Guys
Following input line is from /etc/group file.As we know last entry in a line of /etc/group is userlist (all the users belonging to that group).
I need to splilt this one line into 3 lines as shown below (3 because userlist has 3 names in it).
Input:
lp:!:11:root,lp,printq
Output:
lp<TAB>!<TAB>11<TAB>root
lp<TAB>!<TAB>11<TAB>lp
lp<TAB>!<TAB>11<TAB>printq
This way i need to do for all the entries in /etc/group.
I need code lines to achive it.
Help Appreciated !!
Regards
Abhi
awk '
{
split($0,A,":");
split(A[4],UL,",");
for (U in UL) {print A[1] "\t" A[2] "\t" A[3] "\t" UL}
}' /etc/group
You probably meant "achieve" iso "achive". Help is not always appreciated. 
This will split the last line the way you want
awk -F'[:|,]' 'END{for(i=3;++i<=NF;)print $1,$2,$3,$i}' OFS="<TAB>" file
Thanks for the replies guys.
<Myhost>:$ C='lp:!:11:root,lp,printq'
<Myhost>:$ echo $C
lp:!:11:root,lp,printq
<Myhost>:$ echo $C > c
<Myhost>:$ cat c
lp:!:11:root,lp,printq
<Myhost>:$ awk '
> {
> split($0,A,":");
> split(A[4],UL,",");
> for (U in UL) {print A[1] "\t" A[2] "\t" A[3] "\t" UL}
> }' c
lp ! 11
lp ! 11
lp ! 11
this is not giving me what i want.
Regards
Abhi
Change in the line 5 ->
for (U in UL) {print A[1] "\t" A[2] "\t" A[3] "\t" UL}
U should be in upper case in UL[u]
yup...i shud have seen that...thanks...!!
even i tried writing something ,though its not giving me entirely what i want..
cat /etc/group | nawk '
{
TUP = split($0, field, ":");
USERS = split(field[4],user, ",");
if (USERS == 0) printf "%s\t%s\t%s\n", field[1], field[3], "NULL";
for (i = 1; i <= USERS; i++)
printf "%s\t%s\t%s\n", field[1], field[3], user;
}
'
I m just setting NULL wherever userlist if NULL.
Can someone tell me how do i split the row by removing ':' from it into just a row with n columns
lp:!:11:root,lp,printq
into
lp ! 11 root,lp,printq
such that i can just pick column values using 'awk'.
Regards
Abhi
---------- Post updated at 02:25 PM ---------- Previous update was at 02:02 PM ----------
okay...this might be it...
bash-3.00$ D='lp:!:11:root,lp,printq'
bash-3.00$ echo $D |sed 's/:/ /g'
lp ! 11 root,lp,printq
bash-3.00$ echo $D |sed 's/:/ /g'|awk '{print $1}'
lp
bash-3.00$ echo $D |sed 's/:/ /g'|awk '{print $2}'
!
bash-3.00$ echo $D |sed 's/:/ /g'|awk '{print $3}'
11
bash-3.00$ echo $D |sed 's/:/ /g'|awk '{print $4}'
root,lp,printq
Regards
Abhi
---------- Post updated at 02:39 PM ---------- Previous update was at 02:25 PM ----------
now this one is from /etc/passwd file
abhi:!:34971:3418:abhi k, Sys Admin Level 3:/home/users/abhi:/bin/ksh
As we can see,in comments section i have put something against my user id.
using the same code above pasted
bash-3.00$ echo $D|sed 's/:/ /g'|awk '{print $5 $6 $7 $8 $9 $10}'
abhik,SysAdminLevel3
I want "abhi k ,Sys Admin Level 3" to be put in just one column as it is (no space deletion).Rest columns are easy to get.
Any ideas ?
Regards
Abhi
---------- Post updated at 03:48 PM ---------- Previous update was at 02:39 PM ----------
i thought this wud work...but it isn't....anything missing?
bash-3.00$ echo $D|sed 's/:/ /g'|awk -F' ' '{$1=$2=$3=$4="";sub("^ ","");print}'
abhi k, Sys Admin Level 3 /home/users/abhi/ /bin/ksh
i just need "abhi k, Sys Admin Level 3".....
Regards
Abhi
---------- Post updated at 04:12 PM ---------- Previous update was at 03:48 PM ----------
bash-3.00$ echo $D|sed 's/:/ /g'|awk -F' ' '{$1=$2=$3=$4="";sub("^ ","");print}'|cut -d "/" -f1
abhi k, Sys Admin Level 3
i hope i am not sounding crazy here....
Regards
Abhi
Drop the sed, you are losing important information.
Start with:
awk -F ":"
Thanks !! but i did not understand what important info....
even with awk -F':' ' , i m getting same result...
bash-3.00$ D='abhi:!:34971:3418:abhi k, Sys Admin Level 3:/home/users/abhi:/bin/ksh'
bash-3.00$ echo $D|awk -F':' '{$1=$2=$3=$4="";sub("^ ","");print}'
abhi k, Sys Admin Level 3 /home/users/abhi /bin/ksh
bash-3.00$ echo $D|awk -F':' '{$1=$2=$3=$4="";sub("^ ","");print}'|cut -d "/" -f1
abhi k, Sys Admin Level 3
would appreciate if you can point-out !!
Regards
Abhi
awk -F ':' '{print $5}' /etc/passwd
okay...
now i badly need advice...
I have a fingerprinting tool thru which i generate a lot of data in CSVs.
I need all of information in /etc/group file.
lets say,i want to get following line in my CSV.My requirement here is bit tricky.I want 2 separate entries in CSV for following line.
bin:!:2:root,bin
should look like
bin 2 root
bin 2 bin
now this is how i have written command lines in my fingerprinting tool:
ParentFamily : Host etcGroup
GroupName:
cat /etc/group |awk '{print $1}' | cut -d ":" -f1
GID
A=`cat /etc/group | awk -F':' '{if ($1 == "@@Host etcGroup.GroupName@@" ) print $3}'`
echo $A
UserList
A=`cat /etc/group | awk -F':' '{if ($1 == "@@Host etcGroup.GroupName@@" ) print $4}'`
B=`echo $A|wc -w`
C=`echo $A|grep "\,"|wc -l`
D=`echo $A|awk -F',' '{ print NF }'`
if [ $B = 0 ]
then
echo NULL
fi
if [ $B = 1 ]
then
if [ $C = 0 ]
then
echo $A
else
XXXXXXX
XXXXXXX
fi
fi
As you can see,i am stuck at 'XXXXXX' place.I am not sure what to write there to get what is required .
Regards
Abhi
# echo 'bin:!:2:root,bin' | awk -F'[:|,]' 'END{for(i=3;++i<=NF;)print $1,$3,$i}'
bin 2 root
bin 2 bin
You are not stuck, just make your life complicated.
well...thanks for this !!
echo 'bin:!:2:root,bin' | awk -F'[:|,]' 'END{for(i=3;++i<=NF;)print $1,$3,$i}'
bin 2 root
bin 2 bin
but the fingerprinting engine does not work this way.I have defined 3 columns totally ,'GroupName' being first and am fetching 'GroupName' in next 2columns.
whatever data i want to see in CSV,tool has to generate it...its a complicated thing ....can't help it !!
to get something like 'bin:!:2:root,bin' before passing to 'awk',i need to write additonal lines and even after it may not work.
I'll try this and would post my update.
Regards
Abhi
This problem has been dealt with in a different thread, recently.
Can you point me to that thread pls?
By the way,i found solution to my issue...i divided my code into parent-child relationship and wrote following lines for child fingerprint....the 'while' loop is not mine..i took it from very this forum....
A=`cat /etc/group | awk -F':' '{if ($1 == "@@Host etcGroup.GroupName@@" ) print $4}'`
B=`echo $A|wc -w`
if [ $B = 0 ]
then
echo NULL
else
count=`echo $A| nawk -F, {'print NF'}`
i=1
while [ $i -le $count ]
do
str[$i]=`echo $A| cut -d, -f${i}`
echo "${str[$i]}"
i=`expr $i + 1`
done
fi
Thanks to everyone who responded !!
Regards
Abhi