Re: Strange segfault with improper usage

   From: Paolo <[email protected]>

   On Fri, Mar 21, 2008 at 04:23:54PM +0100, Paolo wrote:
   > On Fri, Mar 21, 2008 at 10:52:17AM -0400, Bill Y wrote:
   > > 
   > > I'm just suggesting -e to see if that tickles (or de-tickles) the bug.

   in crm_osb_bayes.c:1240 we get SEGV if -t is used and one cssfile is missing:

     if (internal_trace)
       fprintf (stderr, " files  %ld  learns #0 %lu  #1 %lu  total %lu cp0 %f cp1 %
   f \n",
	      maxhash,
	      hashes[0][learns_index[0]].value,
	      hashes[1][learns_index[1]].value,
	      total_learns,                    
	      cpcorr [0],
	      cpcorr [1] );

   got it on Milan's vserver.

   This is not the bug we're chasing here. A bug
   nevertheless. Possibly in other classifiers as well.

OK, got a patch?

(speakin' of which, TRE 0.7.5 is _less_ buggy than 0.7.3 and therefore
should be considered the "stable" branch (at least as far as I know).) 

Ville: is this true?

    - Bill 

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/