[ clocc-Bugs-564727 ] Bugfixes and some other things to PORT

Bugs item #564727, was opened at 2002-06-04 22:56
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=101802&aid=564727&group_id=1802

Category: port
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Sam Steingold (sds)
Summary: Bugfixes and some other things to PORT

Initial Comment:
Fixes:
- Some places used #+cmucl instead of #+cmu, or
  #-sbcl instead of #-(and sbcl (or net.sbcl.sockets
db-sockets)).
- net.lisp:deftype socket and defun wait-for-stream had
#+sbcl, but
  no matching #-sbcl in the default.
- net.lisp:open-socket looks strange:
  It converts an ipaddr argument to a hostname, which
means that the
  socket function has to convert it back to an ip
address.  I think
  it shold use just ipaddr-to-dotted instead.  This
patch does that,
  but it leaves a placeholder (SOMETHING host) for sbcl
db-sockets,
  where host is a dotted host address, because I don't
know what
  kind of arguments sockets:socket-connect takes. 
Hopefully you do.
- In net.lisp:socket-accept at the comment
	; who needs WAIT and BIN anyway :-S
  I inserted (ecase wait ((nil))), since non-nil wait
is unsupported for
  sbcl db-sockets.  I don't know if BIN should be
handled the same way.
- Similarly, I replaced (ignore kind) in defun
open-unix-socket with
  (ecase kind (:stream)).
- However, defun open-unix-socket looks *really* weird
to me, in that
  it opens a plain file instead of giving an error if
there is no
  useful socket function.  How is that supposed to
work?  Plain files
  have completely different semantics from sockets, and
it's unlikely
  that anyone would use a socket if a plain file would
do.

- path.lisp:defun load-logical-host-def's documentation
said STYLE
  could be :CMU, but the function actually checked for
:CMUCL.
  I changed the doc string to match the old
functionality, and changed
  the function to accept both :CMU and :CMUCL.

- One other thing:  I'm a bit surprised that you do
	#-(or <foobar>) (error ...)
  _inside_ the functions instead of putting
	#+(or <foobar>)
  in _front_ of the functions.  That would catch such
problems
  earlier.  I'll make a patch if you want.


----------------------------------------------------------------------

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=101802&aid=564727&group_id=1802

_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm