[FIXED] What is the purpose of the `withFdSocket sock setCloseOnExecIfNeeded` call in Network.Socket example?

Issue

In the Network.Socket package, there is an echo server example. In the echo server, a call to withFdSocket sock setCloseOnExecIfNeeded is made just before actually binding to the socket. The relevant function is below:

    open :: AddrInfo -> IO Socket
    open addr = E.bracketOnError (openSocket addr) close $ \sock -> do
        setSocketOption sock ReuseAddr 1
        withFdSocket sock setCloseOnExecIfNeeded
        bind sock $ addrAddress addr
        listen sock 1024
        return sock

The full example can be found on the Network.Socket package documentation.

What is the purpose of this call? It's my understanding that withFdSocket creates a file descriptor associated with the socket, and setCloseOnExecIfNeeded sets the CLOEXEC flag on the descriptor. However this descriptor is immediately 'discarded' and nothing is done with it.

Am I confused that the file descriptor always exists and withFdSocket just provides this existing descriptor, which we have to update with the flag in order for the socket to close when the program exits?

Solution

withFdSocket does not create an Fd; it just reads the one already stored inside the socket data structure.

Answered By - Daniel Wagner

Answer Checked By - Marie Seifert (PHPFixing Admin)