PERL脚本使用CSV文件比较不同主机中从今天到昨天的分区使用情况

Question

我是perl的新手，正在努力构建我的第一个脚本但是，分区模块不起作用。更多细节在代码中进行了评论。也许代码应该使用哈希和引用重写，但我不知道如何做到这一点。有人可以帮帮我吗？

#!/usr/bin/perl -w

#This script compares the today's result and yesterday's results between partitions for each host.
#The "df" output is stored in a CSV file. If any partitions have been changed (mounted/unmounted)
#the script must to warn and don't compare the results for that host.

use warnings;

my $file = "/tmp/df.csv";
my $host = `hostname | awk -F'.' '{print \$1}'`;
my $yesterdays_date = `date -d "-1 day" '+%d.%m.%Y'`;
my $todays_date = `date '+%d.%m.%Y'`;
chomp ($host, $yesterdays_date, $todays_date);

open(HANDLE, "$file") or die "Error opening file $file: $!";
my @array = <HANDLE>;

foreach (@array){
     @columns = split /,/;

        if(/$host/ and /$todays_date/){
            my $todays_result = $columns[5];chomp;
            #my $todays_partition = $columns[6];chomp;

            print "Today\'s disk usage on $host: $columns[5]\n";
            #print "$todays_result <<T.Result      T.Partition>> $todays_partition"
        }

        elsif(/$host/ and /$yesterdays_date/){
            my $yesterdays_result = $columns[5];chomp;
            #my $yesterdays_partition = $columns[6];chomp;
 print "Yesterday\'s disk usage on $host: $columns[5]\n";
            #print "$yesterdays_result <<Y.Result     Y.Partition>>  $yesterdays_partition";
        }

        #Debug: Print differences in mount point (condition must be "ne" instead eq)
        #if ($todays_partition eq $yesterdays_partition){
        #print "$todays_partition <<Partition equal>> $yesterdays_partition";
        #}

        #else{
            #print "Debug: Host or Date DIFFERENT or NOT FOUND for today and yesterday\n";
        #}

        #TO DO: print "The diference: $todays_result-$yesterdays_result", "\n";
};

close HANDLE;

CSV文件包含以下行：

testhost,25.08.2018,100M,0,100M,0,/run/user/0
localhost,01.09.2018,6.7G,1.5G,5.2G,23,/
localhost,01.09.2018,485M,0,485M,0,/dev
localhost,01.09.2018,496M,4.0K,496M,1,/dev/shm
localhost,01.09.2018,496M,6.7M,490M,2,/run
localhost,01.09.2018,496M,0,496M,0,/sys/fs/cgroup
localhost,01.09.2018,497M,110M,387M,23,/boot
localhost,01.09.2018,100M,0,100M,0,/run/user/0
localhost,02.09.2018,6.7G,1.5G,5.2G,23,/
localhost,02.09.2018,485M,0,485M,0,/dev
localhost,02.09.2018,496M,4.0K,496M,1,/dev/shm
localhost,02.09.2018,496M,6.7M,490M,2,/run
localhost,02.09.2018,496M,0,496M,0,/sys/fs/cgroup
localhost,02.09.2018,497M,110M,387M,23,/boot
localhost,02.09.2018,100M,0,100M,0,/run/user/0

奖励：帮助英语语法：D

Answer 1

您需要来自多行的信息，因此您需要在循环之外使用一些变量。

如果CSV中的记录按时间顺序排序，则可以使用以下内容：

use strict;
use warnings;

use DateTime      qw( );
use Sys::Hostname qw( hostname );
use Text::CSV_XS  qw( );

my $qfn = "/tmp/df.csv";
open(my $fh, '<', $qfn)
   or die("Can't open \"$qfn\": $!\n");

my $target_host = hostname =~ s/\..*//rs;   # /

my $today_dt = DateTime->now( time_zone => "local" )->set_time_zone("floating")->truncate( to => "day" );
my $yday_dt  = $today_dt->clone->subtract( days => 1 );

my $today = $today_dt->strftime("%d.%m.%Y");
my $yday  = $yday_dt ->strftime("%d.%m.%Y");

my $csv = Text::CSV_XS->new({ auto_diag => 2, binary => 1 });

my %yday_usage_by_partition;   
while ( my $row = $csv->getline($fh) ) {
   my ($host, $date, $partition, $usage) = @$row[0,1,6,5];
   next if $host ne $target_host;

   if ($date eq $yday) {
      $yday_usage_by_partition{$partition} = $usage;
   }
   elsif ($date eq $today) {
      if (!exists($yday_usage_by_partition{$partition})) {
         warn("No data for $yday for partition $partition\n");
         next;
      }

      print("$partition: $$yday_usage_by_partition{$partition} -> $usage\n");
   }
}