leahneukirchen/nq

Completion of error handling

elfring opened this issue · 4 comments

Would you like to add more error handling for return values from functions like the following?

Ok, let's see what could happen here...

dup2: EBADF for oldfd can't happen, EBADF for newfd could happen (with very low limits), same for EMFILE, EBUSY could happen on Linux I guess, EINTR could happen... probably we can just panic when anything bad happens here.

flock: I guess we should catch EINTR, and panic on ENOLCK... rest should not happen.

I suggest to avoid ignorance of return values a bit more.

Are you interested to apply aspect-oriented software development?
How do you think about to encapsulate error detection and corresponding exception handling as a reusable aspect in your software?

Ok, EINTR cannot happen either because we don't use signals.

I checked the dup2 calls now, I don't think islocked needs change.

I think your suggestions are overkill for a tiny 500 loc C progam. ;)

How do you think about to improve static source code analysis also for your software?