MathGroup Archive 2010

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

Search the Archive

Re: Corruption of formulas after cut-and-paste

  • To: mathgroup at
  • Subject: [mg109994] Re: Corruption of formulas after cut-and-paste
  • From: John Fultz <jfultz at>
  • Date: Fri, 28 May 2010 07:20:49 -0400 (EDT)

You are correct, this is a bug.  The correct behavior isn't to produce a
warning, but to produce correct output.  The problem will occur with any
fraction that has a prefix '+' or '-' in the numerator, and which is part of an
implicit multiplication (i.e., one designated by whitespace as opposed to an
asterisk).  That means you can work around the problem by denoting the
multiplication explicitly with an asterisk.

The problem will be resolved in the next major release of Mathematica.


John Fultz
jfultz at
User Interface Group
Wolfram Research, Inc.

On Thu, 27 May 2010 06:46:21 -0400 (EDT), Yen Lee Loh wrote:
> A warning to people who are trying to write packages: Cut and paste may
> corrupt formulas, especially if they use complicated typesetting.  (My
> version: Mathematica, Linux x86-64)
> Example:  Open a notebook in Mathematica (e.g., a .nb file) and type
> -1
> f[x_] :== x -----;
> 2
> This is interpreted as "negative half x".  Now, open a new package and
> paste
> in that definition.  The formatting is preserved.  Save the package as an
> .m
> file.  Reload it.  Now you have
> f[x_]:==x -1/2;
> which is interpreted as "x minus half".  Not good.
> I didn't see any of the usual warnings about "converting to
> TraditionalForm/StandardForm/paste literally" etc.
> Maybe I am doing something wrong.  Anyway, I have to try to find a
> workaround.  -YLL

  • Prev by Date: Re: Why?
  • Next by Date: Re: exporting publishable 3D graphics
  • Previous by thread: Corruption of formulas after cut-and-paste
  • Next by thread: "Simplifying" expressions to not expand user-defined functions