Shell Programming and Scripting

HI All,

Whenever I write a shell script I always check if a command got executed successfully, even for the commands like cd, mv, rm and others, and even for the cases when there is ALMOST nothing to stop this commands from executing with success. so I am wondering if it is an overkill. I am thinking that there is always a chance that some directory might be accidently removed by a user/process and so on.

So I am just curious to hear about your approach to this.
Thanks

When I write a script that only I will be running and I know that I will pay attention to any diagnostics my script produces and know how to fix it, I may omit some of the error checking.

When I provide sample code for the UNIX & Linux Forums, I occasionally put in error checking but generally assume that I'm writing samples rather than production code. If someone wants to put my samples into their production code, it is their job to make it bulletproof.

If I'm writing code for production use, I check the exit status of almost everything. (However, I seldom check the exit status of writing diagnostic messages. If they fail, what are you going to do -- write another diagnostic that will probably also fail???)

I pretty much follow the same philosophy when verifying that command line parameters and data being processed from user-supplied files are in the format needed to get the job done.