Shell Programming and Scripting

i have file which looks like this:

Code:

file=/var/log/log1.log
file=/var/log/log2.log
cmd=ls -l
cmd=dir > /var/log/mylog.log

now i have to write a script which will read all "file" tag values and copy them and read all "cmd" values and run them etc..

what i was thinking to do is:
read file by line and:
1. use awk or sed to get file value process or run cmd if cmd value.

problem with this aproach is while i am copying file i am blocked and cant process further so if any file copy or cmd takes more time i cant process further.

so i thought of using another approach:
i will process above file and create a array of files and cmds and then i can process them individually.

I am new to shell scripting, so creating array and processing i have to read about.
please suggest me right way if you can define better.

i can also consider changing my above file sturcture if you have better way of copying files and processing cmds written there.

Thanks, Ajay

It's somewhat strange to have a script file reading a file of lines to script -- rather the long way around. If you're storing commands in a file to run, why not put them in the script file in the first place?

What are you actually trying to do -- not the algorithm you're following, but the goal the algorithm is trying to achieve...

thank you so much for feedback.
let me explain my problem.

My project consist of submodules and each module want me to do some kind of same work(copy some of their files , run few command and gather output from them)
the file which i have shown is kind of configuration file where each module will provide this file with their data to me and i will have single script which will read this file and process as i explained.

Does this make sense? or i should have some other approach?

waiting eagerly for the inputs.

I suspect I'd be making individual scripts instead of configuration files, here, and passing in the parameters as arguments instead of storing them in a configuration file, but it's hard to be sure when we know so little about your project.

You've explained in more detail how your config file is used, but still haven't said what it's for. What this project is supposed to be doing. Why you're breaking it up into modules. How these modules are supposed to be used.

ok let me give u details:

1. so our product consists of different sub projects which doesn't talk to each other they work independently.
2. Each sub project has its own directory structure and set of work it does.
3. Now requirement comes from these each sub projects: They want there log files and some data which can be gathered by running commands which only they know. each sub projects has same requirement.
4. So i want to write that script which will take a configuration file which will be given by each projects and i will process for them in this master script.
Note: configuration file contain "file" and "cmd" those are filled by them and it can be many more. so instead of each one doing same thing i was thinking i can go with this approach.

Does it help understanding requirement now? let me know if you have any more question.

Make individual scripts for each of the projects, run these scripts.

Or alternatively, make functions inside these scripts, and source these scripts to set variables and load functions, so you can load code exactly as given instead of having to strip it out of variables and run it with eval.

Perhaps something like this, if you have bash or ksh:

Code:

# Template script file

VAR1=asdf
VAR2=asdf

function do_something
{
        echo "do_something called"
}

function do_something_else
{
        echo "Wow, something else"
}

And you'd use it like

Code:

. /path/to/template ; do_something ; do_something_else