MOO-cows Mailing List Archive


Re: A possible bug and a request...

At 03:17 PM 1/9/96 PST, you wrote:
>I've mentioned this bug to a few people but I don't know whether it ever 
>actually got brought to this list, and now that I'm on this list I 
>thought I'd point it out myself. :)

This is a known bug. I, myself, am not sure on the state of it tho.

>First, we're currently running V1.7.8.pl4 of the LambdaMOO server.

You might consider upgrading to 1.7.9p2.
;;x={1,2,3,4,5,6}; x[2] = x[4] = 8; return x
=> {1, 8, 3, 4, 5, 6}

So, apparently, a bug exists in 1.7.9 (unless these are the correct results;
anyone?), but I don't think it's the same one that you describe.

>Now the request:
>Whenever a traceback occurs it is generally difficult for a programmer 
>to get enough information out of the person it happened to to try to fix 
>it.  What would be really cool is if you could somehow store tracebacks 
>in a special property on #0,  if the property exists. That way whenever 
>a traceback occurs any programmer or wizard on line could take a look at 
>exactly what happened.

I just finished up a patch such that when a traceback is going to be printed,
the server tries to let #0:do_traceback handle it, passing it a list
structure with the data that it would otherwise print (but it's in it's
actual values tho, not as strings).  If you return 0 from this verb, then
the server goes ahead and prints the traceback, if you return 1, the server
assumes you handled it (same rules that exist for :do_command).

If you are interested in this patch, I can give ya diffs.


Home | Subject Index | Thread Index