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

Re: Re: Re: Undo in Mathematica

• To: mathgroup at smc.vnet.net
• Subject: [mg105095] Re: [mg105069] Re: [mg105053] Re: Undo in Mathematica
• From: brien colwell <xcolwell at gmail.com>
• Date: Fri, 20 Nov 2009 06:41:58 -0500 (EST)
• References: <200911191225.HAA19386@smc.vnet.net>

Just my 2 cents ... the practical lack of undo is one of the major 
thorns my team has with developing in M. The other major thorns are with 
the IDE/notebook editor itself, which would benefit from listing the 
source file and line number where error messages are generated, linking 
symbols to their source-code declarations, listing available symbols for 
"smart completion,"  and tighter integration with the debugger.

The Eclipse-based workbench seems a promising replacement for the 
notebook editor, but my understanding is that the workbench currently is 
not as interactive as the notebook editor.

I am really curious how people develop large libraries in M.







John Fultz wrote:
> On Thu, 19 Nov 2009 05:23:42 -0500 (EST), David Bailey wrote:
>   
>> fd wrote:
>>
>>     
>>> All,
>>>
>>> I was wondering if and how I could do any useful "undo" in
>>> Mathematica. The one I get from the edit menu just allows me to undo
>>> the last step (if it is not evaluated). This is pretty unsatisfactory,
>>> even more for a program with this complexity.
>>>
>>> I've tried navigating in the Option Inspector without luck, maybe
>>> someone found something?
>>>
>>>
>>> F.
>>>
>>>
>>>       
>> WRI claim that a general (multi-level) undo is impractical in notebooks
>> because of the complexity of the environment and potential size of the
>> files, but I would claim they could do a LOT better than the current
>> scheme.
>>     
>
> To set the record straight, WRI does not make that claim.
>
> Sincerely,
>  
> John Fultz
> jfultz at wolfram.com
> User Interface Group
> Wolfram Research, Inc.
>
>
>   
>> For example, the could restrict the UNDO mechanism to cells that are not
>> generated (i.e. not output), and possibly the undo stack would only go
>> back as far as the previous evaluation.
>>
>> I would have thought a multi-level undo is pretty vital in an
>> interactive program!
>>
>> David Bailey
>> http://www.dbaileyconsultancy.co.uk
>>     
>
>
>
>
>   

• References:
  • Re: Re: Undo in Mathematica
    • From: John Fultz <jfultz@wolfram.com>