[gnet-dev] Callback arguments in GNet

[David Helder <dhelder umich edu>]

I was writing up some documentation on async write and I noticed that GNet
is fairly inconsistent on what should be the arguments in callback.

We seem to do it two ways:

1. Callback includes return value, status, and user data.

  Ex:

    typedef void (*GInetAddrNewAsyncFunc)(GInetAddr* inetaddr,
  				        GInetAddrAsyncStatus status,
				        gpointer data);

  Associated with:

    GInetAddrNewAsyncID
    gnet_inetaddr_new_async (const gchar* name, gint port,
  			   GInetAddrNewAsyncFunc func, gpointer data);

  Note the callback does not include name and port.


2. Callback includes return value, status, and user data, AS WELL AS
most/all original arguments.


  Ex:

    typedef void (*GInetAddrGetNameAsyncFunc)(GInetAddr* inetaddr,
					  GInetAddrAsyncStatus status,
					  gchar* name,
					  gpointer data);

  Associated with:

    GInetAddrGetNameAsyncID
    gnet_inetaddr_get_name_async (GInetAddr* ia,
			      GInetAddrGetNameAsyncFunc func,
			      gpointer data);

  Note the callback includes inetaddr.



My leaning is more to option #1.  #1 may be easier for the programmer, but
it requires us to keep a lot more state.  Besides, the programmer already
has the data pointer which should be enough.  In #2, inetaddr in the
callback is actually a copy of the inetaddr passed in the original
function.  (Hmm, really it could be a ref, but then I'd have to add
mutexes.)

Are there any opinions on this?  Note I wouldn't make any changes to the
stable API until GNet 1.2 (whenever and whatever that is).

David



-- 
      __          _    __ David Helder - dhelder umich edu
  ___/ /__ __  __(_)__/ / <http://www.eecs.umich.edu/~dhelder>
 / _  / _ `/ |/ / / _  /  Jungle Monkey: <http://www.junglemonkey.net>
 |_,_/|_,_/|___/_/|_,_/   Paper CD Case: <http://www.papercdcase.com>