請(qǐng)教高手能不能給我解釋一下這個(gè)腳本

Iinvincible

root@root$ vmstat 2 3
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
2  0 213728  87600 1183808 849440    0    0    86    85    2    0  3  3 92  2  0
0  0 213728  88864 1183808 849460    0    0     0     0 1022  372  5  3 93  0  0
0  0 213728  88740 1183808 849472    0    0     0     2 1014  285  0  0 100  0  0



use strict;
use RRD::Simple;

my $cmd = "/usr/bin/vmstat 2 3";
my $rrdfile = "vmstat-cpu.rrd";
my $rrd = RRD::Simple->new( file => $rrdfile );

my @keys = ();
my %update = ();
open(PH,"-|",$cmd) or die qq{Unable to open file handle PH for command "$cmd": $!};
while (local $_ = <PH>) {
     next if /---/;
     s/^\s+|\s+$//g;
     if (/\d+/ && @keys) {
         @update{@keys} = split(/\s+/,$_);
     } else { @keys = split(/\s+/,$_); }
}
close(PH) or die qq{Unable to close file handle PH for command "$cmd": $!};

my @cpukeys = splice(@keys,-4,4);
my %labels = (wa => "IO wait", id => "Idle", sy => "System", us => "User");

$rrd->create(map { ($_ => "GAUGE") } @cpukeys) unless -f $rrdfile;
$rrd->update(map { ($_ => $update{$_}) } @cpukeys);
我知道這個(gè)腳本大概就是去vmstat cpu部分的數(shù)據(jù)然后畫圖，但是看不懂加粗的部分，尤其是紅色的部分。請(qǐng)高手幫我講解一下。謝謝啦！

w123456_cu

@update{@keys}   ??

Iinvincible

回復(fù) 2# w123456_cu


    是的，   
if (/\d+/ && @keys) {
         @update{@keys} = split(/\s+/,$_);
     } else { @keys = split(/\s+/,$_); }

stanley_tam

1. @update{@keys}是hash切片，跟數(shù)組切片類似，如@array[0 .. $index]
   
2. 
   
3. 建議你用Data::Dumper模塊把數(shù)據(jù)打印出來(lái)，就一目了然了，如
   
4. 
   
5. use strict;
   
6. # use RRD::Simple;
   
7. 
   
8. use Data::Dumper;
   
9. 
   
10. 
    
11. my $cmd = "/usr/bin/vmstat 2 3";
    
12. # my $rrdfile = "vmstat-cpu.rrd";
    
13. # my $rrd = RRD::Simple->new( file => $rrdfile );
    
14. 
    
15. my @keys = ();
    
16. my %update = ();
    
17. open(PH,"-|",$cmd) or die qq{Unable to open file handle PH for command "$cmd": $!};
    
18. while (local $_ = <PH>) {
    
19.      next if /---/;
    
20.      s/^\s+|\s+$//g;
    
21.      if (/\d+/ && @keys) {
    
22.          @update{@keys} = split(/\s+/,$_);
    
23.      } else { @keys = split(/\s+/,$_); }
    
24. }
    
25. close(PH) or die qq{Unable to close file handle PH for command "$cmd": $!};
    
26. 
    
27. my @cpukeys = splice(@keys,-4,4);
    
28. 
    
29. print Dumper \@keys;
    
30. print Dumper \%update;
    
31. print Dumper \@cpukeys;
    
32. 
    
33. __END__
    
34. 
    
35. my %labels = (wa => "IO wait", id => "Idle", sy => "System", us => "User");
    
36. 
    
37. $rrd->create(map { ($_ => "GAUGE") } @cpukeys) unless -f $rrdfile;
    
38. $rrd->update(map { ($_ => $update{$_}) } @cpukeys);

復(fù)制代碼

回復(fù) 1# Iinvincible


   

xiumu2280

這個(gè)就是哈希切片的用法回復(fù) 3# Iinvincible


   

w123456_cu

For example, consider the bowling scores set individually:

1. $score{"fred"} = 205;
   
2. $score{"barney"} = 195;
   
3. $score{"dino"} = 30;

復(fù)制代碼

This seems rather redundant, and in fact can be shortened to:

1. ($score{"fred"},$score{"barney"},$score{"dino"}) =
   
2.     (205,195,30);

復(fù)制代碼

But even these seems redundant. Let's use a hash slice :

1. @score{"fred","barney","dino"} = (205,195,30);

復(fù)制代碼

There. Much shorter. We can use a hash slice with variable interpolation as well:

1. @players = qw(fred barney dino);
   
2. print "scores are: @score{@players}\n";

復(fù)制代碼

Iinvincible

回復(fù) 6# w123456_cu

多謝！通過(guò)列子看起來(lái)容易些了。

但是這段腳本：
@players = qw(fred barney dino);
print "scores are: @score{@players}\n";

運(yùn)行后打印結(jié)果為：

score are：
沒打印出來(lái)啊


   

Iinvincible

回復(fù) 4# stanley_tam


    多謝啊！明白點(diǎn)了，又學(xué)了一招！

但我不明白 if (/\d+/ && @keys) 這是什么意思。d+代表匹配數(shù)字，@key是什么意思呢？

pitonas

強(qiáng)！4,6 樓 很贊啊

pitonas

請(qǐng)高手幫我講解一下。

1. while (local $_ = <PH>) {
   

復(fù)制代碼

1. while (<PH>) {

復(fù)制代碼

謝謝啦！@stanley_tam@w123456_cu