Re^2: Splitting up quoted/escaped command line arguments

[Tommy]

...Because that sends it through a shell. I need to avoid that.

[salva]

But hose commands would not work unless you pass them through a shell!

Or if you prefer to view it from a different angle, you will have to implement in your program all the shell functionality used by those commands!

Update: And BTW, if those commands involve calling ssh, then a shell would be invoked at the remote side. This is an unavoidable feature of the SSH protocol.

[Tommy]

Yes, I'm keenly aware that a shell is going to be invoked on the remote side. That's fine. I need to avoid it on the local machine. Why do you say that these commands won't work unless passed to a shell vs calling ssh directly? The perldocs don't seem to agree with your assertion.

 

[salva]

Why do you say that these commands won't work unless passed to a shell vs calling ssh directly?

Because it's obvious they have been written to be parsed by a shell.

Can you assure that there would be no entries using redirections, pipes, back-quotes, logical operators, variables, wildcards, etc.? Those are features that any intermediate shell user employs frequently.

[runrig]

I need to avoid that.

Why?

[Tommy]

I'll potentially be doing hundreds of these calls in parallel and I have to avoid the overhead of shelling out every time. I've already tested and the difference is very significant when opening to ssh vs opening to a shell opening to ssh. There's a huge slowdown on the local machine when I involve the shell.

[runrig]

Store the string in the database with some token that you can split on, that won't show up in the command itself. But, the "||" in your example is a shell operation, it won't work in 'exec @args'. Ignore that last part...I see it's part of the remote command passed to ssh.

Update: Isn't the command just one arg then that's passed to ssh?