home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: vmsnet.sources.games
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!unixhub!slacvx.slac.stanford.edu!fairfield
- From: fairfield@slacvx.slac.stanford.edu
- Subject: Re^2: Vmsnetrek 47/47 <- FINALLY!
- Message-ID: <1992Nov22.133954.1@slacvx.slac.stanford.edu>
- Lines: 82
- Sender: news@unixhub.SLAC.Stanford.EDU
- Organization: Stanford Linear Accelerator Center
- References: <1992Nov20.211158.1@acad2.alaska.edu> <21NOV199211183262@spades.aces.com> <1992Nov21.234255.5854@ais.com>
- Date: Sun, 22 Nov 1992 21:39:54 GMT
-
- In article <1992Nov21.234255.5854@ais.com>, bruce@ais.com (Bruce C. Wright) writes:
- >
- > But I've got all the parts, and it appears that part 31 is incorrect --
- > when I build it, the orbit.c and oldbitmap.h files get checksum errors,
- > and looking at part 31 it appears to be from a different vms_share
- > distribution: it appears to pick oldbitmap.h in a slightly different
- > place from where part 30 left off, and it leaves off orbit.c in a
- > different place than where part 32 picks up (duplicating code).
-
- I encountered the same thing: VMS_SHARE indicated a checksum error
- for OLDBITMAPS.H and ORBIT.C on unpacking, so it's likely Don's distribution,
- not a random news/mail error...
-
- > Does anyone have a correct copy of part 31? I'd like to try to build a
- > version that could run on something besides Wollongong... unfortunately
- > it's often not so simple as changing the linkage libraries because
- > not all TCP/IP packages use the same routine names.
-
- I decided to follow up on Don's remark that what _he_ thought was a
- Multinet version was available via FTP from RAVEN.ALASKA.EDU and from
- FTP.ULOWELL.EDU. Both these sites have a set of tar'd and lz'd files
- of the client and server sources and exe's (I haven't tried the exe's).
-
- In addition, RAVEN.ALASKA.EDU has the client sources, however, I was
- unsuccessful in retrieving these: RAVEN is running Ultrix 4.1 and whoever
- put the sources up for distribution totally munged the record structure
- in the copy from VMS. (I'm using Multinet v3.1D and no combination of TYPE
- ASCII/BINARY and/or STREAM ON/OFF made any difference: the source text lines
- lack any end-of-line indicator...)
-
- Recipe: I used TYPE IMAGE and STREAM OFF to GET vmsnetreksrc.tar.z from
- RAVEN.ALASKA.EDU. Then I used Hunter Goatley's (well, his distribution :-)
- LZDCMP with the -B option to make the tar file:
-
- $ LZDCMP -BF vmsnetreksrc.tar_z vmsnetreksrc.tar
-
- Without the -B option, LZDCMP failed. Next I restored the files using
- (Hunter's) VMSTAR:
-
- $ VMSTAR XVF vmsnetreksrc.tar
-
- By the way, this created a [.CLIENT] subdirectory and put all the source
- files there...also gave them file types like .C_47 since the vms version
- numbers had been "preserved" in the tar file as, e.g. orbit.c.3. I wrote
- a little DCL file to rename these files sans the "_nnn" in the file type.
-
- OK, now the point: it looks like this particular vmsnetrek was written
- for UCX. The descrip.mms includes the line:
-
- xlibshr/lib, dwtlibshr/lib, sys$library:ucx$ipc.olb/lib
-
- I also searched the .C (and .H) files for types.h, socket.h and netdb.h.
- Where they occur, most are in the form #include <sys/types.h>. There
- are occurrences of #include <sys/socket.h> _and_ #include <socket.h>, but
- netdb occurs only as #include <netdb.h>.
-
- So two questions for Bruce and Ehud: 1) Is it expected that this code
- could be built and run under Multinet using Multinet's UCX emulation, without
- other changes (we're running Multinet, _not_ UCX); and 2) Barring that, would
- something like the following, suggested by Mike Kienenberger, suffice:
-
- a) Make sure netdb.h is included as: #include <multi/netdb.h>
- b) Make sure socket.h is included as: #include <sys/socket.h>
- c) Before compiling, define the following logicals:
-
- $ define sys multinet_root:[multinet.include.sys]
- $ define vms multinet_root:[multinet.include.vms]
- $ define multi multinet_root:[multinet.include]
-
- d) Link like so:
-
- $ link client,sys$input/opt
- sys$share:vaxcrtl/share
- multinet:multinet_socket_library/share
- <control-z>
-
- --
- Dr. Kenneth H. Fairfield | Internet: Fairfield@Slac.Stanford.Edu
- SLAC, P.O.Box 4349, MS 98 | DECnet: 45537::FAIRFIELD (45537=SLACVX)
- Stanford, CA 94309 | BITNET Fairfield@Slacvx
- ----------------------------------------------------------------------------
- These opinions are mine, not SLAC's, Stanford's, nor the DOE's...
-