2011-04-07 13:03:35 +00:00
|
|
|
/* $NetBSD: grep.c,v 1.4 2011/02/16 01:31:33 joerg Exp $ */
|
|
|
|
/* $FreeBSD$ */
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
/* $OpenBSD: grep.c,v 1.42 2010/07/02 22:18:03 tedu Exp $ */
|
|
|
|
|
|
|
|
/*-
|
2010-08-19 09:28:59 +00:00
|
|
|
* Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
* 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 <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#include <sys/types.h>
|
|
|
|
|
|
|
|
#include <ctype.h>
|
|
|
|
#include <err.h>
|
|
|
|
#include <errno.h>
|
2011-10-05 09:56:43 +00:00
|
|
|
#include <fcntl.h>
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
#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>
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
#include "fastmatch.h"
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
#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",
|
2010-07-29 18:02:57 +00:00
|
|
|
/* 3*/ "unknown %s option",
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
/* 4*/ "usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-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[--null] [pattern] [file ...]\n",
|
2010-07-29 18:02:57 +00:00
|
|
|
/* 8*/ "Binary file %s matches\n",
|
|
|
|
/* 9*/ "%s (BSD grep) %s\n",
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/* Flags passed to regcomp() and regexec() */
|
2011-06-12 12:51:58 +00:00
|
|
|
int cflags = REG_NOSUB;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
int eflags = REG_STARTEND;
|
|
|
|
|
|
|
|
/* Shortcut for matching all cases like empty regex */
|
|
|
|
bool matchall;
|
|
|
|
|
|
|
|
/* Searching patterns */
|
2012-10-19 14:49:42 +00:00
|
|
|
unsigned int patterns;
|
|
|
|
static unsigned int pattern_sz;
|
2011-10-05 09:56:43 +00:00
|
|
|
struct pat *pattern;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
regex_t *r_pattern;
|
2011-10-05 09:56:43 +00:00
|
|
|
fastmatch_t *fg_pattern;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
|
|
|
/* Filename exclusion/inclusion patterns */
|
2012-10-19 14:49:42 +00:00
|
|
|
unsigned int fpatterns, dpatterns;
|
|
|
|
static unsigned int fpattern_sz, dpattern_sz;
|
2010-07-29 00:11:14 +00:00
|
|
|
struct epat *dpattern, *fpattern;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
|
|
|
/* 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 */
|
2011-10-05 09:56:43 +00:00
|
|
|
long long mcount; /* count for -m */
|
2012-12-20 17:38:14 +00:00
|
|
|
long long mlimit; /* requested value for -m */
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
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 */
|
|
|
|
char *label; /* --label */
|
2010-07-25 08:42:18 +00:00
|
|
|
const char *color; /* --color */
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
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 */
|
2010-07-25 08:42:18 +00:00
|
|
|
int devbehave = DEV_READ; /* -D: handling of devices */
|
|
|
|
int dirbehave = DIR_READ; /* -dRr: handling of directories */
|
|
|
|
int linkbehave = LINK_READ; /* -OpS: handling of symlinks */
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
bool dexclude, dinclude; /* --exclude-dir and --include-dir */
|
|
|
|
bool fexclude, finclude; /* --exclude and --include */
|
2010-07-29 00:11:14 +00:00
|
|
|
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
enum {
|
|
|
|
BIN_OPT = CHAR_MAX + 1,
|
|
|
|
COLOR_OPT,
|
|
|
|
HELP_OPT,
|
|
|
|
MMAP_OPT,
|
|
|
|
LINEBUF_OPT,
|
|
|
|
LABEL_OPT,
|
|
|
|
NULL_OPT,
|
|
|
|
R_EXCLUDE_OPT,
|
|
|
|
R_INCLUDE_OPT,
|
|
|
|
R_DEXCLUDE_OPT,
|
|
|
|
R_DINCLUDE_OPT
|
|
|
|
};
|
|
|
|
|
2010-07-25 08:42:18 +00:00
|
|
|
static inline const char *init_color(const char *);
|
|
|
|
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
/* Housekeeping */
|
|
|
|
bool first = true; /* flag whether we are processing the first match */
|
|
|
|
bool prev; /* flag whether or not the previous line matched */
|
|
|
|
int tail; /* lines left to print */
|
2011-12-07 12:25:28 +00:00
|
|
|
bool file_err; /* file reading error */
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Prints usage information and returns 2.
|
|
|
|
*/
|
|
|
|
static void
|
|
|
|
usage(void)
|
|
|
|
{
|
2011-10-11 22:27:23 +00:00
|
|
|
fprintf(stderr, getstr(4), getprogname());
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
fprintf(stderr, "%s", getstr(5));
|
|
|
|
fprintf(stderr, "%s", getstr(6));
|
|
|
|
fprintf(stderr, "%s", getstr(7));
|
|
|
|
exit(2);
|
|
|
|
}
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
static const char *optstr = "0123456789A:B:C:D:EFGHIJMLOPSRUVZabcd:e:f:hilm:nopqrsuvwxXy";
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
2011-12-10 18:21:03 +00:00
|
|
|
static const struct option long_options[] =
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
{
|
|
|
|
{"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},
|
|
|
|
{"label", required_argument, NULL, LABEL_OPT},
|
|
|
|
{"null", no_argument, NULL, NULL_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'},
|
2011-10-05 09:56:43 +00:00
|
|
|
{"lzma", no_argument, NULL, 'M'},
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
{"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'},
|
2011-10-05 09:56:43 +00:00
|
|
|
{"xz", no_argument, NULL, 'X'},
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
{"decompress", 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)
|
|
|
|
{
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
/* Do not add further pattern is we already match everything */
|
|
|
|
if (matchall)
|
|
|
|
return;
|
|
|
|
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
/* Check if we can do a shortcut */
|
2011-10-05 09:56:43 +00:00
|
|
|
if (len == 0) {
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
matchall = true;
|
2011-10-05 09:56:43 +00:00
|
|
|
for (unsigned int i = 0; i < patterns; i++) {
|
|
|
|
free(pattern[i].pat);
|
|
|
|
}
|
|
|
|
pattern = grep_realloc(pattern, sizeof(struct pat));
|
|
|
|
pattern[0].pat = NULL;
|
|
|
|
pattern[0].len = 0;
|
|
|
|
patterns = 1;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
/* Increase size if necessary */
|
|
|
|
if (patterns == pattern_sz) {
|
|
|
|
pattern_sz *= 2;
|
|
|
|
pattern = grep_realloc(pattern, ++pattern_sz *
|
2011-10-05 09:56:43 +00:00
|
|
|
sizeof(struct pat));
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
}
|
|
|
|
if (len > 0 && pat[len - 1] == '\n')
|
|
|
|
--len;
|
|
|
|
/* pat may not be NUL-terminated */
|
2011-10-05 09:56:43 +00:00
|
|
|
pattern[patterns].pat = grep_malloc(len + 1);
|
|
|
|
memcpy(pattern[patterns].pat, pat, len);
|
|
|
|
pattern[patterns].len = len;
|
|
|
|
pattern[patterns].pat[len] = '\0';
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
++patterns;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-07-29 00:11:14 +00:00
|
|
|
* Adds a file include/exclude pattern to the internal array.
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
*/
|
|
|
|
static void
|
2010-07-29 00:11:14 +00:00
|
|
|
add_fpattern(const char *pat, int mode)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
/* Increase size if necessary */
|
2010-07-29 00:11:14 +00:00
|
|
|
if (fpatterns == fpattern_sz) {
|
|
|
|
fpattern_sz *= 2;
|
|
|
|
fpattern = grep_realloc(fpattern, ++fpattern_sz *
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
sizeof(struct epat));
|
|
|
|
}
|
2010-07-29 00:11:14 +00:00
|
|
|
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;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Reads searching patterns from a file and adds them with add_pattern().
|
|
|
|
*/
|
|
|
|
static void
|
|
|
|
read_patterns(const char *fn)
|
|
|
|
{
|
2011-10-05 09:56:43 +00:00
|
|
|
struct stat st;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
FILE *f;
|
|
|
|
char *line;
|
|
|
|
size_t len;
|
|
|
|
|
|
|
|
if ((f = fopen(fn, "r")) == NULL)
|
|
|
|
err(2, "%s", fn);
|
2011-10-05 09:56:43 +00:00
|
|
|
if ((fstat(fileno(f), &st) == -1) || (S_ISDIR(st.st_mode))) {
|
|
|
|
fclose(f);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
while ((line = fgetln(f, &len)) != NULL)
|
|
|
|
add_pattern(line, line[0] == '\n' ? 0 : len);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
if (ferror(f))
|
|
|
|
err(2, "%s", fn);
|
|
|
|
fclose(f);
|
|
|
|
}
|
|
|
|
|
2010-07-25 08:42:18 +00:00
|
|
|
static inline const char *
|
|
|
|
init_color(const char *d)
|
|
|
|
{
|
|
|
|
char *c;
|
|
|
|
|
|
|
|
c = getenv("GREP_COLOR");
|
2011-08-17 13:56:33 +00:00
|
|
|
return (c != NULL && c[0] != '\0' ? c : d);
|
2010-07-25 08:42:18 +00:00
|
|
|
}
|
|
|
|
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
int
|
|
|
|
main(int argc, char *argv[])
|
|
|
|
{
|
|
|
|
char **aargv, **eargv, *eopts;
|
2011-10-11 22:27:23 +00:00
|
|
|
char *ep;
|
|
|
|
const char *pn;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
unsigned long long l;
|
|
|
|
unsigned int aargc, eargc, i;
|
|
|
|
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. */
|
2011-10-11 22:27:23 +00:00
|
|
|
pn = getprogname();
|
2011-10-05 09:56:43 +00:00
|
|
|
if (pn[0] == 'b' && pn[1] == 'z') {
|
|
|
|
filebehave = FILE_BZIP;
|
|
|
|
pn += 2;
|
|
|
|
} else if (pn[0] == 'x' && pn[1] == 'z') {
|
|
|
|
filebehave = FILE_XZ;
|
|
|
|
pn += 2;
|
|
|
|
} else if (pn[0] == 'l' && pn[1] == 'z') {
|
|
|
|
filebehave = FILE_LZMA;
|
|
|
|
pn += 2;
|
|
|
|
} else if (pn[0] == 'z') {
|
|
|
|
filebehave = FILE_GZIP;
|
|
|
|
pn += 1;
|
|
|
|
}
|
|
|
|
switch (pn[0]) {
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
case 'e':
|
|
|
|
grepbehave = GREP_EXTENDED;
|
|
|
|
break;
|
|
|
|
case 'f':
|
|
|
|
grepbehave = GREP_FIXED;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
lastc = '\0';
|
|
|
|
newarg = 1;
|
|
|
|
prevoptind = 1;
|
|
|
|
needpattern = 1;
|
|
|
|
|
|
|
|
eopts = getenv("GREP_OPTIONS");
|
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
/* support for extra arguments in GREP_OPTIONS */
|
|
|
|
eargc = 0;
|
2011-08-17 13:56:33 +00:00
|
|
|
if (eopts != NULL && eopts[0] != '\0') {
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
char *str;
|
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
/* make an estimation of how many extra arguments we have */
|
|
|
|
for (unsigned int j = 0; j < strlen(eopts); j++)
|
|
|
|
if (eopts[j] == ' ')
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
eargc++;
|
|
|
|
|
|
|
|
eargv = (char **)grep_malloc(sizeof(char *) * (eargc + 1));
|
|
|
|
|
|
|
|
eargc = 0;
|
2010-08-15 22:15:04 +00:00
|
|
|
/* parse extra arguments */
|
|
|
|
while ((str = strsep(&eopts, " ")) != NULL)
|
2011-08-17 13:56:33 +00:00
|
|
|
if (str[0] != '\0')
|
|
|
|
eargv[eargc++] = grep_strdup(str);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
2010-07-23 19:36:11 +00:00
|
|
|
aargv = (char **)grep_calloc(eargc + argc + 1,
|
|
|
|
sizeof(char *));
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
aargv[0] = argv[0];
|
|
|
|
for (i = 0; i < eargc; i++)
|
|
|
|
aargv[i + 1] = eargv[i];
|
|
|
|
for (int j = 1; j < argc; j++, i++)
|
|
|
|
aargv[i + 1] = argv[j];
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
aargc = eargc + argc;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
} 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':
|
2010-07-25 08:42:18 +00:00
|
|
|
if (strcasecmp(optarg, "skip") == 0)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
devbehave = DEV_SKIP;
|
2010-07-25 08:42:18 +00:00
|
|
|
else if (strcasecmp(optarg, "read") == 0)
|
|
|
|
devbehave = DEV_READ;
|
2010-07-29 18:02:57 +00:00
|
|
|
else
|
|
|
|
errx(2, getstr(3), "--devices");
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'd':
|
2010-07-25 08:42:18 +00:00
|
|
|
if (strcasecmp("recurse", optarg) == 0) {
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
Hflag = true;
|
|
|
|
dirbehave = DIR_RECURSE;
|
2010-07-25 08:42:18 +00:00
|
|
|
} else if (strcasecmp("skip", optarg) == 0)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
dirbehave = DIR_SKIP;
|
2010-07-25 08:42:18 +00:00
|
|
|
else if (strcasecmp("read", optarg) == 0)
|
|
|
|
dirbehave = DIR_READ;
|
2010-07-29 18:02:57 +00:00
|
|
|
else
|
|
|
|
errx(2, getstr(3), "--directories");
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'E':
|
|
|
|
grepbehave = GREP_EXTENDED;
|
|
|
|
break;
|
|
|
|
case 'e':
|
2013-01-05 14:52:31 +00:00
|
|
|
{
|
|
|
|
char *token;
|
2013-01-20 11:58:49 +00:00
|
|
|
char *string = optarg;
|
2013-01-05 14:52:31 +00:00
|
|
|
|
|
|
|
while ((token = strsep(&string, "\n")) != NULL)
|
|
|
|
add_pattern(token, strlen(token));
|
|
|
|
}
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
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':
|
2011-10-11 22:27:23 +00:00
|
|
|
#ifdef WITHOUT_BZIP2
|
|
|
|
errno = EOPNOTSUPP;
|
|
|
|
err(2, "bzip2 support was disabled at compile-time");
|
|
|
|
#endif
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
filebehave = FILE_BZIP;
|
|
|
|
break;
|
|
|
|
case 'L':
|
|
|
|
lflag = false;
|
2010-07-25 08:42:18 +00:00
|
|
|
Lflag = true;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'l':
|
|
|
|
Lflag = false;
|
2010-07-25 08:42:18 +00:00
|
|
|
lflag = true;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'm':
|
|
|
|
mflag = true;
|
|
|
|
errno = 0;
|
2012-12-20 17:38:14 +00:00
|
|
|
mlimit = mcount = strtoll(optarg, &ep, 10);
|
2011-10-05 09:56:43 +00:00
|
|
|
if (((errno == ERANGE) && (mcount == LLONG_MAX)) ||
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
((errno == EINVAL) && (mcount == 0)))
|
|
|
|
err(2, NULL);
|
|
|
|
else if (ep[0] != '\0') {
|
|
|
|
errno = EINVAL;
|
|
|
|
err(2, NULL);
|
|
|
|
}
|
|
|
|
break;
|
2011-10-05 09:56:43 +00:00
|
|
|
case 'M':
|
|
|
|
filebehave = FILE_LZMA;
|
|
|
|
break;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
case 'n':
|
|
|
|
nflag = true;
|
|
|
|
break;
|
|
|
|
case 'O':
|
|
|
|
linkbehave = LINK_EXPLICIT;
|
|
|
|
break;
|
|
|
|
case 'o':
|
|
|
|
oflag = true;
|
2011-06-12 12:51:58 +00:00
|
|
|
cflags &= ~REG_NOSUB;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'p':
|
|
|
|
linkbehave = LINK_SKIP;
|
|
|
|
break;
|
|
|
|
case 'q':
|
|
|
|
qflag = true;
|
|
|
|
break;
|
|
|
|
case 'S':
|
2010-07-25 08:42:18 +00:00
|
|
|
linkbehave = LINK_READ;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
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:
|
2011-10-05 09:56:43 +00:00
|
|
|
filebehave = FILE_MMAP;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'V':
|
2011-10-11 22:27:23 +00:00
|
|
|
printf(getstr(9), getprogname(), VERSION);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
exit(0);
|
|
|
|
case 'v':
|
|
|
|
vflag = true;
|
|
|
|
break;
|
|
|
|
case 'w':
|
|
|
|
wflag = true;
|
2011-06-12 12:51:58 +00:00
|
|
|
cflags &= ~REG_NOSUB;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case 'x':
|
|
|
|
xflag = true;
|
2011-06-12 12:51:58 +00:00
|
|
|
cflags &= ~REG_NOSUB;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
2011-10-05 09:56:43 +00:00
|
|
|
case 'X':
|
|
|
|
filebehave = FILE_XZ;
|
|
|
|
break;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
case 'Z':
|
|
|
|
filebehave = FILE_GZIP;
|
|
|
|
break;
|
|
|
|
case BIN_OPT:
|
2010-07-25 08:42:18 +00:00
|
|
|
if (strcasecmp("binary", optarg) == 0)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
binbehave = BINFILE_BIN;
|
2010-07-25 08:42:18 +00:00
|
|
|
else if (strcasecmp("without-match", optarg) == 0)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
binbehave = BINFILE_SKIP;
|
2010-07-25 08:42:18 +00:00
|
|
|
else if (strcasecmp("text", optarg) == 0)
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
binbehave = BINFILE_TEXT;
|
|
|
|
else
|
2010-07-29 18:02:57 +00:00
|
|
|
errx(2, getstr(3), "--binary-files");
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case COLOR_OPT:
|
2010-07-25 08:42:18 +00:00
|
|
|
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)
|
2010-07-29 18:02:57 +00:00
|
|
|
errx(2, getstr(3), "--color");
|
2011-06-12 12:51:58 +00:00
|
|
|
cflags &= ~REG_NOSUB;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case LABEL_OPT:
|
|
|
|
label = optarg;
|
|
|
|
break;
|
|
|
|
case LINEBUF_OPT:
|
|
|
|
lbflag = true;
|
|
|
|
break;
|
|
|
|
case NULL_OPT:
|
|
|
|
nullflag = true;
|
|
|
|
break;
|
|
|
|
case R_INCLUDE_OPT:
|
2010-07-29 00:11:14 +00:00
|
|
|
finclude = true;
|
|
|
|
add_fpattern(optarg, INCL_PAT);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case R_EXCLUDE_OPT:
|
2010-07-29 00:11:14 +00:00
|
|
|
fexclude = true;
|
|
|
|
add_fpattern(optarg, EXCL_PAT);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case R_DINCLUDE_OPT:
|
2010-08-15 22:15:04 +00:00
|
|
|
dinclude = true;
|
2010-07-29 00:11:14 +00:00
|
|
|
add_dpattern(optarg, INCL_PAT);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case R_DEXCLUDE_OPT:
|
2010-08-15 22:15:04 +00:00
|
|
|
dexclude = true;
|
2010-07-29 00:11:14 +00:00
|
|
|
add_dpattern(optarg, EXCL_PAT);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
break;
|
|
|
|
case HELP_OPT:
|
|
|
|
default:
|
|
|
|
usage();
|
|
|
|
}
|
|
|
|
lastc = c;
|
|
|
|
newarg = optind != prevoptind;
|
|
|
|
prevoptind = optind;
|
|
|
|
}
|
|
|
|
aargc -= optind;
|
|
|
|
aargv += optind;
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
/* Empty pattern file matches nothing */
|
|
|
|
if (!needpattern && (patterns == 0))
|
|
|
|
exit(1);
|
|
|
|
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
/* Fail if we don't have any pattern */
|
|
|
|
if (aargc == 0 && needpattern)
|
|
|
|
usage();
|
|
|
|
|
|
|
|
/* Process patterns from command line */
|
|
|
|
if (aargc != 0 && needpattern) {
|
2013-01-05 14:52:31 +00:00
|
|
|
char *token;
|
2013-01-20 11:58:49 +00:00
|
|
|
char *string = *aargv;
|
2013-01-05 14:52:31 +00:00
|
|
|
|
|
|
|
while ((token = strsep(&string, "\n")) != NULL)
|
|
|
|
add_pattern(token, strlen(token));
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
--aargc;
|
|
|
|
++aargv;
|
|
|
|
}
|
|
|
|
|
|
|
|
switch (grepbehave) {
|
|
|
|
case GREP_BASIC:
|
|
|
|
break;
|
2011-10-05 09:56:43 +00:00
|
|
|
case GREP_FIXED:
|
|
|
|
/* XXX: header mess, REG_LITERAL not defined in gnu/regex.h */
|
|
|
|
cflags |= 0020;
|
|
|
|
break;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
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));
|
2011-10-05 09:56:43 +00:00
|
|
|
|
|
|
|
/* Check if cheating is allowed (always is for fgrep). */
|
|
|
|
for (i = 0; i < patterns; ++i) {
|
|
|
|
if (fastncomp(&fg_pattern[i], pattern[i].pat,
|
|
|
|
pattern[i].len, cflags) != 0) {
|
|
|
|
/* Fall back to full regex library */
|
|
|
|
c = regcomp(&r_pattern[i], pattern[i].pat, cflags);
|
|
|
|
if (c != 0) {
|
|
|
|
regerror(c, &r_pattern[i], re_error,
|
|
|
|
RE_ERROR_BUF);
|
|
|
|
errx(2, "%s", re_error);
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
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);
|
2010-08-19 22:55:17 +00:00
|
|
|
else
|
2010-07-29 00:11:14 +00:00
|
|
|
for (c = 0; aargc--; ++aargv) {
|
|
|
|
if ((finclude || fexclude) && !file_matching(*aargv))
|
|
|
|
continue;
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
c+= procfile(*aargv);
|
2010-07-29 00:11:14 +00:00
|
|
|
}
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
|
|
|
|
#ifndef WITHOUT_NLS
|
|
|
|
catclose(catalog);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Find out the correct return value according to the
|
|
|
|
results and the command line option. */
|
2011-12-07 12:25:28 +00:00
|
|
|
exit(c ? (file_err ? (qflag ? 0 : 2) : 0) : (file_err ? 2 : 1));
|
Add BSD grep to the base system and make it our default grep.
Deliverables: Small and clean code (1,4 KSLOC vs GNU's 8,5 KSLOC),
lower memory usage than GNU grep, GNU compatibility,
BSD license.
TODO: Performance is somewhat behind GNU grep but it is only
significant for bigger searches. The reason is complex, the
most important factor is that GNU grep uses lots of
optimizations to improve the speed of the regex library.
First, we need a modern regex library (practically by adopting
TRE), add support for GNU-style non-standard regexes and then
reevalute the performance issues and look for bottlenecks. In
the meantime, for those, who need better performance, it is
possible to build GNU grep by setting WITH_GNU_GREP.
Approved by: delphij (mentor)
Obtained from: OpenBSD (http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/grep/),
freegrep (http://github.com/howardjp/freegrep)
Sponsored by: Google SoC 2008
Portbuild tests run by: kris, pav, erwin
Acknowledgements to: fjoe (as SoC 2008 mentor),
everyone who helped in reviewing and testing
2010-07-22 19:11:57 +00:00
|
|
|
}
|