Extract string from many brackets - bash

Question

I have a file with this content:
ok: [10.9.22.122] => {
"out.stdout_lines": [
"cgit-1.1-11.el7.x86_64",
"python-paramiko-2.1.1-0.9.el7.noarch",
"varnish-libs-4.0.5-1.el7.x86_64",
"kernel-3.10.0-862.el7.x86_64"
]
}
ok: [10.9.33.123] => {
"out.stdout_lines": [
"python-paramiko-2.1.1-0.9.el7.noarch"
]
}
ok: [10.9.44.124] => {
"out.stdout_lines": [
"python-paramiko-2.1.1-0.9.el7.noarch",
"kernel-3.10.0-862.el7.x86_64"
]
}
ok: [10.9.33.29] => {
"out.stdout_lines": []
}
ok: [10.9.22.28] => {
"out.stdout_lines": [
"NetworkManager-tui-1:1.12.0-8.el7_6.x86_64",
"java-1.8.0-openjdk-javadoc-zip-debug-1:1.8.0.171-8.b10.el7_5.noarch",
"java-1.8.0-openjdk-src-1:1.8.0.171-8.b10.el7_5.x86_64",
"kernel-3.10.0-862.el7.x86_64",
"kernel-tools-3.10.0-862.el7.x86_64",
]
}
ok: [10.2.2.2] => {
"out.stdout_lines": [
"monitorix-3.10.1-1.el6.noarch",
"singularity-runtime-2.6.1-1.1.el6.x86_64"
]
}
ok: [10.9.22.33] => {
"out.stdout_lines": [
"NetworkManager-1:1.12.0-8.el7_6.x86_64",
"gnupg2-2.0.22-5.el7_5.x86_64",
"kernel-3.10.0-862.el7.x86_64",
]
}
I need to extract the IP between [] if into stout_line contains kernel*.
I want to "emulate" substring, to save a 'block' of content into varible and go through the all file.
How would I use sed, or other, to do this if I have many delimiter?

Answer 1

A GNU awk solution:
awk -F'\\]|\\[' 'tolower($3)~/"out.stdout_lines" *:/ && tolower($4)~/"kernel/{print "The IP " $2 " cointain Kernel"}' RS='}' file
Output:
The IP 10.9.22.122 cointain Kernel
The IP 10.9.44.124 cointain Kernel
The IP 10.9.22.28 cointain Kernel
The IP 10.9.22.33 cointain Kernel
I used ] or [ as FS field separator, and } as RS record separator.
So the IP will just becomes $2.
This solution depends on the structure, that means "out.stdout_lines" needs to be in the field after [ip] like you showed in your example.
Another GNU awk way, no above limitation:
awk -F']' 'match(tolower($0),/"out\.stdout_lines": *\[([^\]]+)/,m){if(m[1]~/"kernel/)print "The IP " substr($1, index($1,"[")+1) " cointain Kernel"}' RS='}' file
Same output. The tolowers are for case insensitive match, If you want exact match, you can remove them or just use solutions from Revision 6.
Combine merits from above two ways, the Third way:
awk -F'\\]|\\[' 'match(tolower($0),/"out\.stdout_lines": *\[([^\]]+)/,m){if(m[1]~/"kernel/)print "The IP " $2 " cointain Kernel"}' RS='}' file
Change tolower($0) to $0 if you don't need case insensitive match.

Answer 2

$ gawk -v RS="ok: " -F " => " '$2 ~ /[Kk]ernel/ { printf "The IP %s contains Kernel\n", $1 }' file
The IP [10.9.22.122] contains Kernel
The IP [10.9.44.124] contains Kernel

Answer 3

since your data are pretty much well-formated, you can use awk(gawk):
awk '
# get the ip address
/ok:/ {ip = gensub(/[^0-9\.]/, "", "g", $2) }
# check the stdout_lines block and print Kernal and ip saved from the above line
/"out.stdout_lines":/,/\]/ { if (/\<[Kk]ernel\>/) print ip}
' file
#10.9.22.122
#10.9.44.124
#10.9.22.28
#10.9.22.28
#10.9.22.33
Note:
I adjusted the regexes to reflect to your updated data.
you might get more than one Kernel files for the same IP under the out.stdout_lines block, which will yield the same IP multiple times. If this happens just pipe the result to | uniq

Answer 4

This might work for you (GNU sed):
sed -n '/ok:/{s/[^0-9.]//g;:a;N;/]/!ba;/stdout_line.*kernel/P}' file
Set the -n to suppress implicit printing
If a line contains the the string ok: this is an IP address, strip the line of everything but integers and periods.
Append further lines until a line containing ] is encountered and if the pattern space contains both stdout_line and kernel, print the first line.

Answer 5

Fast solution:
#!/bin/bash
AWK='
/^ok:/ { gsub(/^.*\[/,""); gsub(/].*$/,""); ip=$0 }
/"Kernel-default/ { if (ip) print ip; ip="" }
'
awk "$AWK" INPUT

Answer 6

Could you please try following, this should work for most of the awks I believe.(I have added [kK] in condition match so it should look for kernal OR Kernal both strings(since OP's previous sample was having capital K and now it has ksmall one, so thought to cover both here).
awk '
/ok/{
gsub(/.*\[|\].*/,"")
ip=$0
}
/stdout_line/{
found=1
next
}
found && /[kK]ernel/{
print ip
}
/}/{
ip=found=""
}
' Input_file
Explanation: Adding explanation for above code.
awk ' ##Starting awk program here.
/ok/{ ##Checking condition if a line contains string ok in it then do following.
gsub(/.*\[|\].*/,"") ##Globally substituting everything till [ and everything till ] with NULL in current line.
ip=$0 ##Creating variable named ip whose values is current line value(edited one).
} ##Closing BLOCK for ok string check condition.
/stdout_line/{ ##Checking condition if a line contains stdout_line then do following.
found=1 ##Set value of variable named found to 1 here.
next ##next will skip all further statements from here.
} ##Closing BLOCK for stdout_line string check condition here.
found && /[kK]ernel/{ ##Checking condition if variable found is NOT NULL and string Kernel found in current line then do following.
print ip ##Printing value of variable ip here.
} ##Closing BLOCK for above condition now.
/}/{ ##Checking condition if a line contains } then do following.
ip=found="" ##Nullify ip and found variable here.
} ##Closing BLOCK for } checking condition.
' Input_file ##Mentioning Input_file name here.
Output will be as follows.
10.9.22.122
10.9.44.124
10.9.22.28
10.9.22.28
10.9.22.33

Answer 7

Using Perl
$ perl -0777 -ne 's!\[(\S+)\].+?\{(.+?)\}!$y=$1;$x=$2;$x=~/kernel/ ? print "$y\n":""!sge' brenn.log
10.9.22.122
10.9.44.124
10.9.22.28
10.9.22.33
$