参考在网上找到的代码,没想到相差那么大,应该是grep比perl的模式匹配更快吧。
[gzhy@nearby stat]$ wc -l 1
234033 1
[gzhy@nearby stat]$ perl 1.pl
cost 1 seconds
zjtel : 32606
[gzhy@nearby stat]$ perl 2.pl
cost 111 seconds
zjtel : 32606
2.pl
#!/usr/bin/perl
my $time=time();
open(file,"1");
while(;)
{
chomp;
if(m/:zjtel:/)
{
$zjtel++;
}
}
close(file);
$time=time()-$time;
print "cost $time seconds\n";
print "zjtel : $zjtel\n"; 1.pl
#!/usr/bin/perl
$time=time();
$count=`grep zjtel 1 | wc -l `;
$time=time()-$time;
print "cost $time seconds\n";
print "zjtel : $count\n":
测试:在一个文件夹下有199个纯文本文件,要扫描265个关键字,分别用pattern-match和grep两种方式进行扫描,统计扫描时间。
结果:均查出6354行。
pattern-match:2173 seconds;
grep1: 888 seconds;
grep2: 193 seconds;
参考代码如下:
pattern-match:
use strict;
use File::Basename;
#在一个目录的文件文件中查找包含关键字的 ::
my ($dir,$keywords)= @ARGV;
my @filenames=glob "$dir*";
open KEY,"
my @keywords=;
close KEY;
my $num_key=scalar @keywords;
my @match_lines;
my $time=time();
foreach my $file(@filenames){
eval{
open FILE,"
};
if($@){
print $@;
next;
}
$n=1;
while my $line(){
chomp $line;
foreach my $key(@keywords){
if($line=~m/$key/){
$context="$file:$n:$line\n";
push @match_lines,$context;
}
}$n++;
}
close(file);
}
open RS,">result_file_pattern";
foreach(@match_lines){
print RS $_;
}
close RS;
$time=time()-$time;
print "Patter-match ($num_key keywords) end:$time seconds\n";
grep1: 分别扫每个keywords
use strict;
use File::Basename;
#在一个目录的文件文件中查找包含关键字的 ::
my ($dir,$keywords)= @ARGV;
my @filename=glob "$dir*";
open KEY,"
my @keywords=;
close KEY;
my $num_key=scalar @keywords;
my @match_lines;
my $time=time();
foreach my $file(@filenames){
foreach $key(@keywords){
chomp $key;
next unless ($key);
my $m_keyword = "\\\|\[\^a-z\]$key\$\|\^$key\[\^a-z\]\|\[\^a-z\]$key\[\^a-z\]";
my @sub_match_lines
eval{
@sub_match_lines=`grep -EnriIHs $m_keyword $file` or die "grep Error:$!"
};
if($@}{
print "$@";
next;
}
push @match_lines,@sub_match_lines;
}
}
open RS,">result_file_grep";
foreach(@match_lines){
print RS $_;
}
close RS;
$time=time()-$time;
print "Grep ($num_key keywords) end : $time seconds\n";
//如果直接将$context print到RS句柄和现在这种方式是否有区别?
grep2: 经265个keywords放在一个匹配字符串中扫描
use strict;
use File::Basename;
my ($dir,$keywords)= @ARGV;
my @filenames=glob "$dir*";
my $num=scalar @filenames;
print $num;
open KEY,"
my $all_keys;
my $i=0;
foreach my $key(){
chomp $key;
next unless($key);
my $m_keyword = "\\\|\[\^a-z\]$key\$\|\^$key\[\^a-z\]\|\[\^a-z\]$key\[\^a-z\]";
if($i==0){
$all_keys.="$m_keyword";
$i=1;
}else{
$all_keys.="\|$m_keyword";
}
}
close KEY;
my @match_lines;
my $time1=time();
foreach my $file(@filenames){
chomp $file;
print $file."\n";
my @sub_match_lines;
eval{
my $grep ="grep -EnriIHs \"$all_keys\" $file";
@sub_match_lines=`$grep` or die "$grep Error:$!";
};
if($@){
print $@;
next;
}
push @match_lines,@sub_match_lines;
}
open RS,">result_file_grep2";
foreach(@match_lines){
print RS $_;
}
close RS;
$time=time()-$time;
print "Grep end:$time\n";
File::Basename模块:
File::Basename - Parse file paths into directory, filename and suffix.
File::Basename中常用的方法有fileparse, basename, dirname。
fileparse方法会传回包含路经名称三个部份的串列。
basename方法传回路经位置。
basename方法传回档案名称。
($name,$path,$suffix) = fileparse($fullname,@suffixlist);
my $filename = fileparse("/foo/bar/baz.txt", qr/\Q.txt\E/);
目录句柄操作:
opendir,readdir,closedir
opendir只能返回目录下文件的不带路径的文件名。
opendir(DIRHANDLE,$dir) or die "Can't open $dir:$!";
my @filenames=sort readdir(DIRHANDLE);
closedir(DIRHANDLE);
Glob:
在shell中会将命令行的文件名模式扩展成所有匹配的文件名,这就成为globbing(文件名匹配模式)。
在Perl中通过glob操作符来实现。
my @all_files=glob "dir/*" ;#保存dir目录下所有文件名和目录,除了以点号开头的隐藏文件。
如:目录/file/下有a.txt,b.txt,和C文件夹,D文件夹;则glob "/file/*" ;则结果是/file/a.txt ;/file/b.txt;/file/C;/file/D;
等同于my @all_files=;
my @file=;#表示读取文件内容;
my @file_dirs=;#表示glob;
另外linux grep -r或-R可以递归遍历文件目录下的文件