Perl学习笔记（2）——模式匹配和正则表达式

一、模式匹配

模式是指在字符串中寻找特定序列的字符或者说是一个 匹配（或不匹配）某字符串的模板，模式由反斜线包含。而一个** 正则表达式**，就是 用某种模式去匹配一类字符串的一个公式。

要进行复杂的模式，就需要借助于元字符和量词。

1.1模式匹配的特殊字符

字符	作用
.	匹配任何单个字符（换行符\n除外）
\	转义字符，用于特殊字符前，使其失去特殊字符的作用变成普通字符
+	匹配该字符前而的字符至少一次
*	匹配该字符前面的字符任意多次（包括0次）
？	匹配该字符前而的字符0次或者1次
.*	匹配任意字符（换行符除外）任意多次
{count}	匹配前而的字符count次
{min}	匹配前面的字符至少min次
{min，max}	匹配前而的字符至少min次，但最多不超过max次
*？	匹配前而的字符0次或更多次
+？	匹配前而的字符一次或者更多次
{min,}?	匹配前扁的字符0次或者至少min次
{min,max}？	匹配前而的字符至少min次，但不超过max次
()	模式分组字符．eg:/(perl)+/会匹配"perlperl" -------1)借助模式分组，可以使用”\n”形式的方向引用来引用园括号中匹配的内容，n和括号的组号匹配;2）Perl5.10可以使用\g{n}这种形式的反向引用，n可以为负数 eg:$_="abba”；print "Matched” if(/(.)\1/);
\	或，即’1"符号的左边匹配或右边匹配都可以 eg：/ perl丨Perl | PERL/

1.2 字符集

字符集是指一串可能出现的字符集合。通过写在方括号“[ ]”内来表示。只用来匹配单个字，但可以是方括号中列出的字符中的任意一个。

• 可以使用连字符“-”进行简写
• 使用脱字符“^”匹配字符集之外的字符

字符集的简写：

简写	等价字符集	含义
\d	[0-9]	匹配任意数字：
\w	[A-Za-z0-9_]	匹配任意字母数字下划线
\s	[\f\t\n\r]	匹配空白 （\f:换页符，\t：制表符，\n：换行符，\r：回车）
\h	[\t]	匹配横向空白
\v	[\f\n\r]	匹配纵向空格
\R		匹配任何类型的断行
\d	\D	[^\d]
\w	\W	[^\w]
\s	\S	[^\s]

下面会使用到的一些字符集：
[\dA-Fa-f] :匹配16进制数；
[\d\D]:匹配所有，包括数字，和数字意外部分；
[^\d\D]:匹配除数字外的所有空集除外；

1.3 模式匹配操作符

使用双斜线（//）来进行匹配是m//（模式匹配）操作符的简写。

模式匹配操作符中可以使用任何成对的定界符

m(fred),m,m{fred},m[fred]
/http:VV/
m%http://%

当使用斜线作为定界符时，开头的m可以省略

• /i
  使用／i修饰符，在进行模式匹配时，将不区分大小写
• /s
  使用/s修饰符时，模式中的点号(.）将匹配包括换行
  符在内的任意字符．
• /x
  使用／x修饰符可以在模式里而随意加上空白，使其
  更容易阅读．此时如果还需要匹配空白与制表符，需要
  在前而补一个反斜线过行转义·
  eg：下曲两个式是等价的

/-?\d+\.?\d*/
/-? \d+ \.? \d* /x

• 多个匹配修饰符组合使用
  例子

#!/usr/bin/perl -w

use strict;

#-----------------------------------------------------
# The example used to show the usage of option mdefiers
# to do the expression matching
#

print "\n Would you like to play a game?\n\n";
chomp ($_ = );
if (/yes/i) {
	print "In that case, I recommend that you go bowling.\n\n";
}

$_ = "I saw Tom\ndown at the bowling alley\nwith Fred\nlast night.\n";
if (/Tom.*Fred/s) {
	print "Matched successfully when use \"\\S\" option \n\n";
}

if (/tom.*fred/si) {
	print "That string mentions Fred after Barney! \n\n";
}

结果：

1.4 锚位

默认情况下，模式匹配的过程开始于待匹配字符串的开头，如果不相符就一直往字符串后面浮动，看其他位置能否匹配但是加人一些锚位，可以让模式直接匹配字符串的某处。

锚位	作用
^	标志字符串的开头
$	标志字符串的结尾
/b	单词（\w+）锚位。匹配任何单词的收尾
/B	非单词锚位。能匹配所有\b不能匹配的位置

例子：

#!/user/bin/perl -w

use strict;

# ----------------------------------------------
# THe example used to show the useage of anchros
# in Perl

$_ = "it is just a test for the useage of anchros in Perl.just a test";
if (/^just/){
	print "\n Match the \"just\"in the head successfully!\n\n";
}
else {
	print "\n Failed to match the \"just\"in the head!\n\n";
}

if (/^it.*test$/){
	print "\n Matched a string starts with \"it\"and end with\"test\"\n\n";
}

$_ = "called fred and so ...";
if (/\bfred\b/) {
	print "\n[1] Matched \"fred\" word successfully!\n\n";
}

$_ = "frederick";
if (/\bfred\b/) {
	print "\n[2] Matched \"fred\" word successfully!\n\n";
}
else {
	print "\n[2] Failed to match \"fred\" word!\n\n";
}

结果：

1.5 绑定操作符 =~

默认情况下，模式匹配的对象是$_。绑定操作符=~用
于告诉Perl拿右边的模式来匹配左边的字符串，而不是匹配$_变量中的字符串。
例子：

#!/usr/bin/perl -w

use strict;

#----------------------------------------------
# The example used to show how to use =~ operator
# in perl languge to do expression matching

$_ = "just for test";
print "\n The value for the default variable \"\$_\" is \"$_\" \n";
if (/^just.*TEST$/i) {
	print "\n[1] Matched a string starts with \"just\" and end with \"test\"\n\n";
}
else {
	print "\n[1] Failed to match a string starts with \"just\" and end with \"test\"\n\n";
}
my $val = "just test it";
print "the specified variable val = \"$val\" \n";
if ($val =~ /^just.*it$/) {
	print "\n[1] Matched a string starts with \"just\" and end with \"it\"\n\n";
}
else {
	print "\n[1] Failed to match a string starts with \"just\" and end with \"it\"\n\n";
}

结果：

1.6 模式串中的变量内插

在Perl正则表达式中可以进行双引号形式的变量内插。
例子：

#!/usr/bin/perl -w

use strict;

my $val = "just";
my $val2 = "This is just a test example";
if ($val2 =~ /($val)/) {
	print "\n \"$val2\" matched with \"$val\"!\n\n";
}
else {
	print "\n \"$val2\" failed to match with \"$val\"!\n\n";
}

1.7 捕获变量

在模式中使用圆括号，可以启动正则表达式处理引擎的捕获功能（圆括号也可以用来对模式进行分组）。捕获功能指的是，把（圆括号中模式所匹配的）部分字符串暂时记忆下来的能力。如果有多对圆括号，就是有多个捕获。每个被捕获的对象是原本的字符串，不是模式。

• 捕获变量
  捕获变量都是标量变量，它们的名字依次是$1，$2,…,$n模式里的园括号有多少个，匹配变量就有多少个。

• 捕获变量的生命周期
  。捕获变量通常能存活到下次成功的模式匹配为止，即失败的匹配不会改动上次工程匹配时捕获的内容，而成功的匹配会将它们配置。
  。捕获变量只应该在匹配成功时使用，否则为之前一次模式匹配的内容。
  。如果需要在数行之外使用捕获变量，最好先将捕获变量的值复制到一个一般的变量里。

例子

#!/usr/bin/perl -w

use strict;

#--------------------------------------------
# The example used to show how to use the 
# "match the variables"
#

my $val = "hello there, neighbor";
my $val2;
if ($val =~ /(\w+)\s*(\w+),\s(\w+)/) {
	print "\n v1 = $1, vs = $2 , v3 = $3 \n\n";
	$val2 = $2;
}

$val = "I fear that I will be extinct after 1000 years.";
 
if ( $val =~ /(\d+)\s*(\w+)/ ) {
	print "[2] v1 = $1 , v2 = $2 \n\n";
}

print "\n After some operations \n\n";
print "The obtained value is \"$val2\" \n";

结果：

• 不匹配模式
  。一般情况下，模式使用的圆括号都会捕获部分的匹配串到捕获变量中，并且Perl按照左圆括号的序号来决定捕获变量名。
  。Perl正则表达式允许使用括号但不做捕获（不捕获括号）。要使用不捕获括号，需要在左括号的后而加上问号和冒号（？：），以告诉Per|这一对括号完全是为了分组而存在的。

my $val = "This is just a simple example";
if ($val =~ /^this\s+is\s+(?:just)\s+(.*)$/i) {
	print "\nThe matched value is \"$1\"\n\n";
}

• 命名捕获

管理$1，$2等这些数字型的捕获变量是比较困难的，尤其对于比较复杂的正则表达式而言。

为了不必记忆$1等这些数字型捕获变量的含义，Perl5.10引入了正则表达式命名捕获的概念。

在命名捕获中，捕获的结果会存入一个特殊的哈希%+中，其中的键就是在捕获时使用的特殊标签，其中的值则是被捕获的串。为捕获串加标签的方法是使用(?PATTERN)这样的写法，此时使用捕获串时需要访问的位置变成了$+{LABEL}，在使用了命名捕获后，可以使用\g(LABEL)或\k< LABEL> 这样的形式来使用反向引用。

Note:使用捕获标签后，就可以随意移动位置并加入更多的捕获括号，不会因为括号的次序变化导致麻烦。
例子:

#!/usr/bin/perl -w

use strict;

#-----------------------------------------------------
# The example used to show the "Name Capture" in perl
#

use 5.010;
my $name = 'Fred or Barney';
if ($name =~ m/(?\w+) (?:and|or) (?\w+)/ ) {
	print "\nI saw $+{name1} and $+{name2} \n\n";
}

$name = 'Fred Flinstone and wilam Flinstone';
if ($name =~ m/(?\w+) and \w+ \g{last_name}/ ) {
	print "\nI saw $+{last_name} \n\n";
}

• 自动匹配变量
  Perl正则匹配中，存在3个自动匹配变量：$&，$`，$’
  $&
  字符串里实际匹配模式的部分会被自动存进$&变量中
  $`
  匹配起始位置之前的字符串会存到$`变量中，或者说$`变量中保存了正则表达式引擎在找到匹配段落之前略过的部分
  $’
  匹配结束位置之后的字符串会存放到$'变量中，或者说$'保存了字符串中剩下的、从来没有匹配到的部分

if ("Hello there, neighbor" =~ /\s(\w+),/ ) {
	print "\nThat was ($`) ($&) ($') .\n\n";
}

二、用正则表达式进行文本处理

2.1 文本替换

1、用 “s/// ”替换：

• s///操作符：模式替换操作符s///会将指定变量合乎模式的那个部分替换为另一个串。s///替换返回的是布尔值，替换成功时为真，否则为假。
• 使用/g修饰符进行全局替换:即使有其他可以替换的部分，s///也只会进行一次替换。使用/g修饰符可以让s///进行所有可能的、不重复的替换。

$_ = "He's out bowling with Barney tonight";
print "\n[1] Before replacement and val = \"$_\"\n";
s/Barney/Fred/;
print "\n[1] After replacement and val = \"$_\"\n";

s/TOnight/this afternoon/;
print "\n[2] After replacement and val = \"$_\"\n";

$_ = "home, sweet home!";
print "\n[3] Before replacement and val = \"$_\"\n";
s/home/cave/;
print "\n[3] After replacement and val = \"$_\"\n";

$_ = "home, sweet home!";
print "\n[4] Before replacement and val = \"$_\"\n";
s/home/cave/g;
print "\n[4] After replacement and val = \"$_\"\n";


$_ = " home, sweet home! ";
print "\n[5] Before replacement and val = \"$_\"\n";
s/^\s*|\s*$//g;
print "\n[5] After replacement and val = \"$_\"\n";

• 不同的替换定界符
  和m//操作符一样，s///操作符也可以使用其他的定界符。
  1、使用没有左右之分的非成对符号跟使用斜线一样，只需要重复3次即可
  eg:s#^https://#http://#;
  2、使用有左右之分的成对符号
  必须使用两对，一对圈引模式，一对圈引替换字符串。在这种情况下，圈引模式的定界符和圈引替换字符串的定界符不必相同。
  eg:s{fred}{barney}; s[fred](barney); s#barney#;
• 替换修饰符
  模式匹配用到的修饰符都可以在替换操作时使用：/g,/i,/x,/s （匹配任意字符，使点号可匹配换行符）另外/m修饰符可以让模式也匹配字符串内的换行符

openFILE,$filename or die "Can't open '$filename'：$";
my $lines = join",;
$lines =～ s/^/\$filename:/gm;

• 绑定操作符
  模式匹配时使用的绑定操作符=~同样适用于替换操作

• 大小写转换：

操作符	作用
\U	该转义字符会将其后的所有字符转换成大写
\L	该转义字符将其后的所有字符转换成小写
\E	该转义字符用于结束\U和\L对大小写转换的影响
\u,\l	也进行大小写转换，但只会影响其后的第一个字符

注意：\U,\L,\u,\l同样适用于任何双引号内的字符串。

#!/usr/bin/perl -w

use strict;

#-----------------------------------------------------
# The example used to show the following items in
# perl:
# 	1) Different Delimiters for repalcement
# 	2) Option modefiers for replacement
# 	3) The Binding operaor for replacement
# 	4) Case shifting operation  
#

# 1) Different Delimiters for repalcement
$_ = "today is 12th Jan. 2014";
print "\n[1] Before replacement and val = \"$_\"\n";
s/today/tommorw/;
print "\n[1] After replacement and val = \"$_\"\n";

$_ = "today is 12th Jan. 2014";
print "\n[2] Before replacement and val = \"$_\"\n";
s#today#tommorw#;
print "\n[2] After replacement and val = \"$_\"\n";


$_ = "today is 12th Jan. 2014";
print "\n[3] Before replacement and val = \"$_\"\n";
s[today]{tommorw};
print "\n[3] After replacement and val = \"$_\"\n";

# 2) Option modefiers for replacement
$_ = "TODAY is 12th Jan. 2014";
print "\n[4] Before replacement and val = \"$_\"\n";
s/today/tommorw/;
print "\n[4] After replacement and val = \"$_\"\n";

s/today/tommorw/i;
print "\n[5] After replacement and val = \"$_\"\n";

$_ = "The following is the final result:\n- status:pass \n- __END__ \n\n Simulation finished\n";
print "\n[6] Before replacement and val = \"$_\"\n";
s/.*(__END__).*/$1/;
print "\n[6] After replacement and val = \"$_\"\n";

$_ = "The following is the final result:\n- status:pass \n- __END__ \n\n Simulation finished\n";
print "\n[7] Before replacement and val = \"$_\"\n";
s/.*(__END__).*/$1/s;
print "\n[7] After replacement and val = \"$_\"\n";

# 3) The Binding operaor for replacement
$_ = "today is 12th Jan. 2014";
my $val = "TODAY is 12th jan. 2014";
print "[8] before replacement:	\n";
print "				\$_ = \"$_\"\n";
print "				\$val = \"$val\"\n";
$val =~ s/today/tommorw/i;
print "[8] after replacement:	\n";
print "				\$_ = \"$_\"\n";
print "				\$val = \"$val\"\n";

# 4) Case shifting operation  

$_ = "I saw Barney with Fred.\n";
print "\n[9] Before replacement and val = \"$_\"\n";
s/(fred|barney)/\U$1/gi;
print "\n[10] After replacement and val = \"$_\"\n";

s/(fred|barney)/\L$1/gi;
print "\n[11] After replacement and val = \"$_\"\n";

s/(\w+) with (\w+)/\U$2\E with $1/i;
print "\n[12] After replacement and val = \"$_\"\n";

s/(fred|barney)/\u$1/ig;
print "\n[13] After replacement and val = \"$_\"\n";

s/(fred|barney)/\u\L$1/ig;
print "\n[14] After replacement and val = \"$_\"\n";

2.2 split操作符

• 作用
  根据指定的分隔符拆开一个字符串，返回子串的列表（即将一个字符串分解为数个片段）。尤其适用于处理被制表符、冒号、空白或其他符号分隔的数据
• 语法规则
  @list=split/separator/,$string or@list=split(/separator/,$string)
  应该注意的是，sp|it会保留开头处的空字段，并省略结尾处的空字段（如[3]）。

#!/usr/bin/perl -w

use strict;

#-----------------------------------------------------
# The example used to show the useage of split operator
# in perl
# 

my @fileds = split /:/,"abc:def:g:h";
print "\n[1] obtained list = @fileds \n\n";

@fileds = split /:/,"abc:def::g:h";
print "\n[2] obtained list = @fileds \n\n";

@fileds = split /:/,":::a:b:c:::";
print "\n[3] obtained list = @fileds \n\n";
my $num = @fileds;
print "\n[3'] number of list  = $num \n\n";


my $some_input = "This is a \t test.\n";
my @args = split(/\s+/,$some_input);
print "\n[4] obtained list = @args \n\n";

$_ = "This is a \t test.\n";
@fileds = split;
print "\n[5] obtained list = @fileds \n\n";

2.3 join函数

• 作用
  join函数用于将数个字符串片段联合成一个字符串
• 语法规则
  my $result = join $glue, @pieces; or my $result = join($glue, @pieces);

my $x = join ":",4,6,8,10,12;
print "\n[1] after join and the value = \"$x\"\n\n";

my @values = split /:/,$x;
print "\n[2] after split and the value = \"@values\"\n\n";

my $z = join "-",@values;
print "\n[3] after join and the value = \"$z\"\n\n";

2.4 列表上下文中的m//

一般情况下，模式匹配操作符m//返回的是布尔值。如果在列表上下文中使用m//，模式匹配成功时返回的是所有捕获变量的列表，匹配失败时返回的是空列表。另外，/g修饰符也可以用在m//操作符上。

my $test = "Hello there, neighbor!";
my ($first,$second,$third) = ($test =~ /(\S+) (\S+), (\S+)/);
print "\nfirst = \"$first\" ,second = \"$second\" ,third = \"$third\"\n\n";

my $test = "Fred dropped a 5 ton granite block on Mr.Slate";
my @words = ($test =~ /([a-z]+)/ig);
print "Result : \"@words\"\n";

• 贪婪量词
  之前提到的+，*，？，{m，n}或{m，}这些模式匹配时会用到的量词都是贪婪量词，即在保证整体匹配的前提下，会尽量匹配长字符串．
• 非贪婪量词
  对于每个贪婪量i司，都有一个对应的非贪婪量同．在贪婪量词后而加上一个问号得到的就是非贪婪量词:+?，*?，??，{m，n}?，{m，}?
  相比较于贪婪量词，非贪婪量词匹配的字符串是越短越好，而不再是越长越好。例如，？？依然是匹配零次或一次，但会优先考虑零次的情况。

三、在命令行执行Perl

eg：
perl -p -i.bak -w -e 's/Randall/Randal/g' fred*.dat

设置/选项	作用
-p	让Perl自动生成一段小程序：while（<>）{print:}
-i.bak	告诉Perl处理之前先将原文件做一个备份，备份文件的后缀名为“.bak”
-w	打开warning
-e	告诉Perl后面接着的是程序代码

四、模式匹配和正则表达式的例子

• 例子描述：
  有几十个文件，里而有部门、电话、名字、时间等信息，要求将名字做相应的修改．时间改成今天的日期，电话信息删除。

4.1 产生这样例子的perl

#!/usr/bin/perl -w
use strict;
use POSIX;
use Getopt::Long;
# ----------------------------------------------------
# Filename 	: gen_ext4.18_logs.pl
#
# Description	:
# 	the file use to generate the log files for
# 	example ext4.18.pl
#
# ---------------------------------------------------

my $file_num = 10;	# how many log file will be generated
my $help = 0;
my $debug = 0;
my @companys 	= qw(vimcrc versilcon amd marvell amlogic mtk qualcomm spreadtrum CSR Parade);
my @names 	= qw(peter david tony allen kurt sam mark brance steven);
my @departs 	= qw(analog digital SoCI SoCII SoCIII video Audio Bluetooth WIFI PR CPU GPU PCIE);
my $name_num 	= @names + 0;
my $company_num = @companys + 0;
my $depart_num	= @departs + 0;
my $tab = ""x4;
GetOptions(
	'file_num=s'	=> \$file_num,
	'help!'		=> \$help,
	'debug!'	=> \$debug,
);

&help_message() if $help;

if (defined $file_num && $file_num > 0 ) {
	if (! -e "./log") {
		system ("mkdir log");
	}
	#use the for loop to generate the specified number of log files
	for (my $i = 0; $i < $file_num; $i++) {
	my $cur_file = "./log/info_${i}.log";
	my $str = "";
	open (LOG,">",$cur_file) or die "Can not open $cur_file file for writing!\n";
	my $name_index = int (rand($name_num));
	my $company_index = int (rand($company_num));
	my $depart_index = int (rand($depart_num));
	my $num = int(rand(86400));
	my $date = getTime(time() - 86400*$i - $num);
	my $month = $date->{month};
	my $day = $date->{day};
	my $year = $date->{year};
	my $time = "${month}/${day}/${year}";
	$str .= "program name 	: program_${i}\n";
	$str .= "Author 		: $names[$name_index]\n";
	$str .= "Company 	: $companys[$company_index]\n";
	$str .= "Department 	: $departs[$depart_index]\n";
	$str .= "Phone	 	: +86 32 1345 123${i}\n";
	$str .= "Date 		: $time \n";
	$str .= "Version		: ${i}.1 \n";
	$str .= "Size		: ${i}k \n";
	$str .= "Status		: Final beta_${i} \n";
	print LOG $str;
	print "INFO -- the $cur_file has been generated \n\n" if $debug;
	close (LOG);
	}
	print "\n -- All the requested log file have been generated and saved into ./log dir! -- \n\n";
}

sub getTime {
	my $time = shift || time();
	my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime ($time);
	
	$year += 1900;
	$mon ++;
	
	$min = '0'.$min if length($min) <2;
	$sec = '0'.$sec if length($sec) <2;
	$mon = '0'.$mon if length($mon) <2;
	$mday = '0'.$mday if length($mday) <2;
	$hour = '0'.$hour if length($hour) <2;

	my $weekday = ('sun','Mon','Tue','Wed','Thu','Fri','Sat') [$wday];

	return {
		'second'=> $sec,
		'minute'=> $min,
		'hour'	=> $hour,
		'day'	=> $mday,
		'month'	=> $mon,
		'year'	=> $year,
		'weekNO'=> $wday,
		'wday'	=> $weekday,
		'date'	=> "$year-$mon-$mday"
	};
}

sub help_message {
	print "\n$0 used to gegerate the log file for example to deal with\n\n";
	print "Usage : 	perl $0 -file_num file_num [-debug]\n";
	print "	     or prtl $0 -help/-h \n\n";
	exit;
}

4.2 例子所要求文本处理的Perl

#!/usr/bin/perl -w
use strict;
# ----------------------------------------------------
# Filename 	: ext4.18.pl
#
# Description	:
# 	
#
# ---------------------------------------------------
my $log_dir;
my @log_files;
if (@ARGV > 0) {
	$log_dir = $ARGV[0];
}
else {
	&help_message();
}
$log_dir =~ s#\|/$##;		#remove "/" or "\" in the last position
@log_files = glob"$log_dir/*.log";
print "[INFO] -- obtain the log files done!\n\n";
print "log_files = @log_files \n\n";
if (defined $log_files[0] && $log_files[0] !~ /^\s*$/) {
	print "[INFO] -- Start to process the obtained log files \n\n";
	
	foreach my $file (@log_files) {
	# -------------------------------------------------------------
	# the following is the example code in the book
	# but can not obtain the expected result in the
	# windows environment.Not clear for the reason.
	# And so use another way to reaslze the function
	# -------------------------------------------------------------
	# chomp(my $date = `date`);
	# $^I = ".bak";
	# while (<$file>) {
	#	s/^Author:.*/Author: Radal L.Schwartz/;
	#	s/^Phone:.*\n//;
	#	s/^Date:.*/Date:$date/;
	#	print;
	# }
	#
	# --------------------------------------------------------------
	# the following is anothor common way to reaszie the function
	
	system ("cp $file ${file}.bak");
	my $tmp_file = "${file}.tmp";
	my $str = "";
	chomp (my $date = `date`);
	open (IN_LINE,"<",$file) or die "Can not open $file for reading!\n";
	while () {
		s/^Author\s*:.*/Author		: Randal L. Schwartz/;
		s/^Phone\s*:.*//;
		s/^Date\s*:.*/Date		: $date/;
		$str .= $_ if ($_ !~ /^s*&/)
	}
	close(IN_LINE);
	open (LOG,">",$tmp_file) or die "Can not open $tmp_file for writing!\n";
	print LOG $str;
	system ("mv $tmp_file $file");
	}
	print "[INFO] -- Complete to process the obtained log files \n\n";
}

sub help_message () {
	print "Usage : perl $0 log_dir\n\n";
	exit;
}