2011-04-07 13:03:35 +00:00
|
|
|
/* $NetBSD: util.c,v 1.9 2011/02/27 17:33:37 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: util.c,v 1.39 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) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org>
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#include <sys/types.h>
|
|
|
|
|
|
|
|
#include <ctype.h>
|
|
|
|
#include <err.h>
|
|
|
|
#include <errno.h>
|
|
|
|
#include <fnmatch.h>
|
|
|
|
#include <fts.h>
|
|
|
|
#include <libgen.h>
|
2010-07-29 00:11:14 +00:00
|
|
|
#include <stdbool.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 <stdio.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
#include <wchar.h>
|
|
|
|
#include <wctype.h>
|
|
|
|
|
2017-04-21 14:36:09 +00:00
|
|
|
#ifndef WITHOUT_FASTMATCH
|
2011-10-05 09:56:43 +00:00
|
|
|
#include "fastmatch.h"
|
2017-04-21 14:36:09 +00:00
|
|
|
#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
|
|
|
#include "grep.h"
|
|
|
|
|
|
|
|
static int linesqueued;
|
|
|
|
static int procline(struct str *l, int);
|
|
|
|
|
2017-04-17 13:36:30 +00:00
|
|
|
static int lasta;
|
|
|
|
static bool ctxover;
|
|
|
|
|
2010-07-29 00:11:14 +00:00
|
|
|
bool
|
|
|
|
file_matching(const char *fname)
|
|
|
|
{
|
2016-07-28 15:19:47 +00:00
|
|
|
char *fname_base, *fname_buf;
|
2010-07-29 00:11:14 +00:00
|
|
|
bool ret;
|
|
|
|
|
|
|
|
ret = finclude ? false : true;
|
2016-07-28 15:19:47 +00:00
|
|
|
fname_buf = strdup(fname);
|
|
|
|
if (fname_buf == NULL)
|
|
|
|
err(2, "strdup");
|
|
|
|
fname_base = basename(fname_buf);
|
2010-07-29 00:11:14 +00:00
|
|
|
|
|
|
|
for (unsigned int i = 0; i < fpatterns; ++i) {
|
2011-04-07 13:01:03 +00:00
|
|
|
if (fnmatch(fpattern[i].pat, fname, 0) == 0 ||
|
|
|
|
fnmatch(fpattern[i].pat, fname_base, 0) == 0) {
|
2016-07-28 15:19:47 +00:00
|
|
|
if (fpattern[i].mode == EXCL_PAT) {
|
|
|
|
ret = false;
|
|
|
|
break;
|
|
|
|
} else
|
2010-07-29 00:11:14 +00:00
|
|
|
ret = true;
|
|
|
|
}
|
|
|
|
}
|
2016-07-28 15:19:47 +00:00
|
|
|
free(fname_buf);
|
2010-07-29 00:11:14 +00:00
|
|
|
return (ret);
|
|
|
|
}
|
|
|
|
|
2010-08-15 22:15:04 +00:00
|
|
|
static inline bool
|
2010-07-29 00:11:14 +00:00
|
|
|
dir_matching(const char *dname)
|
|
|
|
{
|
|
|
|
bool ret;
|
|
|
|
|
|
|
|
ret = dinclude ? false : true;
|
|
|
|
|
|
|
|
for (unsigned int i = 0; i < dpatterns; ++i) {
|
|
|
|
if (dname != NULL &&
|
2011-08-17 13:58:39 +00:00
|
|
|
fnmatch(dpattern[i].pat, dname, 0) == 0) {
|
2010-07-29 00:11:14 +00:00
|
|
|
if (dpattern[i].mode == EXCL_PAT)
|
|
|
|
return (false);
|
|
|
|
else
|
|
|
|
ret = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return (ret);
|
|
|
|
}
|
|
|
|
|
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
|
|
|
/*
|
|
|
|
* Processes a directory when a recursive search is performed with
|
|
|
|
* the -R option. Each appropriate file is passed to procfile().
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
grep_tree(char **argv)
|
|
|
|
{
|
|
|
|
FTS *fts;
|
|
|
|
FTSENT *p;
|
|
|
|
int c, fts_flags;
|
|
|
|
bool ok;
|
2017-04-17 13:22:39 +00:00
|
|
|
const char *wd[] = { ".", NULL };
|
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 = fts_flags = 0;
|
|
|
|
|
|
|
|
switch(linkbehave) {
|
|
|
|
case LINK_EXPLICIT:
|
|
|
|
fts_flags = FTS_COMFOLLOW;
|
|
|
|
break;
|
|
|
|
case LINK_SKIP:
|
|
|
|
fts_flags = FTS_PHYSICAL;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
fts_flags = FTS_LOGICAL;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
fts_flags |= FTS_NOSTAT | FTS_NOCHDIR;
|
|
|
|
|
2017-04-17 13:22:39 +00:00
|
|
|
fts = fts_open((argv[0] == NULL) ?
|
|
|
|
__DECONST(char * const *, wd) : argv, fts_flags, NULL);
|
|
|
|
if (fts == NULL)
|
2010-07-23 19:36:11 +00:00
|
|
|
err(2, "fts_open");
|
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
|
|
|
while ((p = fts_read(fts)) != NULL) {
|
|
|
|
switch (p->fts_info) {
|
|
|
|
case FTS_DNR:
|
|
|
|
/* FALLTHROUGH */
|
|
|
|
case FTS_ERR:
|
2011-12-07 12:25:28 +00:00
|
|
|
file_err = true;
|
2011-11-28 20:04:26 +00:00
|
|
|
if(!sflag)
|
|
|
|
warnx("%s: %s", p->fts_path, strerror(p->fts_errno));
|
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 FTS_D:
|
|
|
|
/* FALLTHROUGH */
|
|
|
|
case FTS_DP:
|
2011-08-17 13:58:39 +00:00
|
|
|
if (dexclude || dinclude)
|
|
|
|
if (!dir_matching(p->fts_name) ||
|
|
|
|
!dir_matching(p->fts_path))
|
|
|
|
fts_set(fts, p, FTS_SKIP);
|
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 FTS_DC:
|
|
|
|
/* Print a warning for recursive directory loop */
|
|
|
|
warnx("warning: %s: recursive directory loop",
|
|
|
|
p->fts_path);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
/* Check for file exclusion/inclusion */
|
|
|
|
ok = true;
|
2010-07-29 00:11:14 +00:00
|
|
|
if (fexclude || finclude)
|
|
|
|
ok &= file_matching(p->fts_path);
|
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 (ok)
|
|
|
|
c += procfile(p->fts_path);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-07-23 19:36:11 +00:00
|
|
|
fts_close(fts);
|
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 (c);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Opens a file and processes it. Each file is processed line-by-line
|
|
|
|
* passing the lines to procline().
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
procfile(const char *fn)
|
|
|
|
{
|
|
|
|
struct file *f;
|
|
|
|
struct stat sb;
|
|
|
|
struct str ln;
|
|
|
|
mode_t s;
|
|
|
|
int c, t;
|
|
|
|
|
2012-12-20 17:38:14 +00:00
|
|
|
mcount = mlimit;
|
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 (strcmp(fn, "-") == 0) {
|
|
|
|
fn = label != NULL ? label : getstr(1);
|
2010-08-18 17:40:10 +00:00
|
|
|
f = grep_open(NULL);
|
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 {
|
|
|
|
if (!stat(fn, &sb)) {
|
|
|
|
/* Check if we need to process the file */
|
|
|
|
s = sb.st_mode & S_IFMT;
|
|
|
|
if (s == S_IFDIR && dirbehave == DIR_SKIP)
|
|
|
|
return (0);
|
|
|
|
if ((s == S_IFIFO || s == S_IFCHR || s == S_IFBLK
|
|
|
|
|| s == S_IFSOCK) && devbehave == DEV_SKIP)
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
f = grep_open(fn);
|
|
|
|
}
|
|
|
|
if (f == NULL) {
|
2011-12-07 12:25:28 +00:00
|
|
|
file_err = 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
|
|
|
if (!sflag)
|
|
|
|
warn("%s", fn);
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
|
|
|
ln.file = grep_malloc(strlen(fn) + 1);
|
|
|
|
strcpy(ln.file, fn);
|
|
|
|
ln.line_no = 0;
|
|
|
|
ln.len = 0;
|
2017-04-17 13:36:30 +00:00
|
|
|
ctxover = false;
|
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
|
|
|
linesqueued = 0;
|
|
|
|
tail = 0;
|
2017-04-17 13:36:30 +00:00
|
|
|
lasta = 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
|
|
|
ln.off = -1;
|
|
|
|
|
|
|
|
for (c = 0; c == 0 || !(lflag || qflag); ) {
|
|
|
|
ln.off += ln.len + 1;
|
2010-08-18 17:40:10 +00:00
|
|
|
if ((ln.dat = grep_fgetln(f, &ln.len)) == NULL || ln.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
|
|
|
if (ln.line_no == 0 && matchall)
|
|
|
|
exit(0);
|
|
|
|
else
|
|
|
|
break;
|
|
|
|
}
|
2017-04-17 13:14:18 +00:00
|
|
|
if (ln.len > 0 && ln.dat[ln.len - 1] == fileeol)
|
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
|
|
|
--ln.len;
|
|
|
|
ln.line_no++;
|
|
|
|
|
|
|
|
/* Return if we need to skip a binary file */
|
|
|
|
if (f->binary && binbehave == BINFILE_SKIP) {
|
|
|
|
grep_close(f);
|
2010-07-23 19:36:11 +00:00
|
|
|
free(ln.file);
|
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
|
|
|
free(f);
|
|
|
|
return (0);
|
|
|
|
}
|
2017-04-17 13:36:30 +00:00
|
|
|
|
|
|
|
/* Process the file line-by-line, enqueue non-matching lines */
|
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 ((t = procline(&ln, f->binary)) == 0 && Bflag > 0) {
|
2017-04-17 13:36:30 +00:00
|
|
|
/* Except don't enqueue lines that appear in -A ctx */
|
|
|
|
if (ln.line_no == 0 || lasta != ln.line_no) {
|
|
|
|
/* queue is maxed to Bflag number of lines */
|
|
|
|
enqueue(&ln);
|
|
|
|
linesqueued++;
|
|
|
|
ctxover = false;
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Indicate to procline() that we have ctx
|
|
|
|
* overlap and make sure queue is empty.
|
|
|
|
*/
|
|
|
|
if (!ctxover)
|
|
|
|
clearqueue();
|
|
|
|
ctxover = 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
|
|
|
}
|
|
|
|
c += t;
|
2011-10-12 01:09:57 +00:00
|
|
|
if (mflag && mcount <= 0)
|
2011-10-05 09:56:43 +00:00
|
|
|
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
|
|
|
}
|
|
|
|
if (Bflag > 0)
|
|
|
|
clearqueue();
|
|
|
|
grep_close(f);
|
|
|
|
|
|
|
|
if (cflag) {
|
|
|
|
if (!hflag)
|
|
|
|
printf("%s:", ln.file);
|
|
|
|
printf("%u\n", c);
|
|
|
|
}
|
2010-07-25 08:42:18 +00:00
|
|
|
if (lflag && !qflag && c != 0)
|
2011-11-28 20:00:31 +00:00
|
|
|
printf("%s%c", fn, nullflag ? 0 : '\n');
|
2010-07-25 08:42:18 +00:00
|
|
|
if (Lflag && !qflag && c == 0)
|
2011-11-28 20:00:31 +00:00
|
|
|
printf("%s%c", fn, nullflag ? 0 : '\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
|
|
|
if (c && !cflag && !lflag && !Lflag &&
|
|
|
|
binbehave == BINFILE_BIN && f->binary && !qflag)
|
2010-07-29 18:02:57 +00:00
|
|
|
printf(getstr(8), fn);
|
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
|
|
|
free(ln.file);
|
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
|
|
|
free(f);
|
|
|
|
return (c);
|
|
|
|
}
|
|
|
|
|
|
|
|
#define iswword(x) (iswalnum((x)) || (x) == L'_')
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Processes a line comparing it with the specified patterns. Each pattern
|
|
|
|
* is looped to be compared along with the full string, saving each and every
|
|
|
|
* match, which is necessary to colorize the output and to count the
|
|
|
|
* matches. The matching lines are passed to printline() to display the
|
|
|
|
* appropriate output.
|
|
|
|
*/
|
2011-04-07 13:01:03 +00:00
|
|
|
static int
|
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
|
|
|
procline(struct str *l, int nottext)
|
|
|
|
{
|
|
|
|
regmatch_t matches[MAX_LINE_MATCHES];
|
2017-04-03 23:16:51 +00:00
|
|
|
regmatch_t pmatch, lastmatch;
|
|
|
|
size_t st = 0, nst = 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
|
|
|
unsigned int i;
|
2017-04-03 23:16:51 +00:00
|
|
|
int c = 0, m = 0, r = 0, lastmatches = 0, leflags = eflags;
|
|
|
|
int startm = 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
|
|
|
|
2017-04-04 13:34:19 +00:00
|
|
|
/* Initialize to avoid a false positive warning from GCC. */
|
|
|
|
lastmatch.rm_so = lastmatch.rm_eo = 0;
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
/* Loop to process the whole line */
|
|
|
|
while (st <= l->len) {
|
2017-04-03 23:16:51 +00:00
|
|
|
lastmatches = 0;
|
|
|
|
startm = m;
|
|
|
|
if (st > 0)
|
|
|
|
leflags |= REG_NOTBOL;
|
2011-10-05 09:56:43 +00:00
|
|
|
/* Loop to compare with all the patterns */
|
|
|
|
for (i = 0; i < patterns; i++) {
|
2017-04-03 23:16:51 +00:00
|
|
|
pmatch.rm_so = st;
|
|
|
|
pmatch.rm_eo = l->len;
|
2017-04-21 14:36:09 +00:00
|
|
|
#ifndef WITHOUT_FASTMATCH
|
2011-10-05 09:56:43 +00:00
|
|
|
if (fg_pattern[i].pattern)
|
|
|
|
r = fastexec(&fg_pattern[i],
|
2017-04-03 23:16:51 +00:00
|
|
|
l->dat, 1, &pmatch, leflags);
|
2011-10-05 09:56:43 +00:00
|
|
|
else
|
2017-04-21 14:36:09 +00:00
|
|
|
#endif
|
2011-10-05 09:56:43 +00:00
|
|
|
r = regexec(&r_pattern[i], l->dat, 1,
|
2017-04-03 23:16:51 +00:00
|
|
|
&pmatch, leflags);
|
2011-10-05 09:56:43 +00:00
|
|
|
r = (r == 0) ? 0 : REG_NOMATCH;
|
|
|
|
if (r == REG_NOMATCH)
|
|
|
|
continue;
|
|
|
|
/* Check for full match */
|
|
|
|
if (r == 0 && xflag)
|
|
|
|
if (pmatch.rm_so != 0 ||
|
|
|
|
(size_t)pmatch.rm_eo != l->len)
|
|
|
|
r = REG_NOMATCH;
|
|
|
|
/* Check for whole word match */
|
2017-04-21 14:36:09 +00:00
|
|
|
#ifndef WITHOUT_FASTMATCH
|
2011-10-05 09:56:43 +00:00
|
|
|
if (r == 0 && (wflag || fg_pattern[i].word)) {
|
2017-04-21 14:36:09 +00:00
|
|
|
#else
|
|
|
|
if (r == 0 && wflag) {
|
|
|
|
#endif
|
2014-07-17 14:51:50 +00:00
|
|
|
wchar_t wbegin, wend;
|
2011-10-05 09:56:43 +00:00
|
|
|
|
|
|
|
wbegin = wend = L' ';
|
|
|
|
if (pmatch.rm_so != 0 &&
|
|
|
|
sscanf(&l->dat[pmatch.rm_so - 1],
|
|
|
|
"%lc", &wbegin) != 1)
|
|
|
|
r = REG_NOMATCH;
|
|
|
|
else if ((size_t)pmatch.rm_eo !=
|
|
|
|
l->len &&
|
|
|
|
sscanf(&l->dat[pmatch.rm_eo],
|
|
|
|
"%lc", &wend) != 1)
|
|
|
|
r = REG_NOMATCH;
|
|
|
|
else if (iswword(wbegin) ||
|
|
|
|
iswword(wend))
|
|
|
|
r = REG_NOMATCH;
|
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
|
|
|
}
|
2011-10-05 09:56:43 +00:00
|
|
|
if (r == 0) {
|
2017-04-03 23:16:51 +00:00
|
|
|
lastmatches++;
|
|
|
|
lastmatch = pmatch;
|
2011-10-05 09:56:43 +00:00
|
|
|
if (m == 0)
|
|
|
|
c++;
|
2017-04-03 23:16:51 +00:00
|
|
|
|
|
|
|
if (m < MAX_LINE_MATCHES) {
|
|
|
|
/* Replace previous match if the new one is earlier and/or longer */
|
|
|
|
if (m > startm) {
|
|
|
|
if (pmatch.rm_so < matches[m-1].rm_so ||
|
|
|
|
(pmatch.rm_so == matches[m-1].rm_so && (pmatch.rm_eo - pmatch.rm_so) > (matches[m-1].rm_eo - matches[m-1].rm_so))) {
|
|
|
|
matches[m-1] = pmatch;
|
|
|
|
nst = pmatch.rm_eo;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* Advance as normal if not */
|
|
|
|
matches[m++] = pmatch;
|
|
|
|
nst = pmatch.rm_eo;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
/* matches - skip further patterns */
|
|
|
|
if ((color == NULL && !oflag) ||
|
|
|
|
qflag || lflag)
|
|
|
|
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
|
|
|
}
|
2011-10-05 09:56:43 +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
|
|
|
|
2011-10-05 09:56:43 +00:00
|
|
|
if (vflag) {
|
|
|
|
c = !c;
|
|
|
|
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
|
|
|
}
|
2011-10-05 09:56:43 +00:00
|
|
|
|
|
|
|
/* One pass if we are not recording matches */
|
2014-08-18 12:29:28 +00:00
|
|
|
if (!wflag && ((color == NULL && !oflag) || qflag || lflag || Lflag))
|
2011-10-05 09:56:43 +00:00
|
|
|
break;
|
|
|
|
|
2017-04-03 23:16:51 +00:00
|
|
|
/* If we didn't have any matches or REG_NOSUB set */
|
|
|
|
if (lastmatches == 0 || (cflags & REG_NOSUB))
|
|
|
|
nst = l->len;
|
|
|
|
|
|
|
|
if (lastmatches == 0)
|
|
|
|
/* No matches */
|
|
|
|
break;
|
|
|
|
else if (st == nst && lastmatch.rm_so == lastmatch.rm_eo)
|
|
|
|
/* Zero-length match -- advance one more so we don't get stuck */
|
|
|
|
nst++;
|
|
|
|
|
|
|
|
/* Advance st based on previous matches */
|
|
|
|
st = nst;
|
2011-10-05 09:56:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/* Count the matches if we have a match limit */
|
|
|
|
if (mflag)
|
|
|
|
mcount -= c;
|
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 (c && binbehave == BINFILE_BIN && nottext)
|
|
|
|
return (c); /* Binary file */
|
|
|
|
|
|
|
|
/* Dealing with the context */
|
2010-07-25 18:57:48 +00:00
|
|
|
if ((tail || c) && !cflag && !qflag && !lflag && !Lflag) {
|
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 (c) {
|
2017-04-17 13:36:30 +00:00
|
|
|
if (!first && !prev && !tail && (Bflag || Aflag) &&
|
|
|
|
!ctxover)
|
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
|
|
|
printf("--\n");
|
|
|
|
tail = Aflag;
|
|
|
|
if (Bflag > 0) {
|
|
|
|
printqueue();
|
2017-04-17 13:36:30 +00:00
|
|
|
ctxover = false;
|
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
|
|
|
}
|
|
|
|
linesqueued = 0;
|
|
|
|
printline(l, ':', matches, m);
|
|
|
|
} else {
|
2017-04-17 13:36:30 +00:00
|
|
|
/* Print -A lines following matches */
|
|
|
|
lasta = l->line_no;
|
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
|
|
|
printline(l, '-', matches, m);
|
|
|
|
tail--;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (c) {
|
|
|
|
prev = true;
|
|
|
|
first = false;
|
|
|
|
} else
|
|
|
|
prev = false;
|
|
|
|
|
|
|
|
return (c);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Safe malloc() for internal use.
|
|
|
|
*/
|
|
|
|
void *
|
|
|
|
grep_malloc(size_t size)
|
|
|
|
{
|
|
|
|
void *ptr;
|
|
|
|
|
|
|
|
if ((ptr = malloc(size)) == NULL)
|
|
|
|
err(2, "malloc");
|
|
|
|
return (ptr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Safe calloc() for internal use.
|
|
|
|
*/
|
|
|
|
void *
|
|
|
|
grep_calloc(size_t nmemb, size_t size)
|
|
|
|
{
|
|
|
|
void *ptr;
|
|
|
|
|
|
|
|
if ((ptr = calloc(nmemb, size)) == NULL)
|
|
|
|
err(2, "calloc");
|
|
|
|
return (ptr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Safe realloc() for internal use.
|
|
|
|
*/
|
|
|
|
void *
|
|
|
|
grep_realloc(void *ptr, size_t size)
|
|
|
|
{
|
|
|
|
|
|
|
|
if ((ptr = realloc(ptr, size)) == NULL)
|
|
|
|
err(2, "realloc");
|
|
|
|
return (ptr);
|
|
|
|
}
|
|
|
|
|
2010-07-29 00:11:14 +00:00
|
|
|
/*
|
|
|
|
* Safe strdup() for internal use.
|
|
|
|
*/
|
|
|
|
char *
|
|
|
|
grep_strdup(const char *str)
|
|
|
|
{
|
|
|
|
char *ret;
|
|
|
|
|
|
|
|
if ((ret = strdup(str)) == NULL)
|
|
|
|
err(2, "strdup");
|
|
|
|
return (ret);
|
|
|
|
}
|
|
|
|
|
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 a matching line according to the command line options.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
printline(struct str *line, int sep, regmatch_t *matches, int m)
|
|
|
|
{
|
|
|
|
size_t a = 0;
|
|
|
|
int i, n = 0;
|
|
|
|
|
2017-04-03 23:16:51 +00:00
|
|
|
/* If matchall, everything matches but don't actually print for -o */
|
|
|
|
if (oflag && 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
|
|
|
if (!hflag) {
|
2011-11-28 20:00:31 +00:00
|
|
|
if (!nullflag) {
|
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
|
|
|
fputs(line->file, stdout);
|
2011-11-28 20:00:31 +00:00
|
|
|
++n;
|
|
|
|
} else {
|
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
|
|
|
printf("%s", line->file);
|
|
|
|
putchar(0);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (nflag) {
|
|
|
|
if (n > 0)
|
|
|
|
putchar(sep);
|
|
|
|
printf("%d", line->line_no);
|
|
|
|
++n;
|
|
|
|
}
|
|
|
|
if (bflag) {
|
|
|
|
if (n > 0)
|
|
|
|
putchar(sep);
|
|
|
|
printf("%lld", (long long)line->off);
|
|
|
|
++n;
|
|
|
|
}
|
|
|
|
if (n)
|
|
|
|
putchar(sep);
|
|
|
|
/* --color and -o */
|
|
|
|
if ((oflag || color) && m > 0) {
|
|
|
|
for (i = 0; i < m; i++) {
|
2017-04-17 14:59:55 +00:00
|
|
|
/* Don't output zero length matches */
|
|
|
|
if (matches[i].rm_so == matches[i].rm_eo)
|
|
|
|
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
|
|
|
if (!oflag)
|
|
|
|
fwrite(line->dat + a, matches[i].rm_so - a, 1,
|
|
|
|
stdout);
|
|
|
|
if (color)
|
2017-04-04 14:17:50 +00:00
|
|
|
fprintf(stdout, "\33[%sm\33[K", 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
|
|
|
|
|
|
|
fwrite(line->dat + matches[i].rm_so,
|
|
|
|
matches[i].rm_eo - matches[i].rm_so, 1,
|
|
|
|
stdout);
|
|
|
|
if (color)
|
2017-04-04 14:17:50 +00:00
|
|
|
fprintf(stdout, "\33[m\33[K");
|
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
|
|
|
a = matches[i].rm_eo;
|
|
|
|
if (oflag)
|
|
|
|
putchar('\n');
|
|
|
|
}
|
|
|
|
if (!oflag) {
|
|
|
|
if (line->len - a > 0)
|
|
|
|
fwrite(line->dat + a, line->len - a, 1, stdout);
|
|
|
|
putchar('\n');
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
fwrite(line->dat, line->len, 1, stdout);
|
2017-04-17 13:14:18 +00:00
|
|
|
putchar(fileeol);
|
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
|
|
|
}
|
|
|
|
}
|