[comp.unix.aux] Comp.sources.aux Comp.binaries.aux

name@portia.Stanford.EDU (tony cooper) (08/05/89)

Porting software to A/UX is a pain since very little source can be compiled
straight off. This is presumable due to the fact that A/UX is closer to
system V than anything else whereas most of the Unix software around is
for BSD. So when someone does port something to A/UX how about making it
available to the rest of the A/UX community?

Is anyone willing to donate some disk space for an anonymous FTP site? How
about a public directory on apple.apple.com?

I have some software to contribute. If you are interested, send me mail and
I can give you:

a whois program
a BSD style df (much better than the SYSV style)
a Pascal to C translator (can convert 1000 line programs reliably)
a Webster client
a fast man program that can do .Z man pages
patch version 2.0.12
compress (for those users who did not get it with A/UX 1.1)
software for manipulating 24 bit graphics images (but not displaying them
in 24 bits, only 8)
numerous X windows programs eg programs that can display gif, macpaint, sun
graphics files etc. X graphics programs are easy to get running under A/UX
but other types of programs usually require work since most X software is
sun or vax specific. I also have Imake and makedepend which Apple left out
of their X distribution.

I am currently porting mh6.6 and zoo to A/UX. Zoo is an archiving program
and might be quite useful to A/UX users who are having problems backing up
using the Apple tape drive.
I also have some magtape manipulation programs that I haven't looked at yet.

I also have tcsh already compiled for the macII. I didn't port it. tcsh is
like csh with a few extra features such as file name completion and command
line editing.

Finally, I am porting GoodNeWS to run under MacNeWS (from Grasshopper). Since
MacNeWS is the same as the version that Sun uses, this port should be easy.
I also have a lot of other NeWS software to look at, but I am only a beginner
in learning Postscript so it's too early to start being productive there.

Tony Cooper
name@portia.stanford.edu

Disclaimer: Half of the above is not true, the rest is false.

aem@ibiza.cs.miami.edu (a.e.mossberg) (08/06/89)

name@portia.Stanford.EDU (tony cooper) writes:
>Porting software to A/UX is a pain since very little source can be compiled
>straight off. This is presumable due to the fact that A/UX is closer to
>system V than anything else whereas most of the Unix software around is
>for BSD. So when someone does port something to A/UX how about making it
>available to the rest of the A/UX community?

I have had few problems porting stuff. Most programs have compilied with
little or *no* work on my part.  For instance, the ones you name:

>a whois program

No problem.

>a BSD style df (much better than the SYSV style)

A shell script posted to alt sources. No problem.

>patch version 2.0.12

No problem.

>compress (for those users who did not get it with A/UX 1.1)

First thing I set up under 1.0. No problem.

>I am currently porting mh6.6 and zoo to A/UX. Zoo is an archiving program
>and might be quite useful to A/UX users who are having problems backing up
>using the Apple tape drive.

Zoo compilied, no problem.

I don't think that anyone needs or should make available binaries of small
programs that compile right away. I mean, if you're talking something like
emacs which takes a little work, and alot of disk space, sure. But compress?
Geez. 

>I also have tcsh already compiled for the macII. I didn't port it. tcsh is
>like csh with a few extra features such as file name completion and command
>line editing.

Is it bug-less? I read there was one on tut.... that has bugs.

Now, if anyone has the Berkeley lpr/lpd working under A/UX, I'd be interested.
I spent a couple hours on it without getting it to work passably.

aem

a.e.mossberg - aem@mthvax.cs.miami.edu/aem@umiami.BITNET - Pahayokee Bioregion
Wine is but single broth; ale is meat, drink, and cloth. - 16th century proverb

allbery@nc386.UUCP (Brandon S. Allbery) (08/11/89)

In article <572@umigw.MIAMI.EDU>, aem@ibiza (a.e.mossberg) writes:
(name@portia.Stanford.EDU (tony cooper) writes in ">".)
+---------------
| >I am currently porting mh6.6 and zoo to A/UX. Zoo is an archiving program
| >and might be quite useful to A/UX users who are having problems backing up
| >using the Apple tape drive.
| 
| Zoo compilied, no problem.
+---------------

MH shouldn't be much of a problem... if it is, blame A/UX.  I'm running it
under System V and under Xenix (modulo compiler botch), and even compiled it
under System III.  I find it hard to imagine that A/UX could be worse than
that....

Maybe the person doing this is the one with the problem, not A/UX.

+---------------
| >I also have tcsh already compiled for the macII. I didn't port it. tcsh is
| >like csh with a few extra features such as file name completion and command
| >line editing.
| 
| Is it bug-less? I read there was one on tut.... that has bugs.
+---------------

So where does one get tcsh source (as opposed to tcsh patches for csh source
that (last I heard) was still to be considered under AT&T copyright)?

+---------------
| Now, if anyone has the Berkeley lpr/lpd working under A/UX, I'd be interested.
| I spent a couple hours on it without getting it to work passably.
+---------------

Why bother, unless Apple gratuitously decided not to provide lp?

++Brandon
-- 
Brandon S. Allbery, moderator of comp.sources.misc	     allbery@NCoast.ORG
uunet!hal.cwru.edu!ncoast!allbery		    ncoast!allbery@hal.cwru.edu
"Why do trans-atlantic transfers take so long?"
"Electrons don't swim very fast."  -john@minster.york.ac.uk and whh@PacBell.COM

aem@ibiza.cs.miami.edu (a.e.mossberg) (08/11/89)

allbery@nc386.UUCP (Brandon S. Allbery) writes:
>+---------------
>| Now, if anyone has the Berkeley lpr/lpd working under A/UX, I'd be interested.
>| I spent a couple hours on it without getting it to work passably.
>+---------------

>Why bother, unless Apple gratuitously decided not to provide lp?


Because we have primarily BSD systems, and System V lp has no provision
for printers attached to other systems. I have no desire to give people
accounts on my vaxstation or Sun 3/60 just so they can print.  I've
been sent a copy of diffs for lpd under A/UX, by the way, but haven't
tried it out yet.


BTW, I've snagged the binary for gcc from apple.com, as well as the
patched sources for a/ux, but can't compile gcc under a/ux, either using
gcc 1.35 (minus patchs) or cc with the big compilier. Has anyone been able
to compile it?

aem
a.e.mossberg - aem@mthvax.cs.miami.edu/aem@umiami.BITNET - Pahayokee Bioregion
Colourless green ideas sleep furiously.			- Noam Chomsky

allbery@nc386.uucp (Brandon S. Allbery) (08/16/89)

As quoted from <625@umigw.MIAMI.EDU> by aem@ibiza.cs.miami.edu (a.e.mossberg):
+---------------
| allbery@nc386.UUCP (Brandon S. Allbery) writes:
| >+---------------
| >| Now, if anyone has the Berkeley lpr/lpd working under A/UX, I'd be interested.
| >| I spent a couple hours on it without getting it to work passably.
| >+---------------
| 
| >Why bother, unless Apple gratuitously decided not to provide lp?
| 
| Because we have primarily BSD systems, and System V lp has no provision
| for printers attached to other systems. I have no desire to give people
| accounts on my vaxstation or Sun 3/60 just so they can print.  I've
+---------------

I found passing a print job to another system via either uux or remsh (you
call it "rsh") in a printer model to be trivial.

However -- we are likely to start supporting networks on our (System V)
systems soon, and I'm considering grabbing the 4.3+Tahoe lpd from uunet and
using it to write Streams/TLI-based programs to go both directions:  a daemon
to wait for remote print jobs and hand them to lp, and a print model which
opens a network connection to another system to print via BSD lpd protocol.

The problem with dropping lp in favor of lpr is that programs like e.g.
WordPerfect have a tendency to expect the lp spooler on System V systems, and
won't install properly if you aren't using lp.  (Yes, on a Mac you're more
likely to run the MacOS version of WP -- assuming A/UX 1.1 can handle it --
but I'm talking generic System V.)

++Brandon
-- 
Brandon S. Allbery, moderator of comp.sources.misc	     allbery@NCoast.ORG
uunet!hal.cwru.edu!ncoast!allbery		    ncoast!allbery@hal.cwru.edu
"Why do trans-atlantic transfers take so long?"
"Electrons don't swim very fast."  -john@minster.york.ac.uk and whh@PacBell.COM

aem@ibiza.cs.miami.edu (a.e.mossberg) (08/16/89)

allbery@nc386.uucp (Brandon S. Allbery) writes:
>I found passing a print job to another system via either uux or remsh (you
>call it "rsh") in a printer model to be trivial.

The problem with remsh is either having to set up a "print" user and doing
convoluted setuid thingies to send the files, or more easily, giving users
accounts on the machines with the printers, which I really didn't want to
do. Besides, my machines were already set up for lpr/lpd. I didn't want to
change anything just because a few a/ux systems were forced on us.

>The problem with dropping lp in favor of lpr is that programs like e.g.
>WordPerfect have a tendency to expect the lp spooler on System V systems, and
>won't install properly if you aren't using lp.  (Yes, on a Mac you're more
>likely to run the MacOS version of WP -- assuming A/UX 1.1 can handle it --
>but I'm talking generic System V.)

We don't, and don't plan to, run any MacOS software.  The few things that users
have asked to run, don't run under A/UX 1.1.  



aem

a.e.mossberg - aem@mthvax.cs.miami.edu/aem@umiami.BITNET - Pahayokee Bioregion
Chances that a homeless American holds a full- or part-time jobs: 1 in 5
							- Harper's Index 2/89