thenumbernine / symmath-lua

Tensor algebra system written in lua
http://christopheremoore.net/symbolic-lua
MIT License
50 stars 7 forks source link

how to install #1

Closed sonoro1234 closed 9 years ago

sonoro1234 commented 9 years ago

Perhaps github root dir should be renamed synmath somewhere in package.path but I see require "ext" but I dont know what ext is

thenumbernine commented 9 years ago

Yeah, the "symmath-lua" project needs to go in a "symmath" directory. That's my "lua-ext" project: https://github.com/thenumbernine/lua-ext -- which needs to go in a "ext" folder name. I still haven't got a good naming scheme set up for my github projects, as you can see. I have a disclaimer with my "lua-ext" project to add "?/init.lua" to your LUA_PATH to get it to work.

On Fri, May 29, 2015 at 12:23 PM, Victor Bombi notifications@github.com wrote:

Perhaps github root dir should be renamed synmath somewhere in package.path but I see require "ext" but I dont know what ext is

— Reply to this email directly or view it on GitHub https://github.com/thenumbernine/symmath-lua/issues/1.

sonoro1234 commented 9 years ago

thanks for quick answer. Have tried two tests and they print html How do I redirect tests output to webbrowser?

thenumbernine commented 9 years ago

Do you mean, how did I embed it in a web-browser? For that you should check out Emscripten: https://github.com/kripken/emscripten . It lets you compile any LLVM bytecode to JavaScript. The project has a build for the Lua interpreter specifically that you can find here: https://kripken.github.io/lua.vm.js/lua.vm.js.html

On Fri, May 29, 2015 at 12:55 PM, Victor Bombi notifications@github.com wrote:

thanks for quick answer. Have tried two tests and they print html How do I redirect tests output to webbrowser?

— Reply to this email directly or view it on GitHub https://github.com/thenumbernine/symmath-lua/issues/1#issuecomment-106915270 .

sonoro1234 commented 9 years ago

No, I mean that I got html printed in the console. I did console redirection ( > test.html) to get it in a file. Is this the normal operation? I have also seen that I need MathJax

thenumbernine commented 9 years ago

Yeah that's normal. That's the default behavior for the test cases i had included - so i could run them in the embedded web browser version.

I'm planning to change it to default to fixed-width text output. In the mean time just comment out the lines in each test that say "symmath.tostring = require 'symmath.ToString.MathJax" or equivalent. On May 29, 2015 1:09 PM, "Victor Bombi" notifications@github.com wrote:

No, I mean that I got html printed in the console. I did console redirection ( > test.html) to get it in a file. Is this the normal operation? I have also seen that I need MathJax

— Reply to this email directly or view it on GitHub https://github.com/thenumbernine/symmath-lua/issues/1#issuecomment-106922255 .

sonoro1234 commented 9 years ago

yes but output is nicer with MathJax. And console output is sometimes as

<br>
geodesic
<br>
\({d^2 x^t} \over {d\tau^2}\) =
    d                                                 d                                                        d
                                         d                                                   d
                  d                                                        d
     d                                                                         d                             d
                                           d                                                        d
           d                                   d
v ----f {d x^z}\over {d\tau} {d x^x}\over {d\tau} + ----v f {d x^z}\over {d\tau} {d x^x}\over {d\tau} -    2 ----v {d x^z}\ov
er {d\tau} {d x^t}\over {d\tau} -  2   ----f {d x^z}\over {d\tau} {d x^t}\over {d\tau} + v ----f {d x^y}\over {d\tau} {d x^x}
\over {d\tau} + ----v f {d x^y}\over {d\tau} {d x^x}\over {d\tau} -    2 ----v {d x^y}\over {d\tau} {d x^t}\over {d\tau} -  2
   ----f {d x^y}\over {d\tau} {d x^t}\over {d\tau} +                     2   ----f +                     2 ----v f -      2
                                         ----v -    2                                             ----f +  2  3
       2 ----v +  3  2                     2 ----f
   dz                                                dz                                                 v f   dz
                                  v  f  dz                                                  dy
                 dy                                                 v f   dy                                               v
 f  dy                                               {d x^x}\over {d\tau}  v  dx     {d x^x}\over {d\tau}   dx       2 v f  {
d x^x}\over {d\tau} {d x^t}\over {d\tau}  dx     2 v  f {d x^x}\over {d\tau} {d x^t}\over {d\tau}  dx     v  f  {d x^t}\over
{d\tau}   dx     v  f  {d x^t}\over {d\tau}   dx
<br>
thenumbernine commented 9 years ago

Aha, looks like the html hardcoding is a deeper problem than I thought in my last email. With that said, most my tests aren't going to be able to switch away from html format so easily. Sorry about that.

On Fri, May 29, 2015 at 1:28 PM, Victor Bombi notifications@github.com wrote:

yes but output is nicer with MathJax. And console output is sometimes as


geodesic
({d^2 x^t} \over {d\tau^2}) = d d d d d d d d d d d d d d v ----f {d x^z}\over {d\tau} {d x^x}\over {d\tau} + ----v f {d x^z}\over {d\tau} {d x^x}\over {d\tau} - 2 ----v {d x^z}\ov er {d\tau} {d x^t}\over {d\tau} - 2 ----f {d x^z}\over {d\tau} {d x^t}\over {d\tau} + v ----f {d x^y}\over {d\tau} {d x^x} \over {d\tau} + ----v f {d x^y}\over {d\tau} {d x^x}\over {d\tau} - 2 ----v {d x^y}\over {d\tau} {d x^t}\over {d\tau} - 2 ----f {d x^y}\over {d\tau} {d x^t}\over {d\tau} + 2 ----f + 2 ----v f - 2 ----v - 2 ----f + 2 3 2 ----v + 3 2 2 ----f dz dz v f dz v f dz dy dy v f dy v f dy {d x^x}\over {d\tau} v dx {d x^x}\over {d\tau} dx 2 v f { d x^x}\over {d\tau} {d x^t}\over {d\tau} dx 2 v f {d x^x}\over {d\tau} {d x^t}\over {d\tau} dx v f {d x^t}\over {d\tau} dx v f {d x^t}\over {d\tau} dx

— Reply to this email directly or view it on GitHub https://github.com/thenumbernine/symmath-lua/issues/1#issuecomment-106927069 .

sonoro1234 commented 9 years ago

You are working for free so please dont feel sorry. I would only love to have some more docs and simpler tests. I was trying to do symbolic calculus with complex rational polynomials (for a work on audio filters) but I dont know how to do that. Thanks for your work

thenumbernine commented 9 years ago

I haven't got a good polynomial factoring algorithm in my symmath project anyways. Check out some of the more tried and true ones out there, like Maxima.

On Sat, May 30, 2015 at 12:17 AM, Victor Bombi notifications@github.com wrote:

You are working for free so please dont feel sorry. I would only love to have some more docs and simpler tests. I was trying to do symbolic calculus with complex rational polynomials (for a work on audio filters) but I dont know how to do that. Thanks for your work

— Reply to this email directly or view it on GitHub https://github.com/thenumbernine/symmath-lua/issues/1#issuecomment-107001514 .

sonoro1234 commented 9 years ago

Yes but it is Lisp. I wish your project could grow to make a Maxima like program (38 developers now) but in Lua. Lua is so good language that there are a lot of one-developer projects: gsl-shell, scilua,...( I do have some of them also, althought smaller than yours) but perhaps this is the cause of being a somehow special comunity with a limit for growing. Thanks alot for your attention. victor bombi

sonoro1234 commented 9 years ago

BTW: I do have a quite naif (I am just a musician) lib for polynomials and filters in https://github.com/sonoro1234/Lua2SC/tree/master/lua2SC/lua/num It allows real, complex, polynomial and rational polynomial variables (the two last almost allow symbolic calculus) with the help of metatables