Re: Let's talk some integrals

2008-04-08 Thread Ondrej Certik
On Sat, Apr 5, 2008 at 6:39 PM, Ryan James <[EMAIL PROTECTED]> wrote: > > On Sat, 2008-04-05 at 18:25 +0200, Ondrej Certik wrote: > > > BTW - is there any application, where you actually get one > > trigonometric function in the other: > > > > sin(cos(x)) > > > > ? > > the integral forms of

Re: compiling functions to machine code

2008-04-08 Thread Vinzent Steinberg
On 3 Apr., 13:35, mabshoff <[EMAIL PROTECTED]> wrote: > On Apr 2, 10:58 pm, Robert Bradshaw <[EMAIL PROTECTED]> > wrote: > > > This is very interesting. I actually implemented something along these > > lines (but not near as complete) before implementing the fast_float > > stuff. Clearly nothing i

Re: compiling functions to machine code

2008-04-08 Thread Vinzent Steinberg
Cython generates this: static double __pyx_f_54C:\WINDOWS\Temp\sympy \c22dfc385147e1fa7b59a1dafc6f623e_cf(double __pyx_v_x) { double __pyx_r; /* "C:\WINDOWS\Temp\sympy\c22dfc385147e1fa7b59a1dafc6f623e.pyx":38 * # define C function * cdef double cf(double x): * return (2*exp(2*x) - 3.

Re: compiling functions to machine code

2008-04-08 Thread mabshoff
On Apr 8, 2:50 pm, Vinzent Steinberg <[EMAIL PROTECTED]> wrote: > On 3 Apr., 13:35, mabshoff <[EMAIL PROTECTED]> wrote: > > Yep, and since there are also a bunch of Sympy users on Windows some > > of the assumptions about the presence of gcc might not hold. Sage is > > getting ported to MSVC

Re: compiling functions to machine code

2008-04-08 Thread Ondrej Certik
On Tue, Apr 8, 2008 at 3:04 PM, Vinzent Steinberg <[EMAIL PROTECTED]> wrote: > > Cython generates this: > > static double __pyx_f_54C:\WINDOWS\Temp\sympy > \c22dfc385147e1fa7b59a1dafc6f623e_cf(double __pyx_v_x) { > double __pyx_r; You mean this ^^^? I don't think so. Try to report it to th

Re: compiling functions to machine code

2008-04-08 Thread Vinzent Steinberg
On 8 Apr., 15:56, mabshoff <[EMAIL PROTECTED]> wrote: > On Apr 8, 2:50 pm, Vinzent Steinberg > > <[EMAIL PROTECTED]> wrote: > > On 3 Apr., 13:35, mabshoff <[EMAIL PROTECTED]> wrote: > > This is not about duplication of efforts in Sage. My point was to > illustrate that the assumption "C compiler

Re: compiling functions to machine code

2008-04-08 Thread Ondrej Certik
On Tue, Apr 8, 2008 at 8:29 PM, Vinzent Steinberg <[EMAIL PROTECTED]> wrote: > > > > On 8 Apr., 15:56, mabshoff <[EMAIL PROTECTED]> wrote: > > On Apr 8, 2:50 pm, Vinzent Steinberg > > > > <[EMAIL PROTECTED]> wrote: > > > On 3 Apr., 13:35, mabshoff <[EMAIL PROTECTED]> wrote: > > > > > This is

Re: compiling functions to machine code

2008-04-08 Thread Vinzent Steinberg
On 8 Apr., 15:25, "Ondrej Certik" <[EMAIL PROTECTED]> wrote: > On Tue, Apr 8, 2008 at 3:04 PM, Vinzent Steinberg > > <[EMAIL PROTECTED]> wrote: > > >  Cython generates this: > > >  static  double __pyx_f_54C:\WINDOWS\Temp\sympy > >  \c22dfc385147e1fa7b59a1dafc6f623e_cf(double __pyx_v_x) { > >  

Re: compiling functions to machine code

2008-04-08 Thread Ondrej Certik
> > BTW, you know your code the best, where and how should we include it > > in sympy? I guess at around the lambdify code? > > Let's do it. > > Yeah, it works exactly like lambdify. It should also be possible to > pass lambda strings directly (to avoid sympy overhead). But I will > need to

Re: compiling functions to machine code

2008-04-08 Thread Robert Kern
On Tue, Apr 8, 2008 at 12:43 PM, Ondrej Certik <[EMAIL PROTECTED]> wrote: > tcc is now in Debian and it's indeed superfast, compared to gcc. And > it can compile simple projects just fine. > > If you make it work with tcc, that'd be really cool. BTW, if you > realize you don't have time to fi

Re: compiling functions to machine code

2008-04-08 Thread Michael Abshoff
On Tue, Apr 8, 2008 at 11:13 PM, Robert Kern <[EMAIL PROTECTED]> wrote: > > On Tue, Apr 8, 2008 at 12:43 PM, Ondrej Certik <[EMAIL PROTECTED]> wrote: > > > tcc is now in Debian and it's indeed superfast, compared to gcc. And > > it can compile simple projects just fine. > > > > If you make it w