MathGroup Archive 2009

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

Search the Archive

Re: Log[]//TraditionalForm

  • To: mathgroup at smc.vnet.net
  • Subject: [mg96515] Re: Log[]//TraditionalForm
  • From: Bill Rowe <readnews at sbcglobal.net>
  • Date: Sun, 15 Feb 2009 03:21:44 -0500 (EST)

On 2/14/09 at 3:09 AM, human at site.pl (slawek) wrote:

>BTW, Log vs. ln is an aesthetic problem. The serious troubles are
>caused by CForm and FortranForm. These "forms" produces output
>unreadable by compilers. Therefore it is not possible to use
>Mathematica to generate source code in C language. You need a proof?
>Here is an example:

>In[1]:= Series[Sin[x],{x,0,5}] Out[1]= x-x^3/6+x^5/120+O[x]^6
>In[2]:= Series[Sin[x],{x,0,5}]//CForm

<snip>

>Neither C/C++ nor Fortran (II/IV/77/95) have got "List" or
>"SeriesData".

Mathematica has a large number of built in functions that simply
have no counterpart in C/C++ or Fortran. So, if you are
expecting CForm or FortranForm to output something
understandable to C/C++ or Fortran when given an arbitrary
Mathematica expression you are doomed to disappointment.

But this particular case is easily resolved with a bit of
thought. That is:

In[10]:= Series[Sin[x], {x, 0, 5}] // Normal // CForm

Out[10]//CForm=
x - Power(x,3)/6. + Power(x,5)/120.

In[11]:= Series[Sin[x], {x, 0, 5}] // Normal // FortranForm

Out[11]//FortranForm=
x - x**3/6. + x**5/120.



  • Prev by Date: Surface fitting error while using Fit
  • Next by Date: Re: newbie here,, need help with parametrics
  • Previous by thread: Re: Surface fitting error while using Fit
  • Next by thread: optimization