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

Re: Initialization Cells Not Evaluated When Notebook Saved to CDF

• To: mathgroup at smc.vnet.net
• Subject: [mg124756] Re: Initialization Cells Not Evaluated When Notebook Saved to CDF
• From: Gregory Lypny <gregory.lypny at videotron.ca>
• Date: Sat, 4 Feb 2012 06:30:15 -0500 (EST)
• Delivered-to: l-mathgroup@mail-archive0.wolfram.com

Thank you, John.  Much appreciated.  I was on my way to discovering as much yesterday as I worked on a fix that involved moving key variables into a Manipulate that has SaveDefinitions set to True.  I'm going to explore your other suggestions.

Regards,

Gregory



On Fri, Feb 3, 2012, at 11:34 AM, John Fultz wrote:

> Actually, I'll bet that it does *not* work in Mathematica.  Which is to say that
> if you do nothing more than open the CDF file in a freshly started session of
> Mathematica, you'll get the same breakages.
>
> CDF files which you expect to deploy cannot rely on Shift+Enter evaluations to
> prime the pump.  You need to make sure that all of the code dependencies are in
> the dynamic evaluations somewhere.  Some possible ways of doing this, all of
> which have been discussed at various points on MathGroup, include:
>
> * Using the Initialization option of Dynamic, DynamicModule, or Manipulate
> * Using the SaveDefinitions option of Manipulate
> * Adding code to the NotebookDynamicExpression option of the notebook (if it's
> initialization code, then wrapped in Refresh[#,None]& to prevent it from
> evaluating more than once per session).
> * Not such a good idea for function definitions, but if you simply have code
> that needs to run before Dynamic code runs, nesting a DynamicWrapper around it
> might be appropriate, too.
>
> Sincerely,
>
> John Fultz
> jfultz at wolfram.com
> User Interface Group
> Wolfram Research, Inc.