This looks good. Maybe we can modify it to use boolean instead of Boolean and just have a separate method with no boolean parameter for the "null" behavior you have now?

I'm not quite sure how to completely get rid of the Boolean, since the current code relies on the buildAddrinfoList() function to process the args array before sending the processed args on to the AddrinfoCabllack() function.

If the concern is over having the getHostAddress(ThreadContext context, InetAddress addr) function call the getHostAddress(ThreadContext context, InetAddress addr, boolean reverse) function and pass in a null for the last parameter, I can do what you suggest and leave the original getHostAddress(ThreadContext context, InetAddress addr) function alone; however, I can't see a way to get around then having to do something ugly like the following inside the AddrInfoCallback() to select the appropriate function call:

                    if (reverse == null) {
                        c[2] = runtime.newString(getHostAddress(context, address));
                    } else {
                        c[2] = runtime.newString(getHostAddress(context, address, reverse));
                    }

I noticed I've made a serious error in doing a String comparison using the == operator in my initial Pull Request. I'm new to the Git paradigm; is the proper procedure here to make a brand new Pull Request, with the fixed code and close this one?

Just amend your fix to your existing commit and do a force push to your socket_getaddrinfo_reverse_lookup branch and it will update on this pull request.

Amending and force pushing worked like a charm, thanks.