Hi,
There's a file with below contents which I have to read based on the input parameter provided by the user.
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
FILE_ID=2
FILE_FTP_ID=ftp.server2.com
FILE_FTP_USER=user2
FILE_FTP_PASS=pass2
FILE_ID=3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
FILE_ID=4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
For e.g. if user entered 1
./ftp_files 1
Then the script should only read the first block i.e. and assign the values for ftp site, user & pass,etc....
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
which I am doing with the below command
grep -A3 "FILE_ID=$1" ftp_info_file
Now the problem is that no. of lines between
FILE_ID=1
FILE_ID=2
may vary (there could be more information to read); so -A3 would not help me. I need to know dynamically some how the no. of lines between the first and the next pattern.
I thought of something like using -m & -n options of grep
grep -m `cat ftp_info_file | wc -l` "FILE_ID" -n ftp_info_file
which displays lines matched with the line numbers and then I can subtract those numbers to get no. of lines between the two contiguous groups of matches. But I am wondering if there's a neater way to do it?
Please suggest.
-dips
Something like this?
awk -F"=" -v id="$1" '$2==id{f=1}f && !NF{exit}f' ftp_info_file
Not sure if this is what you want...
ftpdata
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
FILE_ID=2
FILE_FTP_ID=ftp.server2.com
FILE_FTP_USER=user2
FILE_FTP_PASS=pass2
FILE_FTP_PARAM_1=param1
FILE_FTP_PARAM_2=param2
FILE_ID=3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
FILE_FTP_VALUE_1=val1
FILE_FTP_VALUE_2=val2
FILE_FTP_VALUE_3=val3
FILE_ID=4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
getdata.sh
#!/bin/sh
string="FILE_ID=$1"
sed -n '/'$string'/,/^ /p;' ftpdata | sed '$d'
$ ./getdata.sh 1
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
$ ./getdata.sh 2
FILE_ID=2
FILE_FTP_ID=ftp.server2.com
FILE_FTP_USER=user2
FILE_FTP_PASS=pass2
FILE_FTP_PARAM_1=param1
FILE_FTP_PARAM_2=param2
$ ./getdata.sh 3
FILE_ID=3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
FILE_FTP_VALUE_1=val1
FILE_FTP_VALUE_2=val2
FILE_FTP_VALUE_3=val3
$ ./getdata.sh 4
FILE_ID=4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
$
[root@sistem1lnx first]# cat ftpfile
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
FILE_ID=2
FILE_FTP_ID=ftp.server2.com
FILE_FTP_USER=user2
FILE_FTP_PASS=pass2
FILE_ID=3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
FILE_ID=4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
First of all we change your ftpdata file 
# Let Format changing our ftpdata file
[root@sistem1lnx first]# cat ftpfile | sed -e 's/.*=//g' | sed -e :j -e '$b;N;s/\n/ /;bj' | sed -e 's/ /\n/g' > newftpfile
# Our newftpfile format
[root@sistem1lnx first]# cat newftpfile
1 ftp.server1.com user1 pass1
2 ftp.server2.com user2 pass2
3 ftp.server3.com user3 pass3
4 ftp.server4.com user4 pass4
And then ;
# Let Modifying our script with our newftpdata file
[root@sistem1lnx first]# cat ftp_files
#!/bin/bash
while read ftpid ftpadr ftpusr ftppass
do
if [ $ftpid = $1 ] ; then
FILE_FTP_ID=$ftpadr
FILE_FTP_USER=$ftpusr
FILE_FTP_PASS=$ftppass
fi
done < newftpfile
# test
echo "FILE_FTP_ID="$FILE_FTP_ID
echo "FILE_FTP_USER="$FILE_FTP_USER
echo "FILE_FTP_PASS="$FILE_FTP_PASS
# Now let start our ftp script
[root@sistem1lnx first]# ./ftp_files 3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
And other exa..
[root@sistem1lnx first]# ./ftp_files 1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
[root@sistem1lnx first]# ./ftp_files 4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
I hope this job is OK 
Regards ygemici
Thanks once again Franklin52 your solution works perfectly!
But can you please expalin the command (I understand bits of it but not the whole of it :(; which is useless!)
pseudocoder- I tried your command
sed -n '/'FILE_ID=1'/,/^ /p;' ftp_info_file | sed '$d'
but it still prints the whole file.
FILE_ID=1
FILE_FTP_ID=ftp.server1.com
FILE_FTP_USER=user1
FILE_FTP_PASS=pass1
FILE_ID=2
FILE_FTP_ID=ftp.server2.com
FILE_FTP_USER=user2
FILE_FTP_PASS=pass2
FILE_ID=3
FILE_FTP_ID=ftp.server3.com
FILE_FTP_USER=user3
FILE_FTP_PASS=pass3
FILE_ID=4
FILE_FTP_ID=ftp.server4.com
FILE_FTP_USER=user4
FILE_FTP_PASS=pass4
-dips
Sure.
awk -F"=" -v id="$1" '$2==id{f=1}f && !NF{exit}f'
-F"=" < set fieldseparator
-v id="$1" < set awk variable
$2==id{f=1} < if field 2 == id set flag
f && !NF{exit} < exit if flag is set and record doesn't have a field
f < print the record if flag is set
Thanks Franklin52 for the explanation.
-dips