From 11fefcda40d5c7d0a674441486cbac61a3ba6fc7 Mon Sep 17 00:00:00 2001
From: kevans <kevans@FreeBSD.org>
Date: Wed, 16 Aug 2017 00:47:53 +0000
Subject: bsdgrep: add -z/--null-data support and update NLS catalogs
 accordingly

MFC r317049: bsdgrep: add -z/--null-data support

-z treats input and output data as sequences of lines terminated by a
zero byte instead of a newline. This brings it more in line with GNU grep
and brings us closer to passing the current tests with BSD grep.

MFC r317679: bsdgrep: correct nls usage data after r317049

r317049 added -z/--null-data to BSD grep but missed the update to nls
catalogs.

Approved by:	emaste (mentor, blanket MFC)
Relnotes:	yes
---
 usr.bin/grep/file.c                  |  7 ++++---
 usr.bin/grep/grep.1                  |  9 ++++++---
 usr.bin/grep/grep.c                  | 10 ++++++++--
 usr.bin/grep/grep.h                  |  1 +
 usr.bin/grep/nls/C.msg               |  2 +-
 usr.bin/grep/nls/es_ES.ISO8859-1.msg |  2 +-
 usr.bin/grep/nls/gl_ES.ISO8859-1.msg |  2 +-
 usr.bin/grep/nls/hu_HU.ISO8859-2.msg |  2 +-
 usr.bin/grep/nls/ja_JP.SJIS.msg      |  2 +-
 usr.bin/grep/nls/ja_JP.UTF-8.msg     |  2 +-
 usr.bin/grep/nls/ja_JP.eucJP.msg     |  2 +-
 usr.bin/grep/nls/pt_BR.ISO8859-1.msg |  2 +-
 usr.bin/grep/nls/ru_RU.KOI8-R.msg    |  2 +-
 usr.bin/grep/nls/uk_UA.UTF-8.msg     |  2 +-
 usr.bin/grep/nls/zh_CN.UTF-8.msg     |  2 +-
 usr.bin/grep/util.c                  |  4 ++--
 16 files changed, 32 insertions(+), 21 deletions(-)

(limited to 'usr.bin/grep')

diff --git a/usr.bin/grep/file.c b/usr.bin/grep/file.c
index 34831f1..7ed34d9 100644
--- a/usr.bin/grep/file.c
+++ b/usr.bin/grep/file.c
@@ -197,7 +197,7 @@ grep_fgetln(struct file *f, size_t *lenp)
 	}
 
 	/* Look for a newline in the remaining part of the buffer */
-	if ((p = memchr(bufpos, '\n', bufrem)) != NULL) {
+	if ((p = memchr(bufpos, fileeol, bufrem)) != NULL) {
 		++p; /* advance over newline */
 		ret = bufpos;
 		len = p - bufpos;
@@ -219,7 +219,7 @@ grep_fgetln(struct file *f, size_t *lenp)
 		if (bufrem == 0)
 			/* EOF: return partial line */
 			break;
-		if ((p = memchr(bufpos, '\n', bufrem)) == NULL &&
+		if ((p = memchr(bufpos, fileeol, bufrem)) == NULL &&
 		    filebehave != FILE_MMAP)
 			continue;
 		if (p == NULL) {
@@ -322,7 +322,8 @@ grep_open(const char *path)
 		goto error2;
 
 	/* Check for binary stuff, if necessary */
-	if (binbehave != BINFILE_TEXT && memchr(bufpos, '\0', bufrem) != NULL)
+	if (binbehave != BINFILE_TEXT && fileeol != '\0' &&
+	    memchr(bufpos, '\0', bufrem) != NULL)
 	f->binary = true;
 
 	return (f);
diff --git a/usr.bin/grep/grep.1 b/usr.bin/grep/grep.1
index f06cfaa..f790ce1 100644
--- a/usr.bin/grep/grep.1
+++ b/usr.bin/grep/grep.1
@@ -30,7 +30,7 @@
 .\"
 .\"	@(#)grep.1	8.3 (Berkeley) 4/18/94
 .\"
-.Dd July 28, 2010
+.Dd April 17, 2017
 .Dt GREP 1
 .Os
 .Sh NAME
@@ -40,7 +40,7 @@
 .Sh SYNOPSIS
 .Nm grep
 .Bk -words
-.Op Fl abcdDEFGHhIiJLlmnOopqRSsUVvwxZ
+.Op Fl abcdDEFGHhIiJLlmnOopqRSsUVvwxZz
 .Op Fl A Ar num
 .Op Fl B Ar num
 .Op Fl C Ns Op Ar num
@@ -378,7 +378,10 @@ expression are considered to be matching lines.
 Equivalent to
 .Fl i .
 Obsoleted.
-.It Fl Z , Fl z , Fl Fl decompress
+.It Fl z , Fl Fl null-data
+Treat input and output data as sequences of lines terminated by a
+zero-byte instead of a newline.
+.It Fl Z , Fl Fl decompress
 Force
 .Nm grep
 to behave as
diff --git a/usr.bin/grep/grep.c b/usr.bin/grep/grep.c
index fe8c778..190322c 100644
--- a/usr.bin/grep/grep.c
+++ b/usr.bin/grep/grep.c
@@ -67,7 +67,7 @@ const char	*errstr[] = {
 /* 1*/	"(standard input)",
 /* 2*/	"cannot read bzip2 compressed file",
 /* 3*/	"unknown %s option",
-/* 4*/	"usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A num] [-B num] [-C[num]]\n",
+/* 4*/	"usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n",
 /* 5*/	"\t[-e pattern] [-f file] [--binary-files=value] [--color=when]\n",
 /* 6*/	"\t[--context[=num]] [--directories=action] [--label] [--line-buffered]\n",
 /* 7*/	"\t[--null] [pattern] [file ...]\n",
@@ -110,6 +110,7 @@ bool	 lflag;		/* -l: only show names of files with matches */
 bool	 mflag;		/* -m x: stop reading the files after x matches */
 long long mcount;	/* count for -m */
 long long mlimit;	/* requested value for -m */
+char	 fileeol;	/* indicator for eol */
 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) */
@@ -166,7 +167,7 @@ usage(void)
 	exit(2);
 }
 
-static const char	*optstr = "0123456789A:B:C:D:EFGHIJMLOPSRUVZabcd:e:f:hilm:nopqrsuvwxXy";
+static const char	*optstr = "0123456789A:B:C:D:EFGHIJMLOPSRUVZabcd:e:f:hilm:nopqrsuvwxXyz";
 
 static const struct option long_options[] =
 {
@@ -216,6 +217,7 @@ static const struct option long_options[] =
 	{"word-regexp",		no_argument,		NULL, 'w'},
 	{"line-regexp",		no_argument,		NULL, 'x'},
 	{"xz",			no_argument,		NULL, 'X'},
+	{"null-data",		no_argument,		NULL, 'z'},
 	{"decompress",          no_argument,            NULL, 'Z'},
 	{NULL,			no_argument,		NULL, 0}
 };
@@ -385,6 +387,7 @@ main(int argc, char *argv[])
 	newarg = 1;
 	prevoptind = 1;
 	needpattern = 1;
+	fileeol = '\n';
 
 	eopts = getenv("GREP_OPTIONS");
 
@@ -606,6 +609,9 @@ main(int argc, char *argv[])
 		case 'X':
 			filebehave = FILE_XZ;
 			break;
+		case 'z':
+			fileeol = '\0';
+			break;
 		case 'Z':
 			filebehave = FILE_GZIP;
 			break;
diff --git a/usr.bin/grep/grep.h b/usr.bin/grep/grep.h
index 5613d39..c82e6bb 100644
--- a/usr.bin/grep/grep.h
+++ b/usr.bin/grep/grep.h
@@ -116,6 +116,7 @@ extern bool	 dexclude, dinclude, fexclude, finclude, lbflag, nullflag;
 extern unsigned long long Aflag, Bflag;
 extern long long mcount;
 extern long long mlimit;
+extern char	 fileeol;
 extern char	*label;
 extern const char *color;
 extern int	 binbehave, devbehave, dirbehave, filebehave, grepbehave, linkbehave;
diff --git a/usr.bin/grep/nls/C.msg b/usr.bin/grep/nls/C.msg
index 6327b27..d8a3f08 100644
--- a/usr.bin/grep/nls/C.msg
+++ b/usr.bin/grep/nls/C.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(standard input)"
 2 "cannot read bzip2 compressed file"
 3 "unknown %s option"
-4 "usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A num] [-B num] [-C[num]]\n"
+4 "usage: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n"
 5 "\t[-e pattern] [-f file] [--binary-files=value] [--color=when]\n"
 6 "\t[--context[=num]] [--directories=action] [--label] [--line-buffered]\n"
 7 "\t[--null] [pattern] [file ...]\n"
diff --git a/usr.bin/grep/nls/es_ES.ISO8859-1.msg b/usr.bin/grep/nls/es_ES.ISO8859-1.msg
index 208e91d..8a08079 100644
--- a/usr.bin/grep/nls/es_ES.ISO8859-1.msg
+++ b/usr.bin/grep/nls/es_ES.ISO8859-1.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(entrada est�ndar)"
 2 "no se puede leer el fichero comprimido bzip2"
 3 "opci�n desconocida de %s"
-4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A no] [-B no] [-C[no]]\n"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A no] [-B no] [-C[no]]\n"
 5 "\t[-e pauta] [-f fichero] [--binary-files=valor] [--color=cuando]\n"
 6 "\t[--context[=no]] [--directories=acci�n] [--label] [--line-buffered]\n"
 7 "\t[--null] [pauta] [fichero ...]\n"
diff --git a/usr.bin/grep/nls/gl_ES.ISO8859-1.msg b/usr.bin/grep/nls/gl_ES.ISO8859-1.msg
index 9f3915a..f89532f 100644
--- a/usr.bin/grep/nls/gl_ES.ISO8859-1.msg
+++ b/usr.bin/grep/nls/gl_ES.ISO8859-1.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(entrada est�ndar)"
 2 "non se pode ler o ficheiro comprimido bzip2"
 3 "opci�n desco�ecida de %s"
-4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A no] [-B no] [-C[no]]\n"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A no] [-B no] [-C[no]]\n"
 5 "\t[-e pauta] [-f ficheiro] [--binary-files=valor] [--color=cando]\n"
 6 "\t[--context[=no]] [--directories=acci�n] [--label] [--line-buffered]\n"
 7 "\t[--null] [pauta] [ficheiro ...]\n"
diff --git a/usr.bin/grep/nls/hu_HU.ISO8859-2.msg b/usr.bin/grep/nls/hu_HU.ISO8859-2.msg
index 6d3c1c9..9c8aafd 100644
--- a/usr.bin/grep/nls/hu_HU.ISO8859-2.msg
+++ b/usr.bin/grep/nls/hu_HU.ISO8859-2.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(szabv�nyos bemenet)"
 2 "bzip2 t�m�r�tett f�jl nem olvashat�"
 3 "ismeretlen %s opci�"
-4 "haszn�lat: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A sz�m] [-B sz�m] [-C[sz�m]]\n"
+4 "haszn�lat: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A sz�m] [-B sz�m] [-C[sz�m]]\n"
 5 "\t[-e minta] [-f f�jl] [--binary-files=�rt�k] [--color=mikor]\n"
 6 "\t[--context[=sz�m]] [--directories=m�velet] [--label] [--line-buffered]\n"
 7 "\t[--null] [minta] [f�jl ...]\n"
diff --git a/usr.bin/grep/nls/ja_JP.SJIS.msg b/usr.bin/grep/nls/ja_JP.SJIS.msg
index d18edd5..da0677e 100644
--- a/usr.bin/grep/nls/ja_JP.SJIS.msg
+++ b/usr.bin/grep/nls/ja_JP.SJIS.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(�W������)"
 2 "bzip2 ���k�t�@�C����ǂݍ��ނ��Ƃ��ł��܂���"
 3 "%s �I�v�V�����̎w��l�Ɍ�肪����܂�"
-4 "�g����: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A ����] [-B ����] [-C[����]]\n"
+4 "�g����: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A ����] [-B ����] [-C[����]]\n"
 5 "\t[-e �p�^�[��] [-f �t�@�C����] [--binary-files=�l] [--color=�l]\n"
 6 "\t[--context[=����]] [--directories=����] [--label] [--line-buffered]\n"
 7 "\t[--null] [�p�^�[��] [�t�@�C���� ...]\n"
diff --git a/usr.bin/grep/nls/ja_JP.UTF-8.msg b/usr.bin/grep/nls/ja_JP.UTF-8.msg
index 84c6f45..e842b25 100644
--- a/usr.bin/grep/nls/ja_JP.UTF-8.msg
+++ b/usr.bin/grep/nls/ja_JP.UTF-8.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(標準入力)"
 2 "bzip2 圧縮ファイルを読み込むことができません"
 3 "%s オプションの指定値に誤りがあります"
-4 "使い方: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A 数字] [-B 数字] [-C[数字]]\n"
+4 "使い方: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A 数字] [-B 数字] [-C[数字]]\n"
 5 "\t[-e パターン] [-f ファイル名] [--binary-files=値] [--color=値]\n"
 6 "\t[--context[=数字]] [--directories=動作] [--label] [--line-buffered]\n"
 7 "\t[--null] [パターン] [ファイル名 ...]\n"
diff --git a/usr.bin/grep/nls/ja_JP.eucJP.msg b/usr.bin/grep/nls/ja_JP.eucJP.msg
index e778c5f..922b25a 100644
--- a/usr.bin/grep/nls/ja_JP.eucJP.msg
+++ b/usr.bin/grep/nls/ja_JP.eucJP.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(ɸ������)"
 2 "bzip2 ���̥ե�������ɤ߹��ळ�Ȥ��Ǥ��ޤ���"
 3 "%s ���ץ����λ����ͤ˸��꤬����ޤ�"
-4 "�Ȥ���: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A ����] [-B ����] [-C[����]]\n"
+4 "�Ȥ���: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A ����] [-B ����] [-C[����]]\n"
 5 "\t[-e �ѥ�����] [-f �ե�����̾] [--binary-files=��] [--color=��]\n"
 6 "\t[--context[=����]] [--directories=ư��] [--label] [--line-buffered]\n"
 7 "\t[--null] [�ѥ�����] [�ե�����̾ ...]\n"
diff --git a/usr.bin/grep/nls/pt_BR.ISO8859-1.msg b/usr.bin/grep/nls/pt_BR.ISO8859-1.msg
index 79c026e..c049fde 100644
--- a/usr.bin/grep/nls/pt_BR.ISO8859-1.msg
+++ b/usr.bin/grep/nls/pt_BR.ISO8859-1.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(entrada padr�o)"
 2 "n�o se posso ler o fichero comprimido bzip2"
 3 "opc�o n�o conhecida de %s"
-4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A num] [-B num] [-C[num]]\n"
+4 "uso: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A num] [-B num] [-C[num]]\n"
 5 "\t[-e padr�o] [-f arquivo] [--binary-files=valor] [--color=quando]\n"
 6 "\t[--context[=num]] [--directories=a��o] [--label] [--line-buffered]\n"
 7 "\t[--null] [padr�o] [arquivo ...]\n"
diff --git a/usr.bin/grep/nls/ru_RU.KOI8-R.msg b/usr.bin/grep/nls/ru_RU.KOI8-R.msg
index 6880d51..d6d308f 100644
--- a/usr.bin/grep/nls/ru_RU.KOI8-R.msg
+++ b/usr.bin/grep/nls/ru_RU.KOI8-R.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(����������� ����)"
 2 "�� ���� ��������� ������ � bzip2 ����"
 3 "����������� ���� %s"
-4 "�������������: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A ���] [-B ���] [-C[���]]\n"
+4 "�������������: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A ���] [-B ���] [-C[���]]\n"
 5 "\t[-e ������] [-f ����] [--binary-files=��������] [--color=�����]\n"
 6 "\t[--context[=���]] [--directories=��������] [--label] [--line-buffered]\n"
 7 "\t[--null] [������] [���� ...]\n"
diff --git a/usr.bin/grep/nls/uk_UA.UTF-8.msg b/usr.bin/grep/nls/uk_UA.UTF-8.msg
index 4492d31..e601ca2 100644
--- a/usr.bin/grep/nls/uk_UA.UTF-8.msg
+++ b/usr.bin/grep/nls/uk_UA.UTF-8.msg
@@ -4,7 +4,7 @@ $quote "
 1 "(стандартний ввід)"
 2 "не можу прочитати стиснутий bzip2 файл"
 3 "невiдома опція %s"
-4 "використання: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A чис] [-B чис] [-C[чис]]\n"
+4 "використання: %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A чис] [-B чис] [-C[чис]]\n"
 5 "\t[-e шаблон] [-f файл] [--binary-files=значення] [--color=коли]\n"
 6 "\t[--context[=чис] [--directories=дія] [--label] [--line-buffered]\n"
 7 "\t[--null] [шаблон] [файл ...]\n"
diff --git a/usr.bin/grep/nls/zh_CN.UTF-8.msg b/usr.bin/grep/nls/zh_CN.UTF-8.msg
index d5cd0c2..921ff4b 100644
--- a/usr.bin/grep/nls/zh_CN.UTF-8.msg
+++ b/usr.bin/grep/nls/zh_CN.UTF-8.msg
@@ -5,7 +5,7 @@ $quote "
 1 "(标准输入)"
 2 "读取 bzip2 压缩文件时出错"
 3 "选项 %s 无法识别"
-4 "用法： %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZ] [-A 行数] [-B 行数] [-C[行数]]\n"
+4 "用法： %s [-abcDEFGHhIiJLlmnOoPqRSsUVvwxZz] [-A 行数] [-B 行数] [-C[行数]]\n"
 5 "\t[-e 模式] [-f 文件] [--binary-files=值] [--color=何时]\n"
 6 "\t[--context[=行数]] [--directories=动作] [--label] [--line-buffered]\n"
 7 "\t[--null] [模式] [文件名 ...]\n"
diff --git a/usr.bin/grep/util.c b/usr.bin/grep/util.c
index 46b1dd8..4ab7b06 100644
--- a/usr.bin/grep/util.c
+++ b/usr.bin/grep/util.c
@@ -216,7 +216,7 @@ procfile(const char *fn)
 			else
 				break;
 		}
-		if (ln.len > 0 && ln.dat[ln.len - 1] == '\n')
+		if (ln.len > 0 && ln.dat[ln.len - 1] == fileeol)
 			--ln.len;
 		ln.line_no++;
 
@@ -525,6 +525,6 @@ printline(struct str *line, int sep, regmatch_t *matches, int m)
 		}
 	} else {
 		fwrite(line->dat, line->len, 1, stdout);
-		putchar('\n');
+		putchar(fileeol);
 	}
 }
-- 
cgit v1.1