MathGroup Archive 2007

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Re: spurious $Aborted messages. How to track

  • To: mathgroup at smc.vnet.net
  • Subject: [mg80476] Re: [mg79738] Re: spurious $Aborted messages. How to track
  • From: Chris Hill <chill at wolfram.com>
  • Date: Thu, 23 Aug 2007 01:10:09 -0400 (EDT)
  • References: <f8cgnd$44d$1@smc.vnet.net>

At 05:25 AM 8/3/2007, Nasser Abbasi wrote:
>The same for the normal error messages, why no line number to tell the
>user where the error is?
>
>I spend 90% of my time in Mathematica trying to find the location of
>the error that caused the error message, because Mathematica simply
>tells the user what the error is, NOT where it is as well.
>
>Is the concept of a line number in the code or even a stack printout
>to show the call history to help the user find the cause of the error
>something that conflicts with the design of Mathematica?

There is a feature in the Mathematica version 6 debugger called 
"Break at Messages" which will stop execution when a message is 
issued.  While stopped, the stack can be inspected using the Stack 
window.  This will supply some context for the message, which is 
usually helpful in isolating a problem.

Chris Hill
Wolfram Research 



  • Prev by Date: Re: ParametricPlot3D from 5.2 to 6.0
  • Next by Date: Re: PlotStyle in version 6
  • Previous by thread: Re: ParametricPlot3D from 5.2 to 6.0
  • Next by thread: Manipulate a VectorFieldPlot3D