spaces are missing

SPF · 01-16-2008, 01:19 AM

I'm running bash:

Code:

$ ps
  PID TTY          TIME CMD
28408 pts/20   00:00:00 bash
13908 pts/20   00:00:00 ps
$ bash --version
GNU bash, version 2.04.21(1)-release (i386-redhat-linux-gnu)

This is the input data I want to progress:

Code:

$ top -b -d1 -n1 | tr -d '\r' | head -n 20


  4:55pm  up 40 days,  4:38, 26 users,  load average: 0.23, 0.38, 0.23
313 processes: 312 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  1.1% user,  0.5% system,  0.0% nice,  0.5% idle
Mem:  1033960K av,  882548K used,  151412K free,       0K shrd,   21428K buff
Swap: 1572944K av,   96712K used, 1476232K free                  560116K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
    1 root       9   0   508  472   440 S     0.0  0.0   0:08 init
    2 root       9   0     0    0     0 SW    0.0  0.0   0:00 keventd
    3 root      19  19     0    0     0 SWN   0.0  0.0   0:23 ksoftirqd_CPU0
    4 root       9   0     0    0     0 SW    0.0  0.0   0:10 kswapd
    5 root       9   0     0    0     0 SW    0.0  0.0   0:00 bdflush
    6 root       9   0     0    0     0 SW    0.0  0.0   0:09 kupdated
    7 root       9   0     0    0     0 SW    0.0  0.0   0:00 khubd
    8 root      -1 -20     0    0     0 SW<   0.0  0.0   0:00 mdrecoveryd
  593 root       9   0   612  604   516 S     0.0  0.0  42:51 syslogd

But if I change some lines, the spaces at the beginning of the lines are gone (this is an example that should work):

Code:

$ top -b -d1 -n1 | tr -d '\r' | head -n 30 | while read -r line; do echo -e "$line" ; done

8:05am  up 40 days, 19:48,  8 users,  load average: 0.18, 0.18, 0.17
234 processes: 233 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  1.1% user,  0.5% system,  0.0% nice,  0.8% idle
Mem:  1033960K av,  861712K used,  172248K free,       0K shrd,   17056K buff
Swap: 1572944K av,  105028K used, 1467916K free                  609364K cached
PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
612 rpc       13   0   604  580   516 S     0.9  0.0 506:55 portmap
1 root       9   0   508  472   440 S     0.0  0.0   0:08 init
2 root       9   0     0    0     0 SW    0.0  0.0   0:00 keventd
3 root      19  19     0    0     0 SWN   0.0  0.0   0:23 ksoftirqd_CPU0
4 root       9   0     0    0     0 SW    0.0  0.0   0:10 kswapd
5 root       9   0     0    0     0 SW    0.0  0.0   0:00 bdflush
6 root       9   0     0    0     0 SW    0.0  0.0   0:09 kupdated
7 root       9   0     0    0     0 SW    0.0  0.0   0:00 khubd
8 root      -1 -20     0    0     0 SW<   0.0  0.0   0:00 mdrecoveryd
593 root       9   0   612  604   516 S     0.0  0.0  43:36 syslogd
598 root       9   0  1272  432   432 S     0.0  0.0   0:00 klogd
630 rpcuser    9   0   692  596   596 S     0.0  0.0   0:00 rpc.statd
676 root       9   0     0    0     0 SW    0.0  0.0   0:00 rpciod
677 root       9   0     0    0     0 SW    0.0  0.0   0:00 lockd
728 root       9   0   584  500   484 S     0.0  0.0   0:00 automount
740 daemon     9   0   548  488   476 S     0.0  0.0   0:00 atd
775 root       9   0   680  632   488 S     0.0  0.0   0:37 smartd
787 root       8   0  1084  984   900 S     0.0  0.0   5:23 sshd
808 root       9   0   912  820   728 S     0.0  0.0   2:12 xinetd

If you view it in hex, you'll see that there are indeed spaces (hex code 20):

Code:

$ top -b -d1 -n1 | tr -d '\r' | head -n 30 | od -cx
0001620                   6       r   o   o   t                        
        2020 2020 2036 6f72 746f 2020 2020 2020

You need to use the "" characters to preserve spaces in the variables, but somehow it doesn't work here?

I also tried it with:

Code:

$ bash --version
GNU bash, version 3.1.17(1)-release (i486-pc-linux-gnu)

But with no luck.
What do I want with this printout? I'm getting this printout from various systems. But every printout comes with extra data which I have to delete.
That part is already finished. But I noticed during processing that the spaces at the beginning of the lines were deleted.

Disillusionist · 01-16-2008, 02:10 AM

This is because the read command removes the leading spaces.

Have you though of using Perl?

Code:

#!/usr/bin/perl
##
## f_perl_read
##
## A simple perl script to read from Standard Input
##
@l_array=<STDIN>;
foreach $l_var (@l_array)
{
   ###
   ### Any checks or additional processing can go here
   ###
   print "$l_var";
}

Code:

top -b -d1 -n1 | tr -d '\r' | head -n 30 |{dir_to_script}/f_perl_read

SPF · 01-16-2008, 02:20 AM

Thanks Disillusionist, I didn't know that read removed the leading spaces.

I'll consider using perl.

Disillusionist · 01-18-2008, 02:50 AM

This could (and probably should) be written as:

Code:

#!/usr/bin/perl
##
## Second script example
##
while (<STDIN>) {
   ##
   ## Perform any checks here!
   ##
   print $_;
}

Notes:

$_ is a special variable in Perl and is used to store results of read requests where a variable hasn't been stated.

In this case it contains the current line from STDIN.

The " " symbols are not required in the print statement.

The script loops through STDIN without building an array, this should reduce the amount of RAM being used. (Especially if you are passing in a lot of information)

osor · 01-18-2008, 04:36 PM

If you want to continue using the shell (instead of e.g., perl), you can temporarily disable the field separation functionality of read. For example,

Code:

top -b -d1 -n1 | tr -d '\r' | head -n 30 | (IFS=''; while read -r line; do echo -e "$line" ; done)