Ar file , Fl Fl file Ns = Ns Ar file
 Read one or more newline separated patterns from
 .Ar file .
 Empty pattern lines match every input line.
@@ -175,7 +226,7 @@ Newlines are not considered part of a pattern.
 If
 .Ar file
 is empty, nothing is matched.
-.It Fl G
+.It Fl G , Fl Fl basic-regexp
 Interpret
 .Ar pattern
 as a basic regular expression
@@ -184,30 +235,55 @@ as a basic regular expression
 to behave as traditional
 .Nm grep ) .
 .It Fl H
-If
-.Fl R
-is specified, follow symbolic links only if they were explicitly listed
-on the command line.
-The default is not to follow symbolic links.
-.It Fl h
+Always print filename headers with output lines.
+.It Fl h , Fl Fl no-filename
 Never print filename headers
 .Pq i.e. filenames
 with output lines.
+.It Fl Fl help
+Print a brief help message.
 .It Fl I
 Ignore binary files.
-.It Fl i
+This option is equivalent to
+.Fl Fl binary-file Ns = Ns Ar without-match
+option.
+.It Fl i , Fl Fl ignore-case
 Perform case insensitive matching.
 By default,
 .Nm grep
 is case sensitive.
-.It Fl L
+.It Fl Fl include
+If specified, only files matching the
+given filename pattern are searched.
+Note that
+.Fl Fl exclude
+patterns take priority over
+.Fl Fl include
+patterns.
+Patterns are matched to the full path specified,
+not only to the filename component.
+.It Fl Fl include-dir
+If
+.Fl R
+is specified, only directories matching the
+given filename pattern are searched.
+Note that
+.Fl Fl exclude-dir
+patterns take priority over
+.Fl Fl include-dir
+patterns.
+.It Fl J, Fl Fl bz2decompress
+Decompress the
+.Xr bzip2 1
+compressed file before looking for the text.
+.It Fl L , Fl Fl files-without-match
 Only the names of files not containing selected lines are written to
 standard output.
 Pathnames are listed once per file searched.
 If the standard input is searched, the string
 .Dq (standard input)
 is written.
-.It Fl l
+.It Fl l , Fl Fl files-with-matches
 Only the names of files containing selected lines are written to
 standard output.
 .Nm grep
@@ -217,7 +293,18 @@ Pathnames are listed once per file searched.
 If the standard input is searched, the string
 .Dq (standard input)
 is written.
-.It Fl n
+.It Fl Fl mmap
+Use
+.Xr mmap 2
+instead of
+.Xr read 2
+to read input, which can result in better performance under some
+circumstances but can cause undefined behaviour.
+.It Fl m Ar num, Fl Fl max-count Ns = Ns Ar num
+Stop reading the file after
+.Ar num
+matches.
+.It Fl n , Fl Fl line-number
 Each output line is preceded by its relative line number in the file,
 starting at line 1.
 The line number counter is reset for each file processed.
@@ -229,54 +316,62 @@ or
 .Fl q
 is
 specified.
-.It Fl o
-Always print filename headers with output lines.
-.It Fl P
+.It Fl O
+If
+.Fl R
+is specified, follow symbolic links only if they were explicitly listed
+on the command line.
+The default is not to follow symbolic links.
+.It Fl o, Fl Fl only-matching
+Prints only the matching part of the lines.
+.It Fl p
 If
 .Fl R
 is specified, no symbolic links are followed.
 This is the default.
-.It Fl q
+.It Fl q , Fl Fl quiet , Fl Fl silent
 Quiet mode:
 suppress normal output.
 .Nm grep
 will only search a file until a match has been found,
 making searches potentially less expensive.
-.It Fl R
+.It Fl R , Fl r , Fl Fl recursive
 Recursively search subdirectories listed.
 .It Fl S
 If
 .Fl R
 is specified, all symbolic links are followed.
 The default is not to follow symbolic links.
-.It Fl s
+.It Fl s , Fl Fl no-messages
 Silent mode.
 Nonexistent and unreadable files are ignored
 (i.e. their error messages are suppressed).
-.It Fl U
+.It Fl U , Fl Fl binary
 Search binary files, but do not attempt to print them.
-.It Fl V
-Display version information.
-All other options are ignored.
-.It Fl v
+.It Fl V , Fl Fl version
+Display version information and exit.
+.It Fl v , Fl Fl invert-match
 Selected lines are those
 .Em not
 matching any of the specified patterns.
-.It Fl w
+.It Fl w , Fl Fl word-regexp
 The expression is searched for as a word (as if surrounded by
 .Sq [[:<:]]
 and
 .Sq [[:>:]] ;
 see
 .Xr re_format 7 ) .
-.It Fl x
+.It Fl x , Fl Fl line-regexp
 Only input lines selected against an entire fixed string or regular
 expression are considered to be matching lines.
-.It Fl Z
-Force
-.Nm grep
-to behave as
-.Nm zgrep .
+.It Fl y
+Equivalent to
+.Fl i .
+Obsoleted.
+.It Fl Z , Fl Fl null
+Prints a zero-byte after the file name.
+.It Fl z , Fl Fl null-data
+Use the zero byte (ASCII NUL) as line separator.
 .It Fl Fl binary-files Ns = Ns Ar value
 Controls searching and printing of binary files.
 Options are
@@ -287,13 +382,16 @@ do not search binary files;
 and
 .Ar text :
 treat all files as text.
-.Sm off
-.It Fl Fl context Op = Ar num
-.Sm on
-Print
-.Ar num
-lines of leading and trailing context.
-The default is 2.
+.It Fl Fl decompress
+Detect input files compressed with
+.Xr bzip2 1
+or
+.Xr gzip 1
+and decompress them dynamically.
+This makes
+.Nm grep
+behave like
+.Nm zgrep .
 .It Fl Fl line-buffered
 Force output to be line buffered.
 By default, output is line buffered when standard output is a terminal
@@ -301,7 +399,7 @@ and block buffered otherwise.
 .Pp
 .El
 If no file arguments are specified, the standard input is used.
-.Sh RETURN VALUES
+.Sh EXIT STATUS
 The
 .Nm grep
 utility exits with one of the following values:
@@ -367,7 +465,7 @@ utility is compliant with the
 specification.
 .Pp
 The flags
-.Op Fl AaBbCGHhILoPRSUVwZ
+.Op Fl AaBbCDdGHhIJLmoPRSUVwZ
 are extensions to that specification, and the behaviour of the
 .Fl f
 flag when used with an empty pattern file is left undefined.
diff --git a/usr.bin/grep/grep.c b/usr.bin/grep/grep.c
new file mode 100644 (file)
index 0000000..b3e43f3
--- /dev/null
@@ -0,0 +1,708 @@
+/*     $NetBSD: grep.c,v 1.11 2012/05/06 22:27:00 joerg Exp $  */
+/*     $FreeBSD: head/usr.bin/grep/grep.c 211519 2010-08-19 22:55:17Z delphij $        */
+/*     $OpenBSD: grep.c,v 1.42 2010/07/02 22:18:03 tedu Exp $  */
+
+/*-
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (C) 2008-2009 Gabor Kovesdan <gabor@FreeBSD.org>
+ * 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.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ */
+
+#if HAVE_NBTOOL_CONFIG_H
+#include "nbtool_config.h"
+#endif
+
+#include <sys/cdefs.h>
+__RCSID("$NetBSD: grep.c,v 1.11 2012/05/06 22:27:00 joerg Exp $");
+
+#include <sys/stat.h>
+#include <sys/types.h>
+
+#include <ctype.h>
+#include <err.h>
+#include <errno.h>
+#include <getopt.h>
+#include <limits.h>
+#include <libgen.h>
+#include <locale.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include "grep.h"
+
+#ifndef WITHOUT_NLS
+#include <nl_types.h>
+nl_catd         catalog;
+#endif
+
+/*
+ * Default messags to use when NLS is disabled or no catalogue
+ * is found.
+ */
+const char     *errstr[] = {
+       "",
+/* 1*/ "(standard input)",
+/* 2*/ "cannot read bzip2 compressed file",
+/* 3*/ "unknown %s option",
+/* 4*/ "usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n",
+/* 5*/ "\t[-e pattern] [-f file] [--binary-files=value] [--color=when]\n",
+/* 6*/ "\t[--context[=num]] [--directories=action] [--label] [--line-buffered]\n",
+/* 7*/ "\t[pattern] [file ...]\n",
+/* 8*/ "Binary file %s matches\n",
+/* 9*/ "%s (BSD grep) %s\n",
+};
+
+/* Flags passed to regcomp() and regexec() */
+int             cflags = 0;
+int             eflags = REG_STARTEND;
+
+/* Searching patterns */
+unsigned int    patterns, pattern_sz;
+char           **pattern;
+regex_t                *r_pattern;
+fastgrep_t     *fg_pattern;
+
+/* Filename exclusion/inclusion patterns */
+unsigned int    fpatterns, fpattern_sz;
+unsigned int    dpatterns, dpattern_sz;
+struct epat    *dpattern, *fpattern;
+
+/* For regex errors  */
+char    re_error[RE_ERROR_BUF + 1];
+
+/* Command-line flags */
+unsigned long long Aflag;      /* -A x: print x lines trailing each match */
+unsigned long long Bflag;      /* -B x: print x lines leading each match */
+bool    Hflag;         /* -H: always print file name */
+bool    Lflag;         /* -L: only show names of files with no matches */
+bool    bflag;         /* -b: show block numbers for each match */
+bool    cflag;         /* -c: only show a count of matching lines */
+bool    hflag;         /* -h: don't print filename headers */
+bool    iflag;         /* -i: ignore case */
+bool    lflag;         /* -l: only show names of files with matches */
+bool    mflag;         /* -m x: stop reading the files after x matches */
+unsigned long long mcount;     /* count for -m */
+bool    nflag;         /* -n: show line numbers in front of matching lines */
+bool    oflag;         /* -o: print only matching part */
+bool    qflag;         /* -q: quiet mode (don't output anything) */
+bool    sflag;         /* -s: silent mode (ignore errors) */
+bool    vflag;         /* -v: only show non-matching lines */
+bool    wflag;         /* -w: pattern must start and end on word boundaries */
+bool    xflag;         /* -x: pattern must match entire line */
+bool    lbflag;        /* --line-buffered */
+bool    nullflag;      /* --null */
+bool    nulldataflag;  /* --null-data */
+unsigned char line_sep = '\n'; /* 0 for --null-data */
+char   *label;         /* --label */
+const char *color;     /* --color */
+int     grepbehave = GREP_BASIC;       /* -EFGP: type of the regex */
+int     binbehave = BINFILE_BIN;       /* -aIU: handling of binary files */
+int     filebehave = FILE_STDIO;       /* -JZ: normal, gzip or bzip2 file */
+int     devbehave = DEV_READ;          /* -D: handling of devices */
+int     dirbehave = DIR_READ;          /* -dRr: handling of directories */
+int     linkbehave = LINK_READ;        /* -OpS: handling of symlinks */
+
+bool    dexclude, dinclude;    /* --exclude-dir and --include-dir */
+bool    fexclude, finclude;    /* --exclude and --include */
+
+enum {
+       BIN_OPT = CHAR_MAX + 1,
+       COLOR_OPT,
+       DECOMPRESS_OPT,
+       HELP_OPT,
+       MMAP_OPT,
+       LINEBUF_OPT,
+       LABEL_OPT,
+       R_EXCLUDE_OPT,
+       R_INCLUDE_OPT,
+       R_DEXCLUDE_OPT,
+       R_DINCLUDE_OPT
+};
+
+static inline const char       *init_color(const char *);
+
+/* Housekeeping */
+int     tail;          /* lines left to print */
+bool    notfound;      /* file not found */
+
+extern char    *__progname;
+
+/*
+ * Prints usage information and returns 2.
+ */
+__dead static void
+usage(void)
+{
+       fprintf(stderr, getstr(4), __progname);
+       fprintf(stderr, "%s", getstr(5));
+       fprintf(stderr, "%s", getstr(5));
+       fprintf(stderr, "%s", getstr(6));
+       fprintf(stderr, "%s", getstr(7));
+       exit(2);
+}
+
+static const char optstr[] =
+    "0123456789A:B:C:D:EFGHIJLOPSRUVZabcd:e:f:hilm:nopqrsuvwxyz";
+
+struct option long_options[] =
+{
+       {"binary-files",        required_argument,      NULL, BIN_OPT},
+       {"decompress",          no_argument,            NULL, DECOMPRESS_OPT},
+       {"help",                no_argument,            NULL, HELP_OPT},
+       {"mmap",                no_argument,            NULL, MMAP_OPT},
+       {"line-buffered",       no_argument,            NULL, LINEBUF_OPT},
+       {"label",               required_argument,      NULL, LABEL_OPT},
+       {"color",               optional_argument,      NULL, COLOR_OPT},
+       {"colour",              optional_argument,      NULL, COLOR_OPT},
+       {"exclude",             required_argument,      NULL, R_EXCLUDE_OPT},
+       {"include",             required_argument,      NULL, R_INCLUDE_OPT},
+       {"exclude-dir",         required_argument,      NULL, R_DEXCLUDE_OPT},
+       {"include-dir",         required_argument,      NULL, R_DINCLUDE_OPT},
+       {"after-context",       required_argument,      NULL, 'A'},
+       {"text",                no_argument,            NULL, 'a'},
+       {"before-context",      required_argument,      NULL, 'B'},
+       {"byte-offset",         no_argument,            NULL, 'b'},
+       {"context",             optional_argument,      NULL, 'C'},
+       {"count",               no_argument,            NULL, 'c'},
+       {"devices",             required_argument,      NULL, 'D'},
+        {"directories",                required_argument,      NULL, 'd'},
+       {"extended-regexp",     no_argument,            NULL, 'E'},
+       {"regexp",              required_argument,      NULL, 'e'},
+       {"fixed-strings",       no_argument,            NULL, 'F'},
+       {"file",                required_argument,      NULL, 'f'},
+       {"basic-regexp",        no_argument,            NULL, 'G'},
+       {"no-filename",         no_argument,            NULL, 'h'},
+       {"with-filename",       no_argument,            NULL, 'H'},
+       {"ignore-case",         no_argument,            NULL, 'i'},
+       {"bz2decompress",       no_argument,            NULL, 'J'},
+       {"files-with-matches",  no_argument,            NULL, 'l'},
+       {"files-without-match", no_argument,            NULL, 'L'},
+       {"max-count",           required_argument,      NULL, 'm'},
+       {"line-number",         no_argument,            NULL, 'n'},
+       {"only-matching",       no_argument,            NULL, 'o'},
+       {"quiet",               no_argument,            NULL, 'q'},
+       {"silent",              no_argument,            NULL, 'q'},
+       {"recursive",           no_argument,            NULL, 'r'},
+       {"no-messages",         no_argument,            NULL, 's'},
+       {"binary",              no_argument,            NULL, 'U'},
+       {"unix-byte-offsets",   no_argument,            NULL, 'u'},
+       {"invert-match",        no_argument,            NULL, 'v'},
+       {"version",             no_argument,            NULL, 'V'},
+       {"word-regexp",         no_argument,            NULL, 'w'},
+       {"line-regexp",         no_argument,            NULL, 'x'},
+       {"null",                no_argument,            NULL, 'Z'},
+       {"null-data",           no_argument,            NULL, 'z'},
+       {NULL,                  no_argument,            NULL, 0}
+};
+
+/*
+ * Adds a searching pattern to the internal array.
+ */
+static void
+add_pattern(char *pat, size_t len)
+{
+
+       /* TODO: Check for empty patterns and shortcut */
+
+       /* Increase size if necessary */
+       if (patterns == pattern_sz) {
+               pattern_sz *= 2;
+               pattern = grep_realloc(pattern, ++pattern_sz *
+                   sizeof(*pattern));
+       }
+       if (len > 0 && pat[len - 1] == '\n')
+               --len;
+       /* pat may not be NUL-terminated */
+       pattern[patterns] = grep_malloc(len + 1);
+       memcpy(pattern[patterns], pat, len);
+       pattern[patterns][len] = '\0';
+       ++patterns;
+}
+
+/*
+ * Adds a file include/exclude pattern to the internal array.
+ */
+static void
+add_fpattern(const char *pat, int mode)
+{
+
+       /* Increase size if necessary */
+       if (fpatterns == fpattern_sz) {
+               fpattern_sz *= 2;
+               fpattern = grep_realloc(fpattern, ++fpattern_sz *
+                   sizeof(struct epat));
+       }
+       fpattern[fpatterns].pat = grep_strdup(pat);
+       fpattern[fpatterns].mode = mode;
+       ++fpatterns;
+}
+
+/*
+ * Adds a directory include/exclude pattern to the internal array.
+ */
+static void
+add_dpattern(const char *pat, int mode)
+{
+
+       /* Increase size if necessary */
+       if (dpatterns == dpattern_sz) {
+               dpattern_sz *= 2;
+               dpattern = grep_realloc(dpattern, ++dpattern_sz *
+                   sizeof(struct epat));
+       }
+       dpattern[dpatterns].pat = grep_strdup(pat);
+       dpattern[dpatterns].mode = mode;
+       ++dpatterns;
+}
+
+/*
+ * Reads searching patterns from a file and adds them with add_pattern().
+ */
+static void
+read_patterns(const char *fn)
+{
+       FILE *f;
+       char *line;
+       size_t len;
+       ssize_t rlen;
+
+       if ((f = fopen(fn, "r")) == NULL)
+               err(2, "%s", fn);
+       line = NULL;
+       len = 0;
+       while ((rlen = getline(&line, &len, f)) != -1)
+               add_pattern(line, *line == '\n' ? 0 : (size_t)rlen);
+       free(line);
+       if (ferror(f))
+               err(2, "%s", fn);
+       fclose(f);
+}
+
+static inline const char *
+init_color(const char *d)
+{
+       char *c;
+
+       c = getenv("GREP_COLOR");
+       return (c != NULL ? c : d);
+}
+
+int
+main(int argc, char *argv[])
+{
+       char **aargv, **eargv, *eopts;
+       char *ep;
+       unsigned long long l;
+       unsigned int aargc, eargc, i, j;
+       int c, lastc, needpattern, newarg, prevoptind;
+
+       setlocale(LC_ALL, "");
+
+#ifndef WITHOUT_NLS
+       catalog = catopen("grep", NL_CAT_LOCALE);
+#endif
+
+       /* Check what is the program name of the binary.  In this
+          way we can have all the funcionalities in one binary
+          without the need of scripting and using ugly hacks. */
+       switch (__progname[0]) {
+       case 'e':
+               grepbehave = GREP_EXTENDED;
+               break;
+       case 'f':
+               grepbehave = GREP_FIXED;
+               break;
+       case 'g':
+               grepbehave = GREP_BASIC;
+               break;
+       case 'z':
+               filebehave = FILE_GZIP;
+               switch(__progname[1]) {
+               case 'e':
+                       grepbehave = GREP_EXTENDED;
+                       break;
+               case 'f':
+                       grepbehave = GREP_FIXED;
+                       break;
+               case 'g':
+                       grepbehave = GREP_BASIC;
+                       break;
+               }
+               break;
+       }
+
+       lastc = '\0';
+       newarg = 1;
+       prevoptind = 1;
+       needpattern = 1;
+
+       eopts = getenv("GREP_OPTIONS");
+
+       /* support for extra arguments in GREP_OPTIONS */
+       eargc = 0;
+       if (eopts != NULL) {
+               char *str;
+
+               /* make an estimation of how many extra arguments we have */
+               for (j = 0; j < strlen(eopts); j++)
+                       if (eopts[j] == ' ')
+                               eargc++;
+
+               eargv = (char **)grep_malloc(sizeof(char *) * (eargc + 1));
+
+               eargc = 0;
+               /* parse extra arguments */
+               while ((str = strsep(&eopts, " ")) != NULL)
+                       eargv[eargc++] = grep_strdup(str);
+
+               aargv = (char **)grep_calloc(eargc + argc + 1,
+                   sizeof(char *));
+
+               aargv[0] = argv[0];
+               for (i = 0; i < eargc; i++)
+                       aargv[i + 1] = eargv[i];
+               for (j = 1; j < (unsigned int)argc; j++, i++)
+                       aargv[i + 1] = argv[j];
+
+               aargc = eargc + argc;
+       } else {
+               aargv = argv;
+               aargc = argc;
+       }
+
+       while (((c = getopt_long(aargc, aargv, optstr, long_options, NULL)) !=
+           -1)) {
+               switch (c) {
+               case '0': case '1': case '2': case '3': case '4':
+               case '5': case '6': case '7': case '8': case '9':
+                       if (newarg || !isdigit(lastc))
+                               Aflag = 0;
+                       else if (Aflag > LLONG_MAX / 10) {
+                               errno = ERANGE;
+                               err(2, NULL);
+                       }
+                       Aflag = Bflag = (Aflag * 10) + (c - '0');
+                       break;
+               case 'C':
+                       if (optarg == NULL) {
+                               Aflag = Bflag = 2;
+                               break;
+                       }
+                       /* FALLTHROUGH */
+               case 'A':
+                       /* FALLTHROUGH */
+               case 'B':
+                       errno = 0;
+                       l = strtoull(optarg, &ep, 10);
+                       if (((errno == ERANGE) && (l == ULLONG_MAX)) ||
+                           ((errno == EINVAL) && (l == 0)))
+                               err(2, NULL);
+                       else if (ep[0] != '\0') {
+                               errno = EINVAL;
+                               err(2, NULL);
+                       }
+                       if (c == 'A')
+                               Aflag = l;
+                       else if (c == 'B')
+                               Bflag = l;
+                       else
+                               Aflag = Bflag = l;
+                       break;
+               case 'a':
+                       binbehave = BINFILE_TEXT;
+                       break;
+               case 'b':
+                       bflag = true;
+                       break;
+               case 'c':
+                       cflag = true;
+                       break;
+               case 'D':
+                       if (strcasecmp(optarg, "skip") == 0)
+                               devbehave = DEV_SKIP;
+                       else if (strcasecmp(optarg, "read") == 0)
+                               devbehave = DEV_READ;
+                       else
+                               errx(2, getstr(3), "--devices");
+                       break;
+               case 'd':
+                       if (strcasecmp("recurse", optarg) == 0) {
+                               Hflag = true;
+                               dirbehave = DIR_RECURSE;
+                       } else if (strcasecmp("skip", optarg) == 0)
+                               dirbehave = DIR_SKIP;
+                       else if (strcasecmp("read", optarg) == 0)
+                               dirbehave = DIR_READ;
+                       else
+                               errx(2, getstr(3), "--directories");
+                       break;
+               case 'E':
+                       grepbehave = GREP_EXTENDED;
+                       break;
+               case 'e':
+                       add_pattern(optarg, strlen(optarg));
+                       needpattern = 0;
+                       break;
+               case 'F':
+                       grepbehave = GREP_FIXED;
+                       break;
+               case 'f':
+                       read_patterns(optarg);
+                       needpattern = 0;
+                       break;
+               case 'G':
+                       grepbehave = GREP_BASIC;
+                       break;
+               case 'H':
+                       Hflag = true;
+                       break;
+               case 'h':
+                       Hflag = false;
+                       hflag = true;
+                       break;
+               case 'I':
+                       binbehave = BINFILE_SKIP;
+                       break;
+               case 'i':
+               case 'y':
+                       iflag =  true;
+                       cflags |= REG_ICASE;
+                       break;
+               case 'J':
+                       filebehave = FILE_BZIP;
+                       break;
+               case 'L':
+                       lflag = false;
+                       Lflag = true;
+                       break;
+               case 'l':
+                       Lflag = false;
+                       lflag = true;
+                       break;
+               case 'm':
+                       mflag = true;
+                       errno = 0;
+                       mcount = strtoull(optarg, &ep, 10);
+                       if (((errno == ERANGE) && (mcount == ULLONG_MAX)) ||
+                           ((errno == EINVAL) && (mcount == 0)))
+                               err(2, NULL);
+                       else if (ep[0] != '\0') {
+                               errno = EINVAL;
+                               err(2, NULL);
+                       }
+                       break;
+               case 'n':
+                       nflag = true;
+                       break;
+               case 'O':
+                       linkbehave = LINK_EXPLICIT;
+                       break;
+               case 'o':
+                       oflag = true;
+                       break;
+               case 'p':
+                       linkbehave = LINK_SKIP;
+                       break;
+               case 'q':
+                       qflag = true;
+                       break;
+               case 'S':
+                       linkbehave = LINK_READ;
+                       break;
+               case 'R':
+               case 'r':
+                       dirbehave = DIR_RECURSE;
+                       Hflag = true;
+                       break;
+               case 's':
+                       sflag = true;
+                       break;
+               case 'U':
+                       binbehave = BINFILE_BIN;
+                       break;
+               case 'u':
+               case MMAP_OPT:
+                       /* noop, compatibility */
+                       break;
+               case 'V':
+                       printf(getstr(9), __progname, VERSION);
+                       exit(0);
+               case 'v':
+                       vflag = true;
+                       break;
+               case 'w':
+                       wflag = true;
+                       break;
+               case 'x':
+                       xflag = true;
+                       break;
+               case 'Z':
+                       nullflag = true;
+                       break;
+               case 'z':
+                       nulldataflag = true;
+                       line_sep = '\0';
+                       break;
+               case BIN_OPT:
+                       if (strcasecmp("binary", optarg) == 0)
+                               binbehave = BINFILE_BIN;
+                       else if (strcasecmp("without-match", optarg) == 0)
+                               binbehave = BINFILE_SKIP;
+                       else if (strcasecmp("text", optarg) == 0)
+                               binbehave = BINFILE_TEXT;
+                       else
+                               errx(2, getstr(3), "--binary-files");
+                       break;
+               case COLOR_OPT:
+                       color = NULL;
+                       if (optarg == NULL || strcasecmp("auto", optarg) == 0 ||
+                           strcasecmp("tty", optarg) == 0 ||
+                           strcasecmp("if-tty", optarg) == 0) {
+                               char *term;
+
+                               term = getenv("TERM");
+                               if (isatty(STDOUT_FILENO) && term != NULL &&
+                                   strcasecmp(term, "dumb") != 0)
+                                       color = init_color("01;31");
+                       } else if (strcasecmp("always", optarg) == 0 ||
+                           strcasecmp("yes", optarg) == 0 ||
+                           strcasecmp("force", optarg) == 0) {
+                               color = init_color("01;31");
+                       } else if (strcasecmp("never", optarg) != 0 &&
+                           strcasecmp("none", optarg) != 0 &&
+                           strcasecmp("no", optarg) != 0)
+                               errx(2, getstr(3), "--color");
+                       break;
+               case DECOMPRESS_OPT:
+                       filebehave = FILE_GZIP;
+                       break;
+               case LABEL_OPT:
+                       label = optarg;
+                       break;
+               case LINEBUF_OPT:
+                       lbflag = true;
+                       break;
+               case R_INCLUDE_OPT:
+                       finclude = true;
+                       add_fpattern(optarg, INCL_PAT);
+                       break;
+               case R_EXCLUDE_OPT:
+                       fexclude = true;
+                       add_fpattern(optarg, EXCL_PAT);
+                       break;
+               case R_DINCLUDE_OPT:
+                       dinclude = true;
+                       add_dpattern(optarg, INCL_PAT);
+                       break;
+               case R_DEXCLUDE_OPT:
+                       dexclude = true;
+                       add_dpattern(optarg, EXCL_PAT);
+                       break;
+               case HELP_OPT:
+               default:
+                       usage();
+               }
+               lastc = c;
+               newarg = optind != prevoptind;
+               prevoptind = optind;
+       }
+       aargc -= optind;
+       aargv += optind;
+
+       /* Fail if we don't have any pattern */
+       if (aargc == 0 && needpattern)
+               usage();
+
+       /* Process patterns from command line */
+       if (aargc != 0 && needpattern) {
+               add_pattern(*aargv, strlen(*aargv));
+               --aargc;
+               ++aargv;
+       }
+
+       switch (grepbehave) {
+       case GREP_FIXED:
+       case GREP_BASIC:
+               break;
+       case GREP_EXTENDED:
+               cflags |= REG_EXTENDED;
+               break;
+       default:
+               /* NOTREACHED */
+               usage();
+       }
+
+       fg_pattern = grep_calloc(patterns, sizeof(*fg_pattern));
+       r_pattern = grep_calloc(patterns, sizeof(*r_pattern));
+/*
+ * XXX: fgrepcomp() and fastcomp() are workarounds for regexec() performance.
+ * Optimizations should be done there.
+ */
+               /* Check if cheating is allowed (always is for fgrep). */
+       if (grepbehave == GREP_FIXED) {
+               for (i = 0; i < patterns; ++i)
+                       fgrepcomp(&fg_pattern[i], pattern[i]);
+       } else {
+               for (i = 0; i < patterns; ++i) {
+                       if (fastcomp(&fg_pattern[i], pattern[i])) {
+                               /* Fall back to full regex library */
+                               c = regcomp(&r_pattern[i], pattern[i], cflags);
+                               if (c != 0) {
+                                       regerror(c, &r_pattern[i], re_error,
+                                           RE_ERROR_BUF);
+                                       errx(2, "%s", re_error);
+                               }
+                       }
+               }
+       }
+
+       if (lbflag)
+               setlinebuf(stdout);
+
+       if ((aargc == 0 || aargc == 1) && !Hflag)
+               hflag = true;
+
+       if (aargc == 0)
+               exit(!procfile("-"));
+
+       if (dirbehave == DIR_RECURSE)
+               c = grep_tree(aargv);
+       else
+               for (c = 0; aargc--; ++aargv) {
+                       if ((finclude || fexclude) && !file_matching(*aargv))
+                               continue;
+                       c+= procfile(*aargv);
+               }
+
+#ifndef WITHOUT_NLS
+       catclose(catalog);
+#endif
+
+       /* Find out the correct return value according to the
+          results and the command line option. */
+       exit(c ? (notfound ? (qflag ? 0 : 2) : 0) : (notfound ? 2 : 1));
+}
diff --git a/usr.bin/grep/grep.h b/usr.bin/grep/grep.h
new file mode 100644 (file)
index 0000000..617edac
--- /dev/null
@@ -0,0 +1,158 @@
+/*     $NetBSD: grep.h,v 1.8 2012/05/06 22:27:00 joerg Exp $   */
+/*     $OpenBSD: grep.h,v 1.15 2010/04/05 03:03:55 tedu Exp $  */
+/*     $FreeBSD: head/usr.bin/grep/grep.h 211496 2010-08-19 09:28:59Z des $    */
+
+/*-
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (c) 2008-2009 Gabor Kovesdan <gabor@FreeBSD.org>
+ * 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.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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 <bzlib.h>
+#include <limits.h>
+#include <regex.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <zlib.h>
+
+#ifdef WITHOUT_NLS
+#define getstr(n)       errstr[n]
+#else
+#include <nl_types.h>
+
+extern nl_catd          catalog;
+#define getstr(n)       catgets(catalog, 1, n, errstr[n])
+#endif
+
+extern const char              *errstr[];
+
+#define VERSION                "2.5.1-FreeBSD"
+
+#define GREP_FIXED     0
+#define GREP_BASIC     1
+#define GREP_EXTENDED  2
+
+#define BINFILE_BIN    0
+#define BINFILE_SKIP   1
+#define BINFILE_TEXT   2
+
+#define FILE_STDIO     0
+#define FILE_GZIP      1
+#define FILE_BZIP      2
+
+#define DIR_READ       0
+#define DIR_SKIP       1
+#define DIR_RECURSE    2
+
+#define DEV_READ       0
+#define DEV_SKIP       1
+
+#define LINK_READ      0
+#define LINK_EXPLICIT  1
+#define LINK_SKIP      2
+
+#define EXCL_PAT       0
+#define INCL_PAT       1
+
+#define MAX_LINE_MATCHES       32
+
+struct file {
+       int              fd;
+       bool             binary;
+};
+
+struct str {
+       off_t            off;
+       size_t           len;
+       char            *dat;
+       char            *file;
+       int              line_no;
+};
+
+struct epat {
+       char            *pat;
+       int              mode;
+};
+
+typedef struct {
+       size_t           len;
+       unsigned char   *pattern;
+       int              qsBc[UCHAR_MAX + 1];
+       /* flags */
+       bool             bol;
+       bool             eol;
+       bool             reversed;
+       bool             word;
+} fastgrep_t;
+
+/* Flags passed to regcomp() and regexec() */
+extern int      cflags, eflags;
+
+/* Command line flags */
+extern bool     Eflag, Fflag, Gflag, Hflag, Lflag,
+                bflag, cflag, hflag, iflag, lflag, mflag, nflag, oflag,
+                qflag, sflag, vflag, wflag, xflag;
+extern bool     dexclude, dinclude, fexclude, finclude, lbflag, nullflag, nulldataflag;
+extern unsigned char line_sep;
+extern unsigned long long Aflag, Bflag, mcount;
+extern char    *label;
+extern const char *color;
+extern int      binbehave, devbehave, dirbehave, filebehave, grepbehave, linkbehave;
+
+extern bool     notfound;
+extern int      tail;
+extern unsigned int dpatterns, fpatterns, patterns;
+extern char    **pattern;
+extern struct epat *dpattern, *fpattern;
+extern regex_t *er_pattern, *r_pattern;
+extern fastgrep_t *fg_pattern;
+
+/* For regex errors  */
+#define RE_ERROR_BUF   512
+extern char     re_error[RE_ERROR_BUF + 1];    /* Seems big enough */
+
+/* util.c */
+bool    file_matching(const char *fname);
+int     procfile(const char *fn);
+int     grep_tree(char **argv);
+void   *grep_malloc(size_t size);
+void   *grep_calloc(size_t nmemb, size_t size);
+void   *grep_realloc(void *ptr, size_t size);
+char   *grep_strdup(const char *str);
+void    printline(struct str *line, int sep, regmatch_t *matches, int m);
+
+/* queue.c */
+void    enqueue(struct str *x);
+void    printqueue(void);
+void    clearqueue(void);
+
+/* file.c */
+void            grep_close(struct file *f);
+struct file    *grep_open(const char *path);
+char           *grep_fgetln(struct file *f, size_t *len);
+
+/* fastgrep.c */
+int             fastcomp(fastgrep_t *, const char *);
+void            fgrepcomp(fastgrep_t *, const char *);
+int             grep_search(fastgrep_t *, const unsigned char *, size_t, regmatch_t *);
diff --git a/usr.bin/grep/nls/C.msg b/usr.bin/grep/nls/C.msg
new file mode 100644 (file)
index 0000000..d07aea6
--- /dev/null
@@ -0,0 +1,13 @@
+$ $FreeBSD: head/usr.bin/grep/nls/C.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(standard input)"
+2 "cannot read bzip2 compressed file"
+3 "unknown %s option"
+4 "usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n"
+5 "\t[-e pattern] [-f file] [--binary-files=value] [--color=when]\n"
+6 "\t[--context[=num]] [--directories=action] [--label] [--line-buffered]\n"
+7 "\t[pattern] [file ...]\n"
+8 "Binary file %s matches\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/es_ES.ISO8859-1.msg b/usr.bin/grep/nls/es_ES.ISO8859-1.msg
new file mode 100644 (file)
index 0000000..441ba01
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: es_ES.ISO8859-1.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/es_ES.ISO8859-1.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(entrada estándar)"
+2 "no se puede leer el fichero comprimido bzip2"
+3 "opción desconocida de %s"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A no] [-B no] [-C[no]]\n"
+5 "\t[-e pauta] [-f fichero] [--binary-files=valor] [--color=cuando]\n"
+6 "\t[--context[=no]] [--directories=acción] [--label] [--line-buffered]\n"
+7 "\t[pauta] [fichero ...]\n"
+8 "fichero binario %s se ajusta\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/gl_ES.ISO8859-1.msg b/usr.bin/grep/nls/gl_ES.ISO8859-1.msg
new file mode 100644 (file)
index 0000000..ab50cea
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: gl_ES.ISO8859-1.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/gl_ES.ISO8859-1.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(entrada estándar)"
+2 "non se pode ler o ficheiro comprimido bzip2"
+3 "opción descoñecida de %s"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A no] [-B no] [-C[no]]\n"
+5 "\t[-e pauta] [-f ficheiro] [--binary-files=valor] [--color=cando]\n"
+6 "\t[--context[=no]] [--directories=acción] [--label] [--line-buffered]\n"
+7 "\t[pauta] [ficheiro ...]\n"
+8 "ficheiro binario %s conforma\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/hu_HU.ISO8859-2.msg b/usr.bin/grep/nls/hu_HU.ISO8859-2.msg
new file mode 100644 (file)
index 0000000..69dc2ec
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: hu_HU.ISO8859-2.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/hu_HU.ISO8859-2.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(szabványos bemenet)"
+2 "bzip2 tömörített fájl nem olvasható"
+3 "ismeretlen %s opció"
+4 "használat: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A szám] [-B szám] [-C[szám]]\n"
+5 "\t[-e minta] [-f fájl] [--binary-files=érték] [--color=mikor]\n"
+6 "\t[--context[=szám]] [--directories=mûvelet] [--label] [--line-buffered]\n"
+7 "\t[minta] [fájl ...]\n"
+8 "%s bináris fájl illeszkedik\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/ja_JP.SJIS.msg b/usr.bin/grep/nls/ja_JP.SJIS.msg
new file mode 100644 (file)
index 0000000..99357b6
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: ja_JP.SJIS.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/ja_JP.SJIS.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(\95W\8f\80\93ü\97Í)"
+2 "bzip2 \88³\8fk\83t\83@\83C\83\8b\82ð\93Ç\82Ý\8d\9e\82Þ\82±\82Æ\82ª\82Å\82«\82Ü\82¹\82ñ"
+3 "%s \83I\83v\83V\83\87\83\93\82Ì\8ew\92è\92l\82É\8cë\82è\82ª\82 \82è\82Ü\82·"
+4 "\8eg\82¢\95û: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A \90\94\8e\9a] [-B \90\94\8e\9a] [-C[\90\94\8e\9a]]\n"
+5 "\t[-e \83p\83^\81[\83\93] [-f \83t\83@\83C\83\8b\96¼] [--binary-files=\92l] [--color=\92l]\n"
+6 "\t[--context[=\90\94\8e\9a]] [--directories=\93®\8dì] [--label] [--line-buffered]\n"
+7 "\t[\83p\83^\81[\83\93] [\83t\83@\83C\83\8b\96¼ ...]\n"
+8 "\83o\83C\83i\83\8a\83t\83@\83C\83\8b %s \82É\83}\83b\83`\82µ\82Ü\82µ\82½\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/ja_JP.UTF-8.msg b/usr.bin/grep/nls/ja_JP.UTF-8.msg
new file mode 100644 (file)
index 0000000..3d4aed3
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: ja_JP.UTF-8.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/ja_JP.UTF-8.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(標準入力)"
+2 "bzip2 圧縮ファイルを読み込むことができません"
+3 "%s オプションの指定値に誤りがあります"
+4 "使い方: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A 数字] [-B 数字] [-C[数字]]\n"
+5 "\t[-e パターン] [-f ファイル名] [--binary-files=値] [--color=値]\n"
+6 "\t[--context[=数字]] [--directories=動作] [--label] [--line-buffered]\n"
+7 "\t[パターン] [ファイル名 ...]\n"
+8 "バイナリファイル %s にマッチしました\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/ja_JP.eucJP.msg b/usr.bin/grep/nls/ja_JP.eucJP.msg
new file mode 100644 (file)
index 0000000..08e0e7d
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: ja_JP.eucJP.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/ja_JP.eucJP.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(ɸ½àÆþÎÏ)"
+2 "bzip2 °µ½Ì¥Õ¥¡¥¤¥ë¤òÆɤ߹þ¤à¤³¤È¤¬¤Ç¤­¤Þ¤»¤ó"
+3 "%s ¥ª¥×¥·¥ç¥ó¤Î»ØÄêÃͤ˸í¤ê¤¬¤¢¤ê¤Þ¤¹"
+4 "»È¤¤Êý: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A ¿ô»ú] [-B ¿ô»ú] [-C[¿ô»ú]]\n"
+5 "\t[-e ¥Ñ¥¿¡¼¥ó] [-f ¥Õ¥¡¥¤¥ë̾] [--binary-files=ÃÍ] [--color=ÃÍ]\n"
+6 "\t[--context[=¿ô»ú]] [--directories=Æ°ºî] [--label] [--line-buffered]\n"
+7 "\t[¥Ñ¥¿¡¼¥ó] [¥Õ¥¡¥¤¥ë̾ ...]\n"
+8 "¥Ð¥¤¥Ê¥ê¥Õ¥¡¥¤¥ë %s ¤Ë¥Þ¥Ã¥Á¤·¤Þ¤·¤¿\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/pt_BR.ISO8859-1.msg b/usr.bin/grep/nls/pt_BR.ISO8859-1.msg
new file mode 100644 (file)
index 0000000..7b6c20e
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: pt_BR.ISO8859-1.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/pt_BR.ISO8859-1.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(entrada padrão)"
+2 "não se posso ler o fichero comprimido bzip2"
+3 "opcão não conhecida de %s"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n"
+5 "\t[-e padrão] [-f arquivo] [--binary-files=valor] [--color=quando]\n"
+6 "\t[--context[=num]] [--directories=ação] [--label] [--line-buffered]\n"
+7 "\t[padrão] [arquivo ...]\n"
+8 "arquivo binário %s casa com o padrão\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/ru_RU.KOI8-R.msg b/usr.bin/grep/nls/ru_RU.KOI8-R.msg
new file mode 100644 (file)
index 0000000..4c69f73
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: ru_RU.KOI8-R.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/ru_RU.KOI8-R.msg 210622 2010-07-29 18:02:57Z gabor $
+$
+$set 1
+$quote "
+1 "(ÓÔÁÎÄÁÒÔÎÙÊ ××ÏÄ)"
+2 "ÎÅ ÍÏÇÕ ÐÒÏÞÉÔÁÔØ ÓÖÁÔÙÊ × bzip2 ÆÁÊÌ"
+3 "ÎÅÉÚ×ÅÓÔÎÙÊ ËÌÀÞ %s"
+4 "ÉÓÐÏÌØÚÏ×ÁÎÉÅ: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A ÞÉÓ] [-B ÞÉÓ] [-C[ÞÉÓ]]\n"
+5 "\t[-e ÛÁÂÌÏÎ] [-f ÆÁÊÌ] [--binary-files=ÚÎÁÞÅÎÉÅ] [--color=ËÏÇÄÁ]\n"
+6 "\t[--context[=ÞÉÓ]] [--directories=ÄÅÊÓÔ×ÉÅ] [--label] [--line-buffered]\n"
+7 "\t[ÛÁÂÌÏÎ] [ÆÁÊÌ ...]\n"
+8 "Ä×ÏÉÞÎÙÊ ÆÁÊÌ %s ÓÏ×ÐÁÄÁÅÔ\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/uk_UA.UTF-8.msg b/usr.bin/grep/nls/uk_UA.UTF-8.msg
new file mode 100644 (file)
index 0000000..e6b9f54
--- /dev/null
@@ -0,0 +1,13 @@
+$ $NetBSD: uk_UA.UTF-8.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/uk_UA.UTF-8.msg 210927 2010-08-06 10:34:48Z gabor $
+$set 1
+$quote "
+1 "(стандартний ввід)"
+2 "не можу прочитати стиснутий bzip2 файл"
+3 "невiдома опція %s"
+4 "використання: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A чис] [-B чис] [-C[чис]]\n"
+5 "\t[-e шаблон] [-f файл] [--binary-files=значення] [--color=коли]\n"
+6 "\t[--context[=чис] [--directories=дія] [--label] [--line-buffered]\n"
+7 "\t[шаблон] [файл ...]\n"
+8 "двійковий файл %s співпадає\n"
+9 "%s (BSD grep) %s\n"
diff --git a/usr.bin/grep/nls/zh_CN.UTF-8.msg b/usr.bin/grep/nls/zh_CN.UTF-8.msg
new file mode 100644 (file)
index 0000000..e9d40cd
--- /dev/null
@@ -0,0 +1,14 @@
+$ $NetBSD: zh_CN.UTF-8.msg,v 1.2 2011/04/18 22:46:48 joerg Exp $
+$ $FreeBSD: head/usr.bin/grep/nls/zh_CN.UTF-8.msg 212927 2010-09-20 19:42:52Z delphij $
+$
+$set 1
+$quote "
+1 "(标准输入)"
+2 "读取 bzip2 压缩文件时出错"
+3 "选项 %s 无法识别"
+4 "用法: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A 行数] [-B 行数] [-C[行数]]\n"
+5 "\t[-e 模式] [-f 文件] [--binary-files=值] [--color=何时]\n"
+6 "\t[--context[=行数]] [--directories=动作] [--label] [--line-buffered]\n"
+7 "\t[模式] [文件名 ...]\n"
+8 "二进制文件 %s 包含模式\n"
+9 "%s (BSD grep) %s\n"
similarity index 58%
rename from minix/commands/grep/queue.c
rename to usr.bin/grep/queue.c
index 3a18d38d8fbb34a59a9e1d85996bf658234d4ab1..e3c6be17a89a6d7db3e5c19190678778f075318f 100644 (file)
@@ -1,7 +1,7 @@
-/*     $OpenBSD: queue.c,v 1.4 2003/06/25 05:31:11 deraadt Exp $       */
-
+/*     $NetBSD: queue.c,v 1.5 2011/08/31 16:24:57 plunky Exp $ */
+/*     $FreeBSD: head/usr.bin/grep/queue.c 211496 2010-08-19 09:28:59Z des $   */
 /*-
- * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
 
 /*
  * A really poor man's queue.  It does only what it has to and gets out of
- * Dodge.
+ * Dodge.  It is used in place of <sys/queue.h> to get a better performance.
  */
 
+#if HAVE_NBTOOL_CONFIG_H
+#include "nbtool_config.h"
+#endif
+
+#include <sys/cdefs.h>
+__RCSID("$NetBSD: queue.c,v 1.5 2011/08/31 16:24:57 plunky Exp $");
+
 #include <sys/param.h>
+#include <sys/queue.h>
 
 #include <stdlib.h>
 #include <string.h>
 
 #include "grep.h"
 
-typedef struct queue {
-       struct queue   *next;
-       str_t           data;
-} queue_t;
-
-static queue_t *q_head, *q_tail;
-static int      count;
+struct qentry {
+       STAILQ_ENTRY(qentry)    list;
+       struct str              data;
+};
 
-static queue_t *dequeue(void);
-
-void
-initqueue(void)
-{
-       q_head = q_tail = NULL;
-}
+static STAILQ_HEAD(, qentry)   queue = STAILQ_HEAD_INITIALIZER(queue);
+static unsigned long long      count;
 
-static void
-free_item(queue_t *item)
-{
-       free(item);
-}
+static struct qentry   *dequeue(void);
 
 void
-enqueue(str_t *x)
+enqueue(struct str *x)
 {
-       queue_t *item;
+       struct qentry *item;
 
-       item = grep_malloc(sizeof *item + x->len);
+       item = grep_malloc(sizeof(struct qentry));
+       item->data.dat = grep_malloc(sizeof(char) * x->len);
        item->data.len = x->len;
        item->data.line_no = x->line_no;
        item->data.off = x->off;
-       item->data.dat = (char *)item + sizeof *item;
        memcpy(item->data.dat, x->dat, x->len);
        item->data.file = x->file;
-       item->next = NULL;
 
-       if (!q_head) {
-               q_head = q_tail = item;
-       } else {
-               q_tail->next = item;
-               q_tail = item;
-       }
+       STAILQ_INSERT_TAIL(&queue, item, list);
 
-       if (++count > Bflag)
-               free_item(dequeue());
+       if (++count > Bflag) {
+               item = dequeue();
+               free(item->data.dat);
+               free(item);
+       }
 }
 
-static queue_t *
+static struct qentry *
 dequeue(void)
 {
-       queue_t *item;
+       struct qentry *item;
 
-       if (q_head == NULL)
-               return NULL;
+       item = STAILQ_FIRST(&queue);
+       if (item == NULL)
+               return (NULL);
 
+       STAILQ_REMOVE_HEAD(&queue, list);
        --count;
-       item = q_head;
-       q_head = item->next;
-       if (q_head == NULL)
-               q_tail = NULL;
-       return item;
+       return (item);
 }
 
 void
 printqueue(void)
 {
-       queue_t *item;
+       struct qentry *item;
 
        while ((item = dequeue()) != NULL) {
-               printline(&item->data, '-');
-               free_item(item);
+               printline(&item->data, '-', NULL, 0);
+               free(item->data.dat);
+               free(item);
        }
 }
 
 void
 clearqueue(void)
 {
-       queue_t *item;
+       struct qentry *item;
 
-       while ((item = dequeue()) != NULL)
-               free_item(item);
+       while ((item = dequeue()) != NULL) {
+               free(item->data.dat);
+               free(item);
+       }
 }
diff --git a/usr.bin/grep/util.c b/usr.bin/grep/util.c
new file mode 100644 (file)
index 0000000..c1c17e4
--- /dev/null
@@ -0,0 +1,499 @@
+/*     $NetBSD: util.c,v 1.17 2013/01/21 03:24:43 msaitoh Exp $        */
+/*     $FreeBSD: head/usr.bin/grep/util.c 211496 2010-08-19 09:28:59Z des $    */
+/*     $OpenBSD: util.c,v 1.39 2010/07/02 22:18:03 tedu Exp $  */
+
+/*-
+ * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
+ * Copyright (C) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org>
+ * 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.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
+ */
+
+#if HAVE_NBTOOL_CONFIG_H
+#include "nbtool_config.h"
+#endif
+
+#include <sys/cdefs.h>
+__RCSID("$NetBSD: util.c,v 1.17 2013/01/21 03:24:43 msaitoh Exp $");
+
+#include <sys/stat.h>
+#include <sys/types.h>
+
+#include <ctype.h>
+#include <err.h>
+#include <errno.h>
+#include <fnmatch.h>
+#include <fts.h>
+#include <libgen.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <wchar.h>
+#include <wctype.h>
+
+#include "grep.h"
+
+static bool     first, first_global = true;
+static unsigned long long since_printed;
+
+static int      procline(struct str *l, int);
+
+bool
+file_matching(const char *fname)
+{
+       char *fname_base, *fname_copy;
+       unsigned int i;
+       bool ret;
+
+       ret = finclude ? false : true;
+       fname_copy = grep_strdup(fname);
+       fname_base = basename(fname_copy);
+
+       for (i = 0; i < fpatterns; ++i) {
+               if (fnmatch(fpattern[i].pat, fname, 0) == 0 ||
+                   fnmatch(fpattern[i].pat, fname_base, 0) == 0) {
+                       if (fpattern[i].mode == EXCL_PAT) {
+                               free(fname_copy);
+                               return (false);
+                       } else
+                               ret = true;
+               }
+       }
+       free(fname_copy);
+       return (ret);
+}
+
+static inline bool
+dir_matching(const char *dname)
+{
+       unsigned int i;
+       bool ret;
+
+       ret = dinclude ? false : true;
+
+       for (i = 0; i < dpatterns; ++i) {
+               if (dname != NULL &&
+                   fnmatch(dname, dpattern[i].pat, 0) == 0) {
+                       if (dpattern[i].mode == EXCL_PAT)
+                               return (false);
+                       else
+                               ret = true;
+               }
+       }
+       return (ret);
+}
+
+/*
+ * Processes a directory when a recursive search is performed with
+ * the -R option.  Each appropriate file is passed to procfile().
+ */
+int
+grep_tree(char **argv)
+{
+       FTS *fts;
+       FTSENT *p;
+       char *d, *dir = NULL;
+       int c, fts_flags;
+       bool ok;
+
+       c = fts_flags = 0;
+
+       switch(linkbehave) {
+       case LINK_EXPLICIT:
+               fts_flags = FTS_COMFOLLOW;
+               break;
+       case LINK_SKIP:
+               fts_flags = FTS_PHYSICAL;
+               break;
+       default:
+               fts_flags = FTS_LOGICAL;
+                       
+       }
+
+       fts_flags |= FTS_NOSTAT | FTS_NOCHDIR;
+
+       if (!(fts = fts_open(argv, fts_flags, NULL)))
+               err(2, "fts_open");
+       while ((p = fts_read(fts)) != NULL) {
+               switch (p->fts_info) {
+               case FTS_DNR:
+                       /* FALLTHROUGH */
+               case FTS_ERR:
+                       errx(2, "%s: %s", p->fts_path, strerror(p->fts_errno));
+                       break;
+               case FTS_D:
+                       /* FALLTHROUGH */
+               case FTS_DP:
+                       break;
+               case FTS_DC:
+                       /* Print a warning for recursive directory loop */
+                       warnx("warning: %s: recursive directory loop",
+                               p->fts_path);
+                       break;
+               default:
+                       /* Check for file exclusion/inclusion */
+                       ok = true;
+                       if (dexclude || dinclude) {
+                               if ((d = strrchr(p->fts_path, '/')) != NULL) {
+                                       dir = grep_malloc(sizeof(char) *
+                                           (d - p->fts_path + 1));
+                                       memcpy(dir, p->fts_path,
+                                           d - p->fts_path);
+                                       dir[d - p->fts_path] = '\0';
+                               }
+                               ok = dir_matching(dir);
+                               free(dir);
+                               dir = NULL;
+                       }
+                       if (fexclude || finclude)
+                               ok &= file_matching(p->fts_path);
+
+                       if (ok)
+                               c += procfile(p->fts_path);
+                       break;
+               }
+       }
+
+       fts_close(fts);
+       return (c);
+}
+
+/*
+ * Opens a file and processes it.  Each file is processed line-by-line
+ * passing the lines to procline().
+ */
+int
+procfile(const char *fn)
+{
+       struct file *f;
+       struct stat sb;
+       struct str ln;
+       mode_t s;
+       int c, t;
+
+       if (mflag && (mcount <= 0))
+               return (0);
+
+       if (strcmp(fn, "-") == 0) {
+               fn = label != NULL ? label : getstr(1);
+               f = grep_open(NULL);
+       } else {
+               if (!stat(fn, &sb)) {
+                       /* Check if we need to process the file */
+                       s = sb.st_mode & S_IFMT;
+                       if (s == S_IFDIR && dirbehave == DIR_SKIP)
+                               return (0);
+                       if ((s == S_IFIFO || s == S_IFCHR || s == S_IFBLK
+                               || s == S_IFSOCK) && devbehave == DEV_SKIP)
+                                       return (0);
+               }
+               f = grep_open(fn);
+       }
+       if (f == NULL) {
+               if (!sflag)
+                       warn("%s", fn);
+               if (errno == ENOENT)
+                       notfound = true;
+               return (0);
+       }
+
+       ln.file = grep_malloc(strlen(fn) + 1);
+       strcpy(ln.file, fn);
+       ln.line_no = 0;
+       ln.len = 0;
+       tail = 0;
+       ln.off = -1;
+
+       for (first = true, c = 0;  c == 0 || !(lflag || qflag); ) {
+               ln.off += ln.len + 1;
+               if ((ln.dat = grep_fgetln(f, &ln.len)) == NULL || ln.len == 0)
+                       break;
+               if (ln.len > 0 && ln.dat[ln.len - 1] == line_sep)
+                       --ln.len;
+               ln.line_no++;
+
+               /* Return if we need to skip a binary file */
+               if (f->binary && binbehave == BINFILE_SKIP) {
+                       grep_close(f);
+                       free(ln.file);
+                       free(f);
+                       return (0);
+               }
+               /* Process the file line-by-line */
+               t = procline(&ln, f->binary);
+               c += t;
+
+               /* Count the matches if we have a match limit */
+               if (mflag) {
+                       mcount -= t;
+                       if (mcount <= 0)
+                               break;
+               }
+       }
+       if (Bflag > 0)
+               clearqueue();
+       grep_close(f);
+
+       if (cflag) {
+               if (!hflag)
+                       printf("%s:", ln.file);
+               printf("%u%c", c, line_sep);
+       }
+       if (lflag && !qflag && c != 0)
+               printf("%s%c", fn, line_sep);
+       if (Lflag && !qflag && c == 0)
+               printf("%s%c", fn, line_sep);
+       if (c && !cflag && !lflag && !Lflag &&
+           binbehave == BINFILE_BIN && f->binary && !qflag)
+               printf(getstr(8), fn);
+
+       free(ln.file);
+       free(f);
+       return (c);
+}
+
+#define iswword(x)     (iswalnum((x)) || (x) == L'_')
+
+/*
+ * Processes a line comparing it with the specified patterns.  Each pattern
+ * is looped to be compared along with the full string, saving each and every
+ * match, which is necessary to colorize the output and to count the
+ * matches.  The matching lines are passed to printline() to display the
+ * appropriate output.
+ */
+static int
+procline(struct str *l, int nottext)
+{
+       regmatch_t matches[MAX_LINE_MATCHES];
+       regmatch_t pmatch;
+       size_t st = 0;
+       unsigned int i;
+       int c = 0, m = 0, r = 0;
+
+       /* Loop to process the whole line */
+       while (st <= l->len) {
+               pmatch.rm_so = st;
+               pmatch.rm_eo = l->len;
+
+               /* Loop to compare with all the patterns */
+               for (i = 0; i < patterns; i++) {
+/*
+ * XXX: grep_search() is a workaround for speed up and should be
+ * removed in the future.  See fastgrep.c.
+ */
+                       if (fg_pattern[i].pattern) {
+                               r = grep_search(&fg_pattern[i],
+                                   (unsigned char *)l->dat,
+                                   l->len, &pmatch);
+                               r = (r == 0) ? 0 : REG_NOMATCH;
+                               st = pmatch.rm_eo;
+                       } else {
+                               r = regexec(&r_pattern[i], l->dat, 1,
+                                   &pmatch, eflags);
+                               r = (r == 0) ? 0 : REG_NOMATCH;
+                               st = pmatch.rm_eo;
+                       }
+                       if (r == REG_NOMATCH)
+                               continue;
+                       /* Check for full match */
+                       if (xflag &&
+                           (pmatch.rm_so != 0 ||
+                            (size_t)pmatch.rm_eo != l->len))
+                               continue;
+                       /* Check for whole word match */
+                       if (fg_pattern[i].word && pmatch.rm_so != 0) {
+                               wint_t wbegin, wend;
+
+                               wbegin = wend = L' ';
+                               if (pmatch.rm_so != 0 &&
+                                   sscanf(&l->dat[pmatch.rm_so - 1],
+                                   "%lc", &wbegin) != 1)
+                                       continue;
+                               if ((size_t)pmatch.rm_eo != l->len &&
+                                   sscanf(&l->dat[pmatch.rm_eo],
+                                   "%lc", &wend) != 1)
+                                       continue;
+                               if (iswword(wbegin) || iswword(wend))
+                                       continue;
+                       }
+                       c = 1;
+                       if (m < MAX_LINE_MATCHES)
+                               matches[m++] = pmatch;
+                       /* matches - skip further patterns */
+                       if ((color != NULL && !oflag) || qflag || lflag)
+                               break;
+               }
+
+               if (vflag) {
+                       c = !c;
+                       break;
+               }
+               /* One pass if we are not recording matches */
+               if ((color != NULL && !oflag) || qflag || lflag)
+                       break;
+
+               if (st == (size_t)pmatch.rm_so)
+                       break;  /* No matches */
+       }
+
+       if (c && binbehave == BINFILE_BIN && nottext)
+               return (c); /* Binary file */
+
+       /* Dealing with the context */
+       if ((tail || c) && !cflag && !qflag && !lflag && !Lflag) {
+               if (c) {
+                       if ((Aflag || Bflag) && !first_global &&
+                           (first || since_printed > Bflag))
+                               printf("--\n");
+                       tail = Aflag;
+                       if (Bflag > 0)
+                               printqueue();
+                       printline(l, ':', matches, m);
+               } else {
+                       printline(l, '-', matches, m);
+                       tail--;
+               }
+               first = false;
+               first_global = false;
+               since_printed = 0;
+       } else {
+               if (Bflag)
+                       enqueue(l);
+               since_printed++;
+       }
+       return (c);
+}
+
+/*
+ * Safe malloc() for internal use.
+ */
+void *
+grep_malloc(size_t size)
+{
+       void *ptr;
+
+       if ((ptr = malloc(size)) == NULL)
+               err(2, "malloc");
+       return (ptr);
+}
+
+/*
+ * Safe calloc() for internal use.
+ */
+void *
+grep_calloc(size_t nmemb, size_t size)
+{
+       void *ptr;
+
+       if ((ptr = calloc(nmemb, size)) == NULL)
+               err(2, "calloc");
+       return (ptr);
+}
+
+/*
+ * Safe realloc() for internal use.
+ */
+void *
+grep_realloc(void *ptr, size_t size)
+{
+
+       if ((ptr = realloc(ptr, size)) == NULL)
+               err(2, "realloc");
+       return (ptr);
+}
+
+/*
+ * Safe strdup() for internal use.
+ */
+char *
+grep_strdup(const char *str)
+{
+       char *ret;
+
+       if ((ret = strdup(str)) == NULL)
+               err(2, "strdup");
+       return (ret);
+}
+
+/*
+ * Prints a matching line according to the command line options.
+ */
+void
+printline(struct str *line, int sep, regmatch_t *matches, int m)
+{
+       size_t a = 0;
+       int i, n = 0;
+
+       if (!hflag) {
+               if (nullflag == 0)
+                       fputs(line->file, stdout);
+               else {
+                       printf("%s", line->file);
+                       putchar(0);
+               }
+               ++n;
+       }
+       if (nflag) {
+               if (n > 0)
+                       putchar(sep);
+               printf("%d", line->line_no);
+               ++n;
+       }
+       if (bflag) {
+               if (n > 0)
+                       putchar(sep);
+               printf("%lld", (long long)line->off);
+               ++n;
+       }
+       if (n)
+               putchar(sep);
+       /* --color and -o */
+       if ((oflag || color) && m > 0) {
+               for (i = 0; i < m; i++) {
+                       if (!oflag)
+                               fwrite(line->dat + a, matches[i].rm_so - a, 1,
+                                   stdout);
+                       if (color) 
+                               fprintf(stdout, "\33[%sm\33[K", color);
+
+                               fwrite(line->dat + matches[i].rm_so, 
+                                   matches[i].rm_eo - matches[i].rm_so, 1,
+                                   stdout);
+                       if (color) 
+                               fprintf(stdout, "\33[m\33[K");
+                       a = matches[i].rm_eo;
+                       if (oflag)
+                               putchar('\n');
+               }
+               if (!oflag) {
+                       if (line->len - a > 0)
+                               fwrite(line->dat + a, line->len - a, 1, stdout);
+                       putchar(line_sep);
+               }
+       } else {
+               fwrite(line->dat, line->len, 1, stdout);
+               putchar(line_sep);
+       }
+}