Code Comments
Programming Forum and web based access to our favorite programming groups.
Re: matching column variables from two awks!xxx@x3$ df -h | grep dsk | sort +5
/dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/xp20
/dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/xp21
/dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/xp22
/dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/xp23
/dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/xp24
/dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/xp25
/dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/xp26
/dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/xp27
xxx@x3$ df
/local/ds/xp26(/dev/md/ds/dsk/x26):148445868 blocks 22517420 files
/local/ds/xp12(/dev/md/ds/dsk/x12):144276990 blocks 22608345 files
/local/ds/xp30(/dev/md/ds/dsk/x30):133775594 blocks 22412624 files
/local/ds/xp6 (/dev/md/ds/dsk/x6):121057960 blocks 22383983 files
/local/ds/xp4 (/dev/md/ds/dsk/x4):129103034 blocks 22193600 files
my program foo:
df -h | grep dsk | sort +5 | awk '
BEGIN {
printf("%-30s%12s%12s%12s%10s\n\n", "File System", "bytes", "used",
"available", "capacity")
printf("--------------------------------------------------------------------
----------
\n")
}
{
printf("%-30s%12s%12s%12s%10s\n", $6, $2, $3, $4, $5)
}
END {
}'
xxx@x3$ ./foo
File System bytes used available
capacity
----------------------------------------------------------------------------
--
/global/ds/xp0 200G 139G 59G
71%
/global/ds/xp1 200G 131G 67G
66%
/global/ds/xp10 200G 129G 69G
66%
/global/ds/xp11 200G 133G 65G
68%
/global/ds/xp12 200G 131G 67G
67%
/global/ds/xp13 200G 139G 59G
71%
/global/ds/xp14 200G 139G 59G
71%
/global/ds/xp15 200G 132G 66G
67%
/global/ds/xp16 200G 138G 60G
70%
/global/ds/xp17 200G 128G 70G
65%
/global/ds/xp18 200G 133G 65G
68%
sample of myFile:
11 621 2008/03/21 12:44 xp4 256000 user/foo/INBOX
806 41722 2008/03/21 13:26 xp47 256000 user/bar/INBOX
2391 115428 2008/03/21 13:26 xp5 256000 user/moo/INBOX
2452 122771 2008/03/21 12:38 xp6 256000 user/123/INBOX
638 65484 2008/03/21 12:56 xp6 256000 user/345/INBOX
139 4892 2008/03/21 12:44 xp7 256000 user/789/INBOX
398 13403 2008/03/19 14:05 xp7 256000 user/1234/INBOX
392 32448 2008/03/21 14:14 xp8 256000 user/5678/INBOX
of which this:
xxx@x3$ awk {'print $5'} myFile | sort |uniq -c | sort -n
shows:
1407 xp45
1496 xp40
1503 xp51
1524 xp49
1537 xp50
1539 xp42
where col 1 is the no. of accts. per "xp" partition.
my desired output will show col 1 (as above) on the appropriate row
(e.g. where partition matches, e.g. where col 2 above matches the end
of col 1 from my prog's out)
xxx@x3$ ./foo
File System bytes used available
capacity accts
----------------------------------------------------------------------------
--
/global/ds/xp45 200G 139G 59G
71% 1407
/global/ds/xp40 200G 131G 67G
66% 1496
/global/ds/xp51 200G 129G 69G
66% 1503
/global/ds/xp49 200G 133G 65G
68% 1524
Thanks again!
sg
Post Follow-up to this message
Re: matching column variables from two awks!OK, let's try it this way. Look:
$ cat df_out
/dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/xp20
/dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/xp21
/dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/xp22
/dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/xp23
/dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/xp24
/dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/xp25
/dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/xp26
/dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/xp27
$
$ cat myFile
11 621 2008/03/21 12:44 xp20 256000 user/foo/INBOX
806 41722 2008/03/21 13:26 xp21 256000 user/bar/INBOX
2391 115428 2008/03/21 13:26 xp22 256000 user/moo/INBOX
2452 122771 2008/03/21 12:38 xp23 256000 user/123/INBOX
638 65484 2008/03/21 12:56 xp23 256000 user/345/INBOX
139 4892 2008/03/21 12:44 xp24 256000 user/789/INBOX
398 13403 2008/03/19 14:05 xp24 256000 user/1234/INBOX
392 32448 2008/03/21 14:14 xp25 256000 user/5678/INBOX
$
$ cat tst.awk
BEGIN {
printf "%-15s%12s%10s%10s%10s%10s\n", "File System", "bytes", "used", "avail
",
"capacity", "accounts"
print "-------------------------------------------------------------------"
}
NR==FNR{ accts["/local/ds/"$5]++; next }
/dsk/ { printf "%-15s%12s%10s%10s%10s%10s\n", $6, $2, $3, $4, $5, accts[$6]+
0 }
$
$ cat df_out | awk -f tst.awk myFile -
File System bytes used avail capacity accounts
-------------------------------------------------------------------
/local/ds/xp20 200G 129G 69G 66% 1
/local/ds/xp21 200G 135G 63G 69% 1
/local/ds/xp22 200G 148G 50G 75% 1
/local/ds/xp23 200G 128G 70G 65% 2
/local/ds/xp24 200G 126G 72G 64% 2
/local/ds/xp25 200G 136G 62G 69% 1
/local/ds/xp26 200G 129G 69G 66% 0
/local/ds/xp27 200G 130G 68G 66% 0
Is that what you wanted?
If so, just replace "cat df_out" with "df -h | sort +5" (no need for grep):
df -h | sort +5 | awk -f tst.awk myFile -
If not, what are you looking for?
Ed.
Post Follow-up to this message Re: matching column variables from two awks!On Mar 21, 6:32 pm, Ed Morton <mor...@lsupcaemnt.com> wrote:
> OK, let's try it this way. Look:
>
> $ cat df_out
> /dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/xp20
> /dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/xp21
> /dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/xp22
> /dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/xp23
> /dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/xp24
> /dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/xp25
> /dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/xp26
> /dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/xp27
> $
> $ cat myFile
> 11 621 2008/03/21 12:44 xp20 256000 user/foo/INBOX
> 806 41722 2008/03/21 13:26 xp21 256000 user/bar/INBOX
> 2391 115428 2008/03/21 13:26 xp22 256000 user/moo/INBOX
> 2452 122771 2008/03/21 12:38 xp23 256000 user/123/INBOX
> 638 65484 2008/03/21 12:56 xp23 256000 user/345/INBOX
> 139 4892 2008/03/21 12:44 xp24 256000 user/789/INBOX
> 398 13403 2008/03/19 14:05 xp24 256000 user/1234/INBOX
> 392 32448 2008/03/21 14:14 xp25 256000 user/5678/INBOX
> $
> $ cat tst.awk
> BEGIN {
> printf "%-15s%12s%10s%10s%10s%10s\n", "File System", "bytes", "used", "ava
il",
> "capacity", "accounts"
> print "-------------------------------------------------------------------
"}
>
> NR==FNR{ accts["/local/ds/"$5]++; next }
> /dsk/ { printf "%-15s%12s%10s%10s%10s%10s\n", $6, $2, $3, $4, $5, accts[$6
]+0 }
> $
> $ cat df_out | awk -f tst.awk myFile -
> File System bytes used avail capacity accounts
> -------------------------------------------------------------------
> /local/ds/xp20 200G 129G 69G 66% 1
> /local/ds/xp21 200G 135G 63G 69% 1
> /local/ds/xp22 200G 148G 50G 75% 1
> /local/ds/xp23 200G 128G 70G 65% 2
> /local/ds/xp24 200G 126G 72G 64% 2
> /local/ds/xp25 200G 136G 62G 69% 1
> /local/ds/xp26 200G 129G 69G 66% 0
> /local/ds/xp27 200G 130G 68G 66% 0
>
> Is that what you wanted?
yes - it's what i want - but making those files on my machine - and
running:
cat df_out | awk -f tst.awk myFile -
produces this:
File System bytes used avail capacity accounts
-------------------------------------------------------------------
/local/ds/xp20 200G 129G 69G 66% 0
/local/ds/xp21 200G 135G 63G 69% 0
/local/ds/xp22 200G 148G 50G 75% 0
/local/ds/xp23 200G 128G 70G 65% 0
/local/ds/xp24 200G 126G 72G 64% 0
/local/ds/xp25 200G 136G 62G 69% 0
/local/ds/xp26 200G 129G 69G 66% 0
/local/ds/xp27 200G 130G 68G 66% 0
>
> If so, just replace "cat df_out" with "df -h | sort +5" (no need for grep)
:
>
> df -h | sort +5 | awk -f tst.awk myFile -
>
> If not, what are you looking for?
>
> Ed.
Post Follow-up to this message Re: matching column variables from two awks!
On 3/22/2008 12:00 AM, spacegoose wrote:
> On Mar 21, 6:32 pm, Ed Morton <mor...@lsupcaemnt.com> wrote:
>
>
>
> yes - it's what i want - but making those files on my machine - and
> running:
>
> cat df_out | awk -f tst.awk myFile -
>
> produces this:
>
>
>
> File System bytes used avail capacity accounts
> -------------------------------------------------------------------
> /local/ds/xp20 200G 129G 69G 66% 0
> /local/ds/xp21 200G 135G 63G 69% 0
> /local/ds/xp22 200G 148G 50G 75% 0
> /local/ds/xp23 200G 128G 70G 65% 0
> /local/ds/xp24 200G 126G 72G 64% 0
> /local/ds/xp25 200G 136G 62G 69% 0
> /local/ds/xp26 200G 129G 69G 66% 0
> /local/ds/xp27 200G 130G 68G 66% 0
>
That seems impossible.
Are you SURE you copy/pasted what I posted into files rather than retyping i
t? I
suspect your version of "myFile" is NOT the same one I posted.
Do this:
cat df_out
cat myFile
awk '{print FILENAME,$FNR,$0}' myFile df_out
and copy/paste the result into a posting so we can see what you're working w
ith.
Ed.
Post Follow-up to this message Re: matching column variables from two awks!
> That seems impossible.
>
> Are you SURE you copy/pasted what I posted into files rather than retyping
it? I
> suspect your version of "myFile" is NOT the same one I posted.
>
> Do this:
>
> cat df_out
> cat myFile
> awk '{print FILENAME,$FNR,$0}' myFile df_out
>
> and copy/paste the result into a posting so we can see what you're working
with.
>
> Ed.
I am running Solaris 10. Also tried on Solaris 9 -same result.
xxx@x3$ cat df_out
/dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/xp20
/dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/xp21
/dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/xp22
/dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/xp23
/dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/xp24
/dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/xp25
/dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/xp26
/dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/xp27
xxx@x3$ cat myFile
11 621 2008/03/21 12:44 xp20 256000 user/foo/INBOX
806 41722 2008/03/21 13:26 xp21 256000 user/bar/INBOX
2391 115428 2008/03/21 13:26 xp22 256000 user/moo/INBOX
2452 122771 2008/03/21 12:38 xp23 256000 user/123/INBOX
638 65484 2008/03/21 12:56 xp23 256000 user/345/INBOX
139 4892 2008/03/21 12:44 xp24 256000 user/789/INBOX
398 13403 2008/03/19 14:05 xp24 256000 user/1234/INBOX
392 32448 2008/03/21 14:14 xp25 256000 user/5678/INBOX
xxx@x3$ awk '{print FILENAME,$FNR,$0}' myFile df_out
myFile 11 621 2008/03/21 12:44 xp20 256000 user/foo/
INBOX 11 621 2008/03/21 12:44 xp20 256000 user/foo/
INBOX
myFile 806 41722 2008/03/21 13:26 xp21 256000 user/bar/
INBOX 806 41722 2008/03/21 13:26 xp21 256000 user/bar/
INBOX
myFile 2391 115428 2008/03/21 13:26 xp22 256000 user/moo/
INBOX 2391 115428 2008/03/21 13:26 xp22 256000 user/moo/
INBOX
myFile 2452 122771 2008/03/21 12:38 xp23 256000 user/123/
INBOX 2452 122771 2008/03/21 12:38 xp23 256000 user/123/
INBOX
myFile 638 65484 2008/03/21 12:56 xp23 256000 user/345/
INBOX 638 65484 2008/03/21 12:56 xp23 256000 user/345/
INBOX
myFile 139 4892 2008/03/21 12:44 xp24 256000 user/789/
INBOX 139 4892 2008/03/21 12:44 xp24 256000 user/789/
INBOX
myFile 398 13403 2008/03/19 14:05 xp24 256000 user/1234/
INBOX 398 13403 2008/03/19 14:05 xp24 256000 user/1234/
INBOX
myFile 392 32448 2008/03/21 14:14 xp25 256000 user/5678/
INBOX 392 32448 2008/03/21 14:14 xp25 256000 user/5678/
INBOX
myFile
df_out /dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/
xp20 /dev/md/ds/dsk/x20 200G 129G 69G 66% /local/ds/xp20
df_out /dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/
xp21 /dev/md/ds/dsk/x21 200G 135G 63G 69% /local/ds/xp21
df_out /dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/
xp22 /dev/md/ds/dsk/x22 200G 148G 50G 75% /local/ds/xp22
df_out /dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/
xp23 /dev/md/ds/dsk/x23 200G 128G 70G 65% /local/ds/xp23
df_out /dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/
xp24 /dev/md/ds/dsk/x24 200G 126G 72G 64% /local/ds/xp24
df_out /dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/
xp25 /dev/md/ds/dsk/x25 200G 136G 62G 69% /local/ds/xp25
df_out /dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/
xp26 /dev/md/ds/dsk/x26 200G 129G 69G 66% /local/ds/xp26
df_out /dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/
xp27 /dev/md/ds/dsk/x27 200G 130G 68G 66% /local/ds/xp27
also:
xxx@x3$ more tst.awk
BEGIN {
printf "%-15s%12s%10s%10s%10s%10s\n", "File System", "bytes", "used",
"avail", "capacity", "accounts"
print
"-------------------------------------------------------------------"
}
NR==FNR{ accts["/local/ds/"$5]++; next }
/dsk/ { printf "%-15s%12s%10s%10s%10s%10s\n", $6, $2, $3, $4, $5,
accts[$6]+0 }
Thanks,
sg
Post Follow-up to this message Re: matching column variables from two awks!On 3/22/2008 9:37 AM, spacegoose wrote: > I am running Solaris 10. Also tried on Solaris 9 -same result. Ahh, you're using old, broken awk (/usr/bin/awk on Solaris). Use a modern aw k, e.g. GNU awk (gawk), New awk (nawk) or /usr/xpg4/bin/awk on Solaris. Ed.
Post Follow-up to this message Re: matching column variables from two awks!In article <47E59349.30308@lsupcaemnt.com>, Ed Morton <morton@lsupcaemnt.com> wrote: >On 3/22/2008 9:37 AM, spacegoose wrote: > >Ahh, you're using old, broken awk (/usr/bin/awk on Solaris). Use a modern a wk, >e.g. GNU awk (gawk), New awk (nawk) or /usr/xpg4/bin/awk on Solaris. > > Ed. > This thread went on for quite a while, didn't it? Amazing how many of these back-and-forths eventually boil down to: Oh, by the way, I'm using Solaris. (And if not that, the other one is: Oh, I'm writing the code on a Windows box [and running/testing it on Unix])
Post Follow-up to this message Re: matching column variables from two awks!On 3/22/2008 6:23 PM, Kenny McCormack wrote: > In article <47E59349.30308@lsupcaemnt.com>, > Ed Morton <morton@lsupcaemnt.com> wrote: > > > > This thread went on for quite a while, didn't it? > > Amazing how many of these back-and-forths eventually boil down to: > Oh, by the way, I'm using Solaris. > > (And if not that, the other one is: Oh, I'm writing the code on a > Windows box [and running/testing it on Unix]) > The other biggy is "post sample input and expected output". I always forget to say "post sample input and THE expected output FROM THAT INPUT". Got to reme mber that.... Ed.
Post Follow-up to this message Re: matching column variables from two awks!On Mar 22, 7:44 pm, Ed Morton <mor...@lsupcaemnt.com> wrote: > On 3/22/2008 6:23 PM, Kenny McCormack wrote: > > > > > > > > > > > > The other biggy is "post sample input and expected output". I always forge t to > say "post sample input and THE expected output FROM THAT INPUT". Got to re member > that.... > > Ed. That did it. Thanks. One last thing: Can this be combined into one file to act on myFile: so that everything is contained in one file (with a reference to myFile inside)?
Post Follow-up to this message Re: matching column variables from two awks!On 3/23/2008 12:13 AM, spacegoose wrote:
<snip>
> That did it. Thanks. One last thing: Can this be combined into one
> file to act on myFile:
> so that everything is contained in one file (with a reference to
> myFile inside)?
I think what you want is to put this in a file:
------------
df -h | sort +5 | awk '
BEGIN {
printf "%-15s%12s%10s%10s%10s%10s\n", "File System", "bytes", "used", "avail
",
"capacity", "accounts"
print "-------------------------------------------------------------------"
}
NR==FNR{ accts["/local/ds/"$5]++; next }
/dsk/ { printf "%-15s%12s%10s%10s%10s%10s\n", $6, $2, $3, $4, $5, accts[$6]+
0 }
' myFile -
-------------
Regards,
Ed.
Post Follow-up to this message
Sponsored Links
Powered by vBulletin
Copyright 2000-2006 Jelsoft Enterprises Limited.
