Re: doubtful RPMs?

[Alex Lancaster]

>>>>> "SM" == Sandor Megyaszai <address@hidden> writes:

SM> Hello!  I would have tried to use Swarm. I would use it through a
SM> higher level interface, MAML (Multi-Agent Modelling Language,
SM> http://www.syslab.ceu.hu/maml, probably you know about it). I'm
SM> not an expert of this MAML, so I cannot strongly state that the
SM> problem is not in that, but the signs say to me that the Swarm
SM> part is the problematic.

The MAML project has no formal affiliation with the Swarm project, and
although we have collaborated, we don't maintain MAML.  It's also
difficult to predict in advance how MAML is going to interact with any
given version of Swarm, and the only sure way of tracking down where
the bug lies, is actually debugging it, and not trying to second-guess
or assume that you know where the problem lies.  This is where Paul's
suggestion to use gdb, in the other followup to this post is
invaluable.

SM> MAML need Swarm version 1.4.1, but as a developer of it told me
SM> higher versions should have been also OK.  

This might be true anecdotally, but as far as I know, since MAML
hasn't been formally tested with more recent versions of Swarm,
there's no guarantee that more recent versions of Swarm will work
correctly.  In fact I'm not sure how actively MAML is being worked on
at the moment (any MAML developers lurking out there?).

SM> After I use RedHat Linux, I downloaded the only RPM version from
SM> your ftp site, swarm-hdf5-2.0.1-3.rh6.i386.rpm

SM> I installed it successfully, but I have problem with it when I try
SM> to compile and run a ~HelloWorld MAML example.

SM> When I tried to run the program:

>> *** event raised for error: OffsetOutOfRange *** function:
>> indexAtOffset(), file:
>> /home/alex/redhat/BUILD/swarm-2.0.1/src/collections/Collection.m,
>> line: 110 *** execution terminating due to error
>> /home/alex/redhat/BUILD/swarm-2.0.1/src/defobj/Symbol.m:173
>> -[Error(c) _raiseEvent:] Aborted (core dumped)

SM> Why I thought the mistake could be in the Swarm package: it
SM> mentions /home/alex directory which shouldn't be in an RPM
SM> package...

You are right, it isn't in the RPM package itself, but the
`/home/alex' path is on my build system (where the RPMs were built).
It is of no consequence as far as the operation of Swarm goes, it's
just the hardcoded path to the source that is installed in the binary
version of Swarm.

SM> Later I found the older versions of Swarm packed in RPMs, so I
SM> downloaded and installed now swarm-with-hdf5-1.4.1-3.rh6. The
SM> error message is a bit different but rather similar:

>> *** event raised for error: OffsetOutOfRange *** function:
>> indexAtOffset(), file: Collection.m, line: 109 *** execution
>> terminating due to error Aborted (core dumped)

SM> I really don't know what to do any more. It would be very great if
SM> you could help me, this is an important task of me.

One other avenue to try, is to download the matching swarmapps
packages (1.4.1. and 2.0.1) for each of the RPMs, try and compile
heatbugs for each of these versions, and post your results.  

These swarmapps are independent of MAML (and are the canonical way of
verifying an installation of Swarm is working correctly), and will
tell us whether the problem is with your Swarm installation itself, or
in some subtle interaction of Swarm and MAML.  They can be found at:

 ftp://ftp.swarm.org/pub/swarm/

Again, Paul's gdb suggestions will help you (and us) debugging your
problem a great deal.

Alex
-- 
Alex Lancaster * address@hidden * www.santafe.edu/~alex * 505 984-8800 x242
Santa Fe Institute (www.santafe.edu) & Swarm Development Group (www.swarm.org)

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.